Received 5 July 2023, accepted 4 August 2023, date of publication 11 August 2023, date of current version 21 August 2023.

Digital Object Identifier 10.1109/ACCESS.2023.3304332

A Comprehensive Review of Green Computing:

Past, Present, and Future Research
SHOWMICK GUHA PAUL 1 , ARPA SAHA 1 ,
MOHAMMAD SHAMSUL AREFIN 1,2 , (Senior Member, IEEE), TOUHID BHUIYAN1 ,
AL AMIN BISWAS 3 , AHMED WASIF REZA 4 , (Member, IEEE), NAIF M. ALOTAIBI5 ,
SALEM A. ALYAMI 5 , (Member, IEEE), AND MOHAMMAD ALI MONI 6
1 Department of Computer Science and Engineering, Daffodil International University, Dhaka 1216, Bangladesh
2 Department of Computer Science and Engineering, Chittagong University of Engineering and Technology, Chattogram 4349, Bangladesh
3 Department of Computer Science and Engineering, Bangabandhu Sheikh Mujibur Rahman University, Kishoreganj 2300, Bangladesh
4 Department of Computer Science and Engineering, East West University, Dhaka 1212, Bangladesh
5 Department of Mathematics and Statistics, Faculty of Science, Imam Mohammad Ibn Saud Islamic University, Riyadh 11564, Saudi Arabia
6 Artificial Intelligence and Cyber Futures Institute, Charles Stuart University, Bathurst, NSW 2795, Australia

Corresponding authors: Mohammad Ali Moni ([email protected]) and Mohammad Shamsul Arefin ([email protected])
This work was supported by the Deanship of Scientific Research, Imam Mohammad Ibn Saud Islamic University (IMSIU), through the
Research Partnership Program under Grant RP-21-09-09.

ABSTRACT Green computing, also called sustainable computing, is the process of developing and
optimizing computer chips, systems, networks, and software in such a manner that can maximize efficiency
by utilizing energy more efficiently and minimizing the negative environmental influence on the surrounding.
The term ‘‘green computing’’ refers to practices that lessen the negative effects of technology on the
environment. Due to the improvements in modern technology, various devices, mechanisms, and software
have been developed, and lots of studies have been conducted to optimize and increase those technologies’
green computing abilities. Thus, review and summarization of green computing-based studies are required
to identify the current advancements, challenges, and future research opportunities. This study reviewed
and summarized green computing in each area studies, by exploring green computing’s twelve areas.
Current research trends, datasets or testing mechanisms, and the construction or implementation of various
technologies to accomplish green computing and sustainable development have been discussed. This study,
after conducting a thorough comparison and analysis, provides responses to the proposed state-of-the-art
research questions. Furthermore, this study presents the current challenges and future research opportunities
with respect to each green computing area. This study will provide organizations, researchers, and institutions
conducting research on green computing with insights and ideas. Furthermore, environmental organizations,
companies, and government agencies concerned with reducing carbon emissions and energy consumption
will also benefit from this review study.

INDEX TERMS Carbon emissions, eco-friendly computing, energy efficiency, green computing, green
computing area, sustainable development.

I. INTRODUCTION since gained widespread acceptance in the IT industry [1].

Green computing is a concept that emerged in response to It refers to the design, development, use, and disposal of com-
growing concerns about the negative environmental impact puting systems in an environmentally sustainable way [2].
of information and communication technology (ICT). Green In other words, green computing aims to reduce the com-
computing term was coined in the early 1990s, and it has puting systems’ carbon footprint while ensuring their optimal
performance [3].
The associate editor coordinating the review of this manuscript and The phrase ‘‘green computing’’ can also be referred to
approving it for publication was Jie Tang . as the environmentally-friendly aspects of cloud computing,
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.
For more information, see https://creativecommons.org/licenses/by-nc-nd/4.0/
VOLUME 11, 2023 87445
 S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

such as high resource usage, cost savings from not having development of energy-efficient distributed computing sys-
to construct as many computers, and environmental pro- tems, such as cloud computing and virtualization, to reduce
tection [4]. Green computing has several benefits, includ- energy consumption and promote environmental sustainabil-
ing reducing energy consumption, minimizing electronic ity. Finally, other ICT direct sustainability includes a range
waste, and lowering operational costs [5]. The domain of of practices, such as sustainable software design and green
green computing includes various aspects of computing sys- procurement, aimed at reducing the environmental impact of
tems, such as hardware, software, networks, and data cen- ICT systems. The green ICT taxonomy plays an important
ters. Green computing practices involve optimizing hardware role in promoting environmental sustainability in the digital
design, reducing power consumption, using renewable energy age and advancing the field of green computing.
sources, improving software efficiency, virtualizing servers, Green computing differs from traditional IT systems in
and managing e-waste [6], [7], [8], [9]. Green computing several ways. Traditional IT systems focus on performance
has both positive and negative impacts. On the positive and functionality, whereas green computing emphasizes envi-
side, green computing can reduce energy consumption, lower ronmental sustainability. Traditional IT systems may use
operational costs, and enhance environmental sustainability. energy-intensive hardware, consume high amounts of power,
Furthermore, green computing can enhance the effectiveness and generate significant e-waste. In contrast, green com-
and dependability of computing systems by optimizing hard- puting systems use energy-efficient hardware, consume less
ware and software. On the negative side, green computing power, and generate minimal e-waste. Furthermore, tradi-
may require additional investments in energy-efficient hard- tional IT systems may rely on non-renewable energy sources,
ware and renewable energy sources. Furthermore, e-waste such as fossil fuels, whereas green computing systems use
disposal and implementing green computing practices may renewable energy sources. Green computing is good for
require additional training and resources. Green computing the environment as it reduces carbon emissions, minimizes
is a broad field that encompasses a range of technologies and electronic waste, and promotes renewable energy sources.
practices that aim to mitigate the environmental impact of By adopting green computing practices, organizations can
computing systems. Furthermore, there are several linkages improve their environmental sustainability and reduce their
between green computing and other technological fields or carbon footprint [12], [13], [14]. Furthermore, green comput-
domains. ing can lead to cost savings, improved energy efficiency, and
Green ICT and green computing are closely related con- enhanced CSR profile. If we do not adopt green computing,
cepts that share a common goal of promoting environmental it can lead to increased carbon emissions, e-waste generation,
sustainability. Green ICT refers to the use of ICT to improve and environmental degradation. Furthermore, organizations
environmental sustainability, while green computing focuses may incur higher energy costs, regulatory fines, and rep-
on the design, development, use, and disposal of computing utational damage. By adopting green computing practices,
systems in an environmentally sustainable way. Both con- organizations can avoid these negative outcomes and pro-
cepts are critical for promoting environmental sustainability mote environmental sustainability. The necessity of green
in the digital age, and they complement each other in several computing is driven by the increasing use of ICT in various
ways. Such green ICT can facilitate the deployment of green sectors, such as healthcare, finance, education, and manu-
computing solutions by enabling remote access, virtualiza- facturing [15], [16], [17]. ICT accounts for a considerable
tion, and cloud computing. Green computing, on the other amount of the world’s energy consumption and carbon emis-
hand, can enhance the energy efficiency and sustainability of sions and its use is projected to increase quickly over the next
ICT infrastructure by optimizing hardware and software [10]. few years. Therefore, it is essential to adopt green computing
Moreover, both concepts can contribute to the development practices to reduce the negative environmental impact of ICT
of smart and sustainable cities by promoting energy-efficient and promote sustainable development.
buildings, transportation systems, and public services. Green We have taken the initiative to explore and study the area
ICT taxonomy is a classification system that is used to cat- of green computing. This study holds a prominent position
egorize the different aspects of environmental sustainability within the current technical state of the art in green com-
in ICT. This taxonomy is essential for identifying the various puting due to its innovative contributions and comprehensive
areas of focus that are critical for promoting environmental analysis. By critically examining the existing literature and
sustainability in the digital age. One way to classify the synthesizing key findings, this research adds significant value
different areas of green ICT is to divide them into three broad to the field. Its novel insights, methodologies, and empir-
categories: data center-related sustainability, distributed ICT- ical evidence make it a valuable resource for researchers
related sustainability, and other ICT-direct sustainability [11]. and practitioners seeking to advance the frontiers of green
Fig 1., illustrates the green ICT taxonomy. Data center-related computing. Furthermore, by addressing emerging challenges
sustainability refers to the usage of energy-efficient data and identifying novel solutions, this study paves the way for
center technologies, such as cooling and power man- future advancements and sets a new benchmark for excellence
agement systems, to reduce data centers’ environmental in the field of green computing. Some of the contributions that
impact. Distributed ICT-related sustainability focuses on the this review study has made include the following:

87446 VOLUME 11, 2023

S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

FIGURE 1. Green ICT taxonomy.

I. This study has identified the area of green computing strategies, methodologies, and recommendations can guide
and thoroughly analyzed the current research works the design, implementation, and management of energy-
in each area. efficient computing systems and infrastructure. Thus, the
II. Various technologies and mechanisms that are used findings of this study may serve as a catalyst for further
in the area of green computing have been explored research in this area.
and the current research trends and most optimal The structure of a review study is crucial to convey infor-
techniques for achieving green computing have been mation to readers effectively. This study is organized into
filtered. seven sections that follow a logical and structural flow to
III. This study presents a thorough analysis of green explore various aspects of green computing. Section I of
computing-related research work, where the most sig- this study constitutes the introduction, providing an initial
nificant contributions and findings, utilized data set overview of green computing. Section II contextualizes the
and testing mechanism, and current research advance- review by discussing related studies and identifying the
ment are presented and summarized. present research gap. Section III outlines the review method-
IV. The study proposed state-of-art questionnaires that ology, which includes the selection procedure of the studies,
uncover the current obstacle and achievement and and research questions. In section IV, the selected study
summarize various other perspectives of green distribution is analyzed, followed by the exploration of areas
computing-related research works. of green computing in section V. Section VI presents the
V. This study provides future research directions and findings and analysis of the reviewed studies, highlight-
challenges for researchers in relevant fields. ing the challenges that emerge in the research on green
The research outcomes offer practical implications for computing and proposing future research directions. Finally,
industry professionals, policymakers, and researchers work- in section VII, the study has been concluded.
ing in the field of green computing. This study facilitates
the establishment of a framework for forthcoming research II. RELATED STUDIES
in the field of green computing by identifying potential In recent years, numerous studies have been performed in the
areas of investigation. As such, further study on this subject green computing area. Furthermore, various studies have ana-
can enhance understanding of green computing and aid in lyzed the uses of green computing and reviewed the current
the further generalization of green computing areas. This studies performed in the area of green computing, which are
study provides a comprehensive analysis of the industry-wide discussed here to identify and explore the current research
shift towards sustainable computing practices, including the gap.
adoption of innovative solutions aimed at mitigating the Dhaini et al. [18], conducted a review study on green com-
environmental impact of computing systems. The identified puting based on five areas such as software engineering,

VOLUME 11, 2023 87447

 S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

cloud computing, mobile computing, data centers, and the analysis of various green computation integration-based stud-
educational sector. In the study, a systematic literature review ies with respect to various fields has not been performed.
of every area has been conducted, highlighting each area’s Sagar and Pradhan [23], in another study, identify the
current research trends, and limitations. However, there has recent trends in green computing. In the study, various algo-
been a lack of discussion regarding the utilization of datasets rithms used in the field of green computing have been iden-
and their characteristics, testing mechanisms, and contri- tified. Furthermore, various perspectives on those algorithms
butions in relation to various perspectives on the current have been discussed. Lastly, a comprehensive analysis of the
technologies’ applications in each area. current limitations has been outlined.
As green computing is strongly associated with other Due to the increased demand for cloud computing, energy
domains or fields, examining the association between the two consumption is increasing, which can lead to an increase in
fields is necessary to achieve environment-friendly modern carbon footprint. Therefore, it is essential to explore tech-
computing systems. Castillo and Melin [19] have reviewed niques that may reduce future energy consumption by utiliz-
the interactions between green computing and computational ing energy efficiently. Aslam and Kaur [24] reviewed studies
intelligence. The study reviews and summarizes the current that proposed energy efficiency techniques in the green cloud
studies performed to find computational intelligence that is computing area. In the study, various techniques that can
environmentally responsible and eco-friendly. In the study, minimize high operational costs and be more energy effi-
a total of 255 studies have been reviewed, and various distri- cient, reducing the carbon footprint, have been explored and
butions related to the reviewed studies have been presented. analyzed. Furthermore, future research guidance has been
Furthermore, the study has analyzed the current trends of provided by analyzing the current relevant studies. Nonethe-
the existing works and the possibilities of future research less, only a small number of studies have been analyzed to
trends. Nevertheless, the current technologies utilized in each uncover energy-efficient cloud computing techniques. Addi-
study, along with their corresponding impacts, approaches, tionally, other green computing areas that can be associated
and limitations, have not been underscored. with green cloud computing, such as green data centers and
Hypotheses are the first step toward building and test- green IoT, have not been explored.
ing the feasibility of any new technology. By implementing The review of studies related to the implementation of
and analyzing the hypothesis, advanced and time-appropriate green cloud computing is performed by Radu [25]. In the
applications or systems can be built. Hamizi et al. [20] have study, recent green cloud computing-based studies have been
explored the current hypotheses in the green computing field. summarized, and environmental issues have been discussed.
In the study, the past 11 years of studies have been analyzed, The present study involved a comprehensive examination of
and the associated green computing hypotheses are outlined. 90 studies. An in-depth analysis consisting of green cloud
Initially, 340 studies had been considered for the study. How- computing benefits, research trends, and current obstacles
ever, after analyzing the studies, plausible hypotheses were has been discussed. Furthermore, future research directions
found in only 78 studies. The hypotheses have been catego- and challenges have been presented. Despite the extensive
rized into three groups such as tools for estimation, design nature of the investigation, the study falls short of providing
choice, and platform issues. The review studies are organized a comprehensive and comprehensible account of the methods
into taxonomies, and a comprehensive outlook on potential employed, dataset utilization, testing mechanism, and contri-
avenues for future research is provided. bution of each individual study. Moreover, the investigation
Another study performed by Khelifi et al. [21] analyzed has not delved into the association between green cloud com-
the economic and social contribution of green comput- puting and other areas.
ing with 5G and blockchain. Studies mainly published In another study, Yuan et al. [26] provided a detailed
from 2016 to 2020, have been analyzed, and a comprehensive review of research relating to data centers and multimedia
analysis of the study’s various perspectives has been detailed services. In the study, data centers and cloud-based multi-
in the study. A taxonomy of green blockchain and green media services, energy efficiency techniques, and approaches
computing based on 5G has been presented, and a power have been analyzed. In the study, various perspectives on the
consumption solution for blockchain 5G computation has energy management of data centers and multimedia services
been discussed in the study. Moreover, various limitations and have been addressed by providing a systemic literature review
future research directions have been detailed. However, other of each perspective. Further, the current challenges in main-
technologies that are closely related to 5G communication taining and designing data centers, as well as future research
technology and blockchain, such as IoT, data centers, and opportunities for green streaming services, have been out-
cloud computing, have not been presented. lined.
In another study, Geetha [22] reviewed the application Benhamaid et al. [27] have performed a comprehensive
of green computing with bots. In the study, green comput- review of studies on energy management techniques used to
ing applications across many sectors have been analytically achieve green IoT. In the study, current challenges in the field
described. Green computing’s potential fields, with their lim- of green IoT for energy consumption have been presented,
itations, have been deliberated. However, a comprehensive followed by the approaches that various studies have utilized

87448 VOLUME 11, 2023

S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

for energy management. The study provides a comprehen- energy administration, assimilation of renewable sources, and
sive overview of recent energy management systems for the integration of the smart grid in relation to other domains of
IoT ecosystem. Furthermore, the present research trends and environmentally-friendly computing, have not been explored.
future perspectives have been described for energy manage- As green computing is an emerging field, limited studies
ment to achieve green IoT. have been conducted in this field so far. As such, the outline
Reviewing implementation studies can provide insight, of proper green computing areas has not yet been generalized.
identify current challenges, and provide directions for poten- Therefore, to understand the progress of green computing-
tial future research work not only in the green computing field related work, it is necessary to explore all the potential green
but also in other fields. For instance, Showmick et al. [28] computing areas. Furthermore, to get a clear understanding
summarized the machine learning (ML) and deep learning of current research achievements and the challenges that
(DL) technologies utilized during the COVID-19 pandemic. researchers face while performing green computing-related
The study delves into six research questions that explore research work, there is a need to examine and explore all
different aspects of ML and DL in relation to the COVID-19 the potential areas regarding green computing. This study
pandemic. Furthermore, the study has outlined current chal- proposed the potential areas of green computing and analyzed
lenges and provided future directions, which can be essential the studies with respect to each area to acquire an in-depth
for combating the COVID-19 pandemic. knowledge of current trends and limitations as well as furnish
Mekala and Viswanathan [29] have conducted a sur- researchers with appropriate guidance for future research
vey of X-IoT applications in green computing, where X directions.
represents industrial/agriculture/environment. In the study,
energy-efficient sensors and virtual machine (VM) selection III. REVIEW METHODOLOGY
approaches have been presented. Based on the three research As was described by Brereton et al. [33], a review of studies
questions, review studies have been selected. Out of an initial is a method that involves locating, evaluating, and interpreting
sample of 160 studies, 72 were selected based on their quality, all of the data that is readily available on a certain research
and 32 were subsequently identified as potential sources subject or area of interest. In this investigation, in response to
for addressing the research questions. Furthermore, energy- a number of different robust research questions, an exhaus-
efficient sensors with respect to various technologies and tive search of the relevant literature has been carried out.
approaches have been discussed. In addition, VM selection In addition to that, a method that is risk-free, reliable, and
approaches based on various criteria have been presented. measurable has been used in order to offer a response to those
Finally, the challenges and open issues have been discussed. issues.
However, this study contains a limited number of studies and
emphasizes specific IoT concerns. A. SEARCH STRATEGY
As COVID-19 led to virtualization, Lee et al. [30] have Finding relevant research required a number of different
reviewed recent studies on green data centers. In the study, academic search engine usage, including Scopus, ERIC,
recent engineering-oriented studies on green data centers PubMed, DOAJ, Science Direct, Web of Science, IEEE
(GDC) have been reviewed. In the study, the authors found Xplore, and Google Scholar. Table 1 provides a summary of
that current technical efforts to create energy-efficient DCs the keywords that were used to retrieve the relevant works
are unable to achieve their goal. Furthermore, although vir- from the database. The majority of the time is spent utilizing
tualization may help reduce negative environmental effects, these keywords in a variety of different combinations to select
GDC needs to be eco-friendlier. the relevant studies.
In another study, a green procurement bibliometric review
was performed by Masudin et al. [31]. In the study, the imple- B. INCLUSION AND EXCLUSION CRITERIA
mentation of green procurement through supplier selection The criteria of inclusion and exclusion for evaluating the
has been analyzed using 220 studies from 1994 to 2022, and relevant studies are as follows:
numerous findings have been identified. However, this study Inclusion Criteria:
doesn’t cover the various perspectives related to the area of
a. Only studies are written in English that were relevant to
green procurement.
green computing.
Massaoudi et al. [32] have explored the usage of DL
b. Studies that include one or more green computing-
in smart technology-based studies through a systematic
related ideas.
approach. The review and analysis have been conducted
c. Studies that address at least one of the environmental
on studies conducted between 2015 and 2020, with a total
concerns related to green computing.
sample size of 220 studies. The study examines the imple-
d. Studies include either experimental or analytical work
mentation and integration of DL in smart grid systems and
and solving any green computing area problem.
outlines DL’s major applications in smart grids. Further-
more, many challenges associated with the application and Exclusion Criteria:
development of DL for smart grids are discussed. However, a. Research articles not published before 2007 and after
alternative viewpoints regarding the smart grid, including 2022.

VOLUME 11, 2023 87449

 S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

TABLE 1. Applied keyword.

b. Green computing-related methods are discussed in ‘‘Finding and Contribution’’, mentions the contribution, uti-
research journals but are not associated with considered lized procedures, and findings of the study.
green computing areas.
c. Green computing issues are mentioned in research that E. RESEARCH QUESTIONS (RQS)
does not employ any green computing-related technolo- This exhaustive and in-depth study focuses mostly on sum-
gies or any analytical work. marizing, analyzing, and synthesizing the numerous studies
d. Studies that review others’ green computing-related that have been performed in the field of green computing.
research work. The primary objective of this investigation is to collect the
appropriate response to the following seven research ques-
C. SELECTION OF THE STUDY tions through an in-depth analysis and provide an informative,
The selection of appropriate studies is essential to conduct a insightful assessment and response. The following are the
high-quality systematic review. Therefore, the selection of proposed research questions:
studies for the review should be performed with the utmost • RQ 1. What are the most recent advancements in the area
care and following a standardized procedure. This study’s of green computing research, and how do they stack up
primary goal is to systematically identify and evaluate the against earlier efforts?
available studies on green computing. • RQ 2. What approaches have shown to be the most suc-
The selection of studies has been carried out through cessful implementation in the areas of green computing?
multiple steps outlined in Fig 2. To begin with, an initial • RQ 3. Which areas of green computing are most under-
pool of 800 studies had been identified using the search developed, and what are the obstacles researchers in
strategy. Duplicates had been removed, leaving 680 unique those areas are facing?
studies for further screening. Abstract and dataset analysis, • RQ 4. How do recent developments in artificial intelli-
along with inclusion and exclusion criteria, had been used to gence and machine learning affect the amount of energy
narrow down the eligible studies to 328 for full-text review. saved by computer systems?
Of these, 87 studies were removed after examination, result- • RQ 5. What areas of green computing research are the
ing in 241 studies for methodological quality evaluation. most promising?
After careful evaluation, 25 studies were excluded, and only • RQ 6. What will be the most difficult obstacles for those
216 studies had been included in the systematic review. researchers who are now intending to do research on
The selection process helps to ensure that the studies green computing?
included in the review are of good quality and relevant to the • RQ 7. What role does green computing research and
proposed research question. This study employed a rigorous development play in achieving the aims of sustainability
selection procedure, including the use of predefined inclusion across the world?
and exclusion criteria and multiple levels of screening to
identify the most appropriate studies.
IV. STUDY DISTRIBUTION ANALYSIS
Before analyzing the studies’ contributions and achievements
D. EXTRACTION OF THE DATA towards green computing, it is important to analyze the distri-
After studies have been selected, the data extraction step bution of the final selected studies. This section presents the
is very vital for accurate analysis and interpretation of the final selected studies’ various distributions.
research work. To achieve useful results, the extraction of From Fig 3., it can be observed that the China per-
data from studies has to follow a logical structure. Therefore, formed the highest number of green computing studies,
tables with predetermined attributes have been generated, with sixty-four studies published in the field. India is the
and diverse data acquired from the study endeavor has been second-highest study publisher country with twenty-eight
filled. The first attribute, ‘‘References and Year’’, of the tables research works, followed by the USA with twenty research
contains the study reference and the publication year. works. Fourteen studies are contributed by Qatar, Lebanon,
The next attribute, ‘‘Perform work’’, indicates the subject Ireland, Brazil, Malta, etc., referred to as ‘‘Others’’. Other
of the research works. The third attribute, ‘‘Dataset/Testing nations, most of which are developing countries, have con-
mechanism’’ describes the dataset characteristics and utilized ducted fewer studies on green computing. While some coun-
testing mechanisms of the studies. Lastly, the last attribute, tries have contributed more than others, it is reassuring to see

87450 VOLUME 11, 2023

S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

FIGURE 2. Diagram of the study selection procedure.

FIGURE 4. Distribution of green computing studies by year.

FIGURE 3. Country-wise distribution of green computing area.

(see Fig 4.), the year 2007 observed one study published in
the field, which had been followed by two studies in 2008 and
that nations worldwide are committed to developing a greener two in 2009. The number of publications started to rise in
future through technology. 2010, with four studies published that year. In 2011, two
Green computing has been a rapidly growing field, with studies were published, and in 2012, the number increased
a significant increase in research output in recent years to ten. The trend continued, with twenty studies published

VOLUME 11, 2023 87451

 S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

FIGURE 5. Word cloud of studies keyword.

FIGURE 6. Area-wise distribution of green computing studies.

in 2013 and twenty-four in 2014. In 2015, there had been
fifteen studies published, and the following year, 2016, there
were twenty-three studies published. In 2017, fourteen stud-
ies were published, and the number increased to eighteen in
2018. The year 2019 saw twenty-one studies published, and
the number increased further to twenty-four in 2020. The year
2021 had twenty-two publications, while fourteen research
works were published in 2022. Overall, the field of green
computing has seen a steady increase in research output since
2007. The peak of this increase has been observed in the
last decade, with the highest number of publications in 2020.
This increase in research output is indicative of the growing
interest and importance of the field, with researchers actively
FIGURE 7. Distribution of journal publication related to green computing
contributing to developing sustainable computing solutions. studies.
Fig 5. Illustrate the word cloud of selected studies
keywords. The keyword analysis demonstrates that the key-
words ‘‘data-center’’, ‘‘green’’, ‘‘energy’’, ‘‘green cloud’’,
‘‘cloud computing’’, ‘‘computing’’, ‘‘network’’, and ‘‘opti-
mization’’ were among the most commonly used. These
keywords represent various aspects of green computing,
such as energy-efficient data centers, cloud computing ser-
vices with reduced environmental impact, and optimization
techniques for energy-efficient computing. On the other
hand, the keywords ‘‘power’’, ‘‘management’’, ‘‘energy con-
sumption’’, ‘‘green procurement’’, ‘‘green transportation’’,
‘‘virtualization’’, and ‘‘smart grid’’ were among the least
commonly used. These less frequently used keywords reflect
the lesser emphasis on areas such as green transportation and FIGURE 8. Distribution of conference publication related to green
energy-efficient procurement in green computing research. computing studies.
Overall, the analysis of keywords used in green computing
research provides insights into the research trends and prior-
ities in this field. is published in journals, while the remaining around 37%
Over the years, research in the green computing field has are published in conferences. The preference for journal
led to the development of various areas, each with its own publications over conference studies indicates the growing
set of challenges and opportunities. From Fig 6., it can be maturity and stability of the field. Fig 7., presents the journal
concluded that the most prominent areas in terms of col- publication house of green computing’s final selected studies.
lected studies are GDC, green cloud computing (GCC), green Elsevier, Springer, IEEE Xplore, and MDPI are the leading
transportation systems, and green IoT. On the other hand, journal publishing houses in the area of green computing.
green machine learning, green data mining (DM) and infor- However, a number of other publishing houses, referred to
mation retrieval (IR), green e-commerce and e-government, collectively as ‘‘Other,’’ play an important role in the publi-
and green databases are some of the most unexplored areas cation of green computing studies. On the other side, Fig 8.
and contain fewer studies. depicts the publishing houses of conference studies, with
In the analysis of publication trends in green comput- IEEE Xplore playing the dominant role, followed by ACM
ing, it is observed that around 63% of the research work and other conferences.

87452 VOLUME 11, 2023

S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

FIGURE 9. Categorization of green computing areas.

V. AREAS OF GREEN COMPUTING and information retrieval that are energy-efficient and can
Studies of green computing try to lessen the environmental help lessen the environmental impact of data processing and
effect of ICT. Green computing covers a wide variety of areas decision-making. The field of green machine learning focuses
(see Fig 9.). The selection of twelve areas in green com- on machine learning methods that are energy-efficient and
puting is driven by their scope, relevance, and the need for can help lessen the environmental impact of data processing
comprehensive coverage of key aspects within the field. After and decision-making. The field of green e-commerce and
gathering numerous research studies, a rigorous application green e-government focuses on environmentally friendly and
of inclusion-exclusion criteria resulted in the finalization of resource-conserving methods of performing e-business and e-
relevant studies. During the in-depth review process, relevant government operations. The sustainable purchase of products
studies were classified into the same categories, resulting in and services is the focus of the green procurement area. The
the formulation of twelve areas of green computing that are field of green agriculture focuses on using ICTs to enhance
more likely to facilitate the alignment of forthcoming studies sustainable agricultural methods and decrease farming’s envi-
with one of the twelve specified areas. This approach ensures ronmental effects. The field of green education focuses on
depth and focus in examining specific challenges, advance- using ICTs to encourage environmentally friendly teaching
ments, and implications associated with each area, while methods. The field of smart grids focuses on using ICTs
also considering practicality, manageability, consensus, and to optimize and control the electrical grid in a sustainable
the existing body of research. These identified areas include and energy-efficient manner. The field of green transportation
GDC, GCC, green IoT, green database, green DM and IR, systems focuses on the use of ICTs to manage and improve
green ML, green e-commerce and green e-government, green transportation networks in a manner that is both sustainable
procurement, green agriculture, green education, smart grid, and energy-efficient. The overall focus of green comput-
and green transportation systems. Each of these areas has ing research is to reduce the environmental impact of ICT
its own unique challenges and opportunities for promoting systems and to encourage environmentally friendly business
environmental sustainability in the digital age. By focusing practices across a range of sectors, from data centers and
on these areas, researchers and practitioners can contribute to cloud computing to transportation and agriculture.
the development of sustainable computing systems and help
mitigate the negative environmental impact of ICT. The field A. GREEN DATA CENTER TECHNOLOGIES
of green data center technologies focuses on creating and run- GDC are built with a focus on energy efficiency and low envi-
ning data centers in an environmentally friendly and resource- ronmental impact. Reducing energy use and carbon footprint
conserving manner, using renewable energy sources, and while preserving or even enhancing performance and avail-
reducing carbon emissions. In the area of green cloud com- ability is the aim of a GDC. To reduce the need for physical
puting, energy efficiency, electronic waste reduction, and servers and improve energy efficiency, server virtualization is
increased reliance on renewable energy sources are explored one of the primary technologies used in GDC. Server virtual-
as ways to leverage cloud computing to lessen the environ- ization enables numerous virtual servers to operate on a single
mental impact of IT infrastructure. The green IoT sector physical server. This is accomplished by dividing a single
examines how to leverage low-power gadgets and improved physical server into many virtual servers, each of which may
communication protocols to make the IoT more sustainable run its own operating system and applications using virtual-
and energy-efficient. The design, usage, and administration ization software. According to studies, server virtualization
of databases with regard to energy efficiency is the emphasis may result in energy savings of at least 50% [34]. Another key
of the green database field. The field of green data mining technology utilized in green data centers is power manage-
and information retrieval focuses on methods for data mining ment. Using renewable energy sources and dynamic voltage

VOLUME 11, 2023 87453

 S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

and frequency scaling, which adapts server power consump-

tion depending on workload, are two power management
strategies that may help data centers use less energy and have
a less negative effect on the environment [35]. Furthermore,
a combination of traditional and renewable power sources
can effectively reduce CO2 emissions while preserving power
security [36].
Table 2, presents some of the key elements of the current
studies conducted in the area of the GDC. In the area of
GDC, various types of datasets, testing software, and data
centers have been utilized for conducting the studies. Among
all the testing mechanisms, CloudSim is the most utilized
FIGURE 10. Country-wise distribution of green data center studies.
software. However, various types of datasets with respect
to the study’s objective have been utilized. From Table 2,
it can be observed that diverse types of work, such as cool-
ing system development, the design and implementation of studies have been conducted in this context to ensure the
green architecture, various types of algorithm development, maximization of the green data center profit [46], [47].
and server virtualization have been performed in the area of Lastly, data centers that are run by green energy sources
the GDC in recent years. The cooling system is an essen- such as biomass, solar, wind, and geothermal energy will
tial component of green data centers. Green data centers assist in lowering the data centers energy consumption as well
employ methods like liquid cooling and outside air cooling to as the carbon footprint. The needs and peculiarities of various
decrease energy usage. Data centers need extensive cooling data centers should be taken into consideration, since the best
to avoid overheating. Traditional air-conditioning systems, technologies and solutions will vary based on the particular
which may be a significant source of energy use in data situation. Data center operators must thus carefully evaluate
centers, may not be as energy-efficient as these alternatives. their unique needs and requirements before introducing green
Using the nature-based cooling system mechanism can effi- data center solutions.
ciently tackle the overheating problem while reducing energy From Fig 10., it is observable that China and USA lead
requirements compared to traditional systems [37]. Another the area with nine and eight studies, respectively. How-
crucial component of green data center is their energy- ever, other countries are also actively researching sustainable
efficient architecture. Data centers may be built to consume data centers, with India and Malaysia each having three
less energy and leave a lower carbon footprint by improving research works. Singapore and Vietnam each having two.
airflow and cutting energy usage. Raised flooring, hot aisle Other countries on the list include France, Mauritius, Italy,
and cold aisle layouts, and other design elements may be used Iran, Saudi Arabia, Pakistan, Hong Kong, Qatar, Indonesia,
to accomplish this. Furthermore, architecture that is capable Bangladesh, and Taiwan, which each have one green data
of online monitoring, live VM migration, or maximizing center-related study, which indicates the global interest in
network power conservation can be implemented to reduce the development of sustainable green data center technology.
energy consumption [38], [39]. Evaluating an architecture Fig 11. illustrated the year-wise study publication in the area
for the GDC is essential for ensuring the effectiveness of of GDC. Starting from the lowest, the year 2009 had only one
the architecture or framework. Various studies have proposed research work published, followed by 2010, and 2012, which
simulators that can be utilized to validate and identify the also had only one research work each. The most recent year,
system’s effectiveness [40]. Storage space optimization is one 2022, had a single research work published, while 2020 and
of the key attributes for achieving a GDC. The multi-valued 2021 had two research works each. The years 2015 and
table structure is an important criterion to optimize the stor- 2019 had three research works of each published. Finally,
age space, thus achieving a GDC [41]. The GDC researcher 2013 had the highest number of research works, with eight
should be concerned not only with decreasing CO2 emis- published.
sions and energy consumption, but also with strengthening
the data center’s security while minimizing cost, which can B. GREEN CLOUD COMPUTING
be accomplished by a distributed data center or by building The purpose of GCC is to lessen the negative effects of
a cyber security solution [42], [43]. Furthermore, develop- cloud computing (CC) on the environment by using sustain-
ing energy-aware algorithms and task schedulers can aid able and energy-efficient methods. Data centers, which are
in optimizing the data center network, and reducing energy significant energy users, are growing in number due to the
consumption is one of the major key components towards increasing popularity of CC. This has caused worries about
achieving a GDC [44], [45]. Profit maximization is an impor- the rising environmental effects of cloud computing. With
tant aspect for industrialists and financial institutes when the fundamental principle of sustainable development, the
developing and investing in green data centers, and various notion of ‘‘green computing’’ first emerged in 1987, the main

87454 VOLUME 11, 2023

S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

TABLE 2. Summary of green data center technologies studies.

VOLUME 11, 2023 87455

 S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

TABLE 2. (Continued.) Summary of green data center technologies studies.

87456 VOLUME 11, 2023

S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

TABLE 2. (Continued.) Summary of green data center technologies studies.

the quality of services (QoS). Allocation of resources and

management of power using algorithms can optimized server
response time, meet service level agreements, and improve
power efficiency [75], [76]. Task scheduling is one of the
key components of cloud computing, therefore, optimization
and improvement of task scheduling algorithms can aid to
achieved green cloud computing. Using various mechanism
based task scheduling algorithms, resource load balancing
can achieve, and execution time and resource consumption
can reduce while delivering better performance [77], [78],
[79]. Furthermore, job scheduling algorithms can save energy
usage, and improved resource utilization. Cloud computing’s
other key components such as VM migration, using various
FIGURE 11. Year-wise distribution of green data center studies.
techniques such as Ant Colony Optimization and constraints
such as traffic-aware VMs migration can reduce energy usage
and improve resource utilization [80], [81]. These various
aim is to lower energy usage. Green computing refers to the procedures may improve network traffic, cut down on data
environmentally friendly use of computers and associated center energy use, and lessen the environmental impact of
technology [74]. Energy consumption and efficiency, power cloud services. Organizations may lessen their negative envi-
management, greenhouse gas and CO2 emissions, virtualiza- ronment effects and increase the overall sustainability of their
tion of servers, and natural resources preservation are some operations by employing green cloud computing techniques.
of the main issues with cloud computing. By using ecological This may include lowering the energy requirements of data
and energy-efficient approaches, GCC tackles these issues. centers, lowering the environmental impact of cloud services,
From Table 3, which presents the current studies con- and raising the general energy effectiveness of cloud comput-
ducted in the area of GCC, it can be observed that the ing.
majority of studies have utilized the CloudSim framework Furthermore, the development of an energy-efficient
for evaluating the proposed solutions. In the most recent framework and management system for cloud computing can
years, there has been a proliferation of research in the devel- assist to achieve green computing. Green cloud computing
opment of energy-efficient hybrid frameworks, as well as has advantages for the economy as well as the environment.
the creation of algorithms for VM migration and selection Organizations may reduce operational expenses and boost
policies, and optimization algorithms. GCC studies have profitability by lowering their energy use.
considered renewable energy sources usages to power data Fig 12. illustrates the country-wise green cloud com-
centers, the optimization of data center energy consumption puting study distributions, which include eighteen different
through server virtualization, framework modeling, power countries. Among the countries, the Netherlands, Australia,
management strategies, effective cooling systems, task and Italy, Mauritius, France, Bangladesh, Singapore, Lebanon,
job scheduling algorithms, the creation of energy-efficient South Korea, Ireland, and Brazil have one study each on
algorithms, and protocols for cloud computing. Renewable the area of green cloud computing. Meanwhile, the USA,
energy usages to power up the data center responsible for pro- Finland, Saudi Arabia, and Egypt have two studies each. Iran
viding cloud computing service is one of the basic and effec- stands out with three research works, while China and India
tive approaches. The usage of various cost and energy-aware have the highest number of publications with six studies each.
algorithms can reduce energy consumption while maintaining The research efforts of these countries highlight the growing

VOLUME 11, 2023 87457

 S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

TABLE 3. Summary green cloud computing studies.

87458 VOLUME 11, 2023

S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

TABLE 3. (Continued.) Summary green cloud computing studies.

VOLUME 11, 2023 87459

 S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

FIGURE 14. Country-wise distribution of green IoT studies.

FIGURE 12. Country-wise distribution of green cloud computing studies.

energy-efficient protocols and algorithms that may be used

on IoT devices and networks is one of the main difficulties
in this sector. Different strategies have been put forward by
researchers to lower the energy use and carbon footprint of
IoT networks and devices [108].
Table 4, presents the current studies performed in the field
of green IoT in recent years. In recent years, various con-
trol schemes and design methods, including energy-efficient
caching, spectrum sharing, and offloading computation stud-
ies, have been conducted. MATLAB and NS2 have been
commonly utilized software in the studies of green IoT. How-
FIGURE 13. Year-wise distribution of green cloud computing studies.
ever, based on the studies’ objectives, various other datasets
and testing mechanisms, such as sensor data, responses,
and software, have also been employed. The performed
importance of green cloud computing as an emerging field work in the area of green IoT includes developing control
and its potential to address the environmental challenges schemes, designing, energy-related management systems,
posed by the massive energy consumption of cloud comput- power management, monitoring systems, energy- efficient
ing services and systems. communication protocols, and green IoT networks. Using
Among the years presented in Fig 13., the least number of IoT systems, various control mechanisms, such as smart con-
studies were published in 2011 and 2021, with only one study gestion, energy consumption, and monitoring systems, have
each year. In contrast, 2014 had the highest number of studies, been developed. To reduce and optimize energy consumption,
with six publications. Years 2012, 2015, 2016, and 2020 had various energy management and harvesting schemes have
the highest number of studies after 2014, with four green been developed. To create a more sustainable future for IoT
cloud computing-related studies. The focus on sustainability systems, green IoT computing strives to achieve a balance
and eco-friendliness cloud computing is driving research in between environmental sustainability and technical growth.
this area, leading to new technologies development and strate- The field of green IoT is growing rapidly, as evidenced
gies for energy-efficient and environmentally-friendly cloud by the country’s involvement and the number of studies pub-
computing. lished on the subject. From Fig 14., China has emerged as
a leader in green IoT research, with ten studies published
C. GREEN IoT in the area. India is also making significant contributions,
Green IoT are the development and implementation with four studies published. Other countries have a smaller
of IoT systems using sustainable and energy-efficient but still noteworthy number of studies in the field. Further-
approaches. The usages of low-power gadgets, the adoption more, in 2007, 2015, 2017, and 2022, only one study per
of energy-efficient communication protocols, and the use of year has been published (see Fig 15.). Two studies had been
renewable energy sources to power IoT gadgets are some published in 2018, while four studies had been published in
examples of these best practices. It is crucial to create green both 2019 and 2021. However, the highest number of studies,
IoT techniques to reduce these devices’ negative environ- a total of nine had been published in 2020.
mental effects since the energy consumption in IoT devices This indicates a global interest in the potential of IoT to
has increased drastically. Green IoT may also lead to cost support sustainability initiatives.
reductions and enhanced system performance. Green IoT
computing is a field of study that deals with the difficulties D. GREEN PROCUREMENT
of lowering energy use and carbon footprint while preserving Green procurement is the process of buying products and
the performance and functionality of IoT devices. Designing services that have a less negative effect on the environment.

87460 VOLUME 11, 2023

S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

TABLE 4. Summary of green IoT studies.

VOLUME 11, 2023 87461

 S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

TABLE 4. (Continued.) Summary of green IoT studies.

businesses are already using green procurement as a means

of achieving sustainability objectives and staying compliant
with international environmental requirements [132], [133].
In order to achieve an ecologically sustainable future, green
procurement is critical, and it is also important to analyze the
currently performed studies in the area.
Table 5, presents the current studies conducted in the area
of green procurement. In the most recent years, various
factors that impact green procurement, such as leadership
style, top management support, and green procurement prac-
tices with respect to various industries, have been examined.
FIGURE 15. Year-wise distribution of green IoT studies.
By examining the overall studies, it is noticeable that for most
of the studies related to green procurement, the researchers
collected data from people or companies. Most of the studies’
Products with a smaller carbon footprint, items made of primary goal is to identify the drivers and barriers related to
recycled materials, and services that support sustainable prac- various industries and sectors. The furniture, real estate, con-
tices may all fall under this category. Green procurement’s struction, and photovoltaic industries’ green procurements
two main goals are to reduce the environmental impact of have been analyzed by the researcher, and the barriers and
the procurement process while also encouraging sustainable drivers for green procurement among those industries have
development. Along with helping businesses comply with been outlined. Lack of knowledge and awareness, lack of
regulations, green procurement may also help them gain green building materials, little market benefit, and insuffi-
favor with consumers and other stakeholders. Green pro- cient incentive policies are the major and most common barri-
curement methods include life cycle thinking, eco-labeling, ers to achieving green procurement in the real estate industry
and environmentally preferred purchasing. A product or ser- and construction projects. Furthermore, green procurement
vice’s environmental effect is taken into account throughout relations with leadership style, financial performance, buyer-
its entire lifespan, from the procurement of raw materials supplier relationships, and management support have been
through disposal. The process of eco-labeling involves con- analyzed by various studies. Various other works, such as a
firming that items adhere to strict environmental criteria. process model for green procurement based on IoT, supplier
Choosing goods or services that have less of an effect on selection, and government green procurement, have been
the environment than those that are already on the market performed in recent years.
is known as environmentally preferred purchasing. Organi- China leads the world with eight studies in the field of
zations may find it difficult to implement green procurement green procurement, followed by the United Arab Emirates
since it often calls for a change in perspective and in the way and South Africa with two studies each (see Fig 16.). Other
that procurement procedures are performed. However, it may countries with one study each include India, Italy, Malaysia,
also bring about financial savings and other advantages, like Thailand, Hong Kong, and Saudi Arabia. Overall, the stud-
a boost to reputation and more creativity. Additionally, many ies in the green procurement area are still limited, but the

87462 VOLUME 11, 2023

S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

TABLE 5. Summary of green procurement studies.

VOLUME 11, 2023 87463

 S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

TABLE 5. (Continued.) Summary of green procurement studies.

control systems, and then utilize that data to enhance grid per-
formance. One of the main advantages of a smart grid is that
it may enable utilities to better control the supply and demand
of power, which can assist in lowering energy usage and
prices [153]. Power usage trends can be observed and altered
by deploying smart meters that monitor power consump-
tion in real-time. This approach may lessen the requirement
for costly peak power production and increase the grid’s
general effectiveness. Smart grids also make it possible for
renewable energy sources like solar and wind power to be
integrated more effectively into the power system. Advanced
FIGURE 16. Country-wise distribution of green procurement studies.
control systems may be used in smart grids to automatically
alter the flow of power from renewable sources. This may
enhance renewable energy usage and decrease the demand
for fossil fuels. The capacity of smart grids to increase
the resilience and dependability of the electrical system is
another significant feature. Smart grids may utilize sophisti-
cated monitoring and control systems to swiftly identify and
react to power outages, decreasing the length and severity
of service disruptions. To increase the dependability of the
power supply during crises, the smart grid may also deploy
FIGURE 17. Year-wise distribution of green procurement studies.
distributed energy resources like battery storage and micro-
grids. In general, smart grids have the potential to greatly
diverse countries’ involvement demonstrates that interest in increase the effectiveness, dependability, and sustainability of
this area is growing globally. In the area of green procure- the electricity system. Furthermore, smart grids are crucial for
ment, research has been relatively limited, with only a few tackling environmental problems and can aid in sustainable
studies published each year (see Fig 17.). The earliest study development.
has been published in 2008, and there has been a steady Several smart grid studies have been conducted in recent
stream of research in this area over the years. In 2008, 2010, years that cover various aspects of smart grid systems.
2014, 2015, 2018, and 2020, a single study was published Table 6, presents the recent studies conducted in the area
on a variety of sustainable procurement-related issues. In of smart grids. In the most recent years, renewable energy
2017 and 2021, there were two research works published in trading schemes, downlink communication schemes, and
the area of green procurements. The years with the highest data processing frameworks have been developed. Various
number of studies published have been 2016 and 2019, with simulator-based software and numerical evaluations have
four studies published in each of these years. been performed to evaluate the smart grid system’s devel-
opment and implementation. Among the used software,
E. SMART GRID MATLAB is the most common, with other software such
A modernized electrical grid, commonly known as the as Castalia and Simulink being used based on the study’s
‘‘smart grid’’, makes use of digital technology to improve objective. Furthermore, various types of studies have been
the dependability, efficiency, and sustainability of the power conducted in the area of the smart grid, such as energy
system [152]. Smart grids collect and analyze power use data management, resource management, security requirements
using modern metering infrastructure, communications, and analysis, protocol and algorithm development, and frame-

87464 VOLUME 11, 2023

S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

the negative environmental effects of transportation. These

systems seek to increase overall sustainability in the trans-
portation industry while reducing carbon emissions and air
pollution brought on by conventional modes of transporta-
tion [172]. Electric and hybrid cars, clean fuel substitutes,
and intelligent transportation systems used to ease traffic and
boost energy efficiency are some of the technologies and
procedures employed in green transportation systems. Aside
from integrating public transportation systems, green trans-
portation systems can also incorporate active transportation
FIGURE 18. Country-wise distribution of smart grid studies. options such as bicycling and walking. Green transporta-
tion systems aim to provide the same level of accessibility
and mobility as conventional systems while reducing the
environmental impact of transportation. Green transporta-
tion systems include electric cars, bike-sharing programs,
and high-speed rail networks [173]. While offering a more
sustainable and energy-efficient alternative, these systems
have the potential to drastically reduce the carbon emis-
sions and air pollution caused by conventional transportation
systems.
From Table 7, it can be observed that various types of stud-
ies have been conducted in the area of green transportation.
In the most recent years, green transportation various policy
developments, factor analyses, scheduling problem solutions,
FIGURE 19. Year-wise distribution of smart grid studies.
urban transportation system development, and analyses of
the regional companies’ green transportation have been per-
work design. Energy management is one of the key attributes formed. To perform overall studies in this field, various
of smart grids, and various studies have been conducted types of datasets containing data about people’s opinions
focusing on the issue. Furthermore, resource management and surveys, field studies, traffic data, industry data, city
and distribution and residential energy management are some data, and CO2 emission data have been utilized. Moreover,
of the common applications of smart grid systems. Other various software such as MATLAB, TransCAD, HOMER,
studies focusing on developing efficient authentication pro- VISSIM, and GAMS has been utilized. Using those datasets
tocols, monitoring and controlling algorithms, and renewable and software, various types of research work, such as analysis
energy trading schemes have a great impact on achieving of the regional company adaptation of green transportation,
secure smart grid systems that can reduce energy consump- the cyber-physical bike system, social systems for vehicles,
tion and are environmentally friendly. However, the adoption strategy and framework development, ecological highway
of smart grids requires cooperation among several parties, development, algorithms, and mechanism development have
including the government, the business community, and con- been conducted in the area of green transportation. Various
sumers. green transportation policies have been taken into considera-
In the context of the smart grid, Germany, Pakistan, the tion in recent years, such as motorcycle bans and subsidizing
United Arab Emirates, and the United Kingdom have each electric scooters, which can reduce city pollution effectively,
contributed one study to this area of research (see Fig 18.). reduce social congestion, and deter social crime. Further-
China, France, and Singapore each have two studies on the more, to reduce the usage of traditionally fueled vehicles in
area, while India and the USA both have four studies. From the city, renting motorcycles for short distances and electric
Fig 19., it can be observed that research in the smart grid area vehicles for medium and long distances is a more environ-
has been ongoing for several years, with studies published in mentally friendly solution. Traffic management and ecolog-
various years. Four studies were published in 2013, while the ical highway development are some major factors in green
highest number of studies was published in 2014 with count transportation systems, which not only save non-renewable
of five studies. A single study has been published in 2015, resources, cost, and time but also ensure safe, eco-friendly,
whereas two were published in 2012. In 2016 and 2019, three sustainable development. Overall, researchers in this area
studies were published, respectively. are concentrated on the creation of new technologies and
regulations that promote the use of environmentally friendly
F. GREEN TRANSPORTATION SYSTEMS transportation methods, as well as the assessment of current
Green transportation systems are those that employ sustain- methods to ascertain their efficacy. Therefore, green trans-
able and energy-saving technology and procedures to reduce portation systems can aid in building a sustainable future and

VOLUME 11, 2023 87465

 S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

TABLE 6. Summary of smart grid studies.

87466 VOLUME 11, 2023

S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

technology and procedures. Due to the growth of technolo-

gies and data, green databases (GDB) are becoming more
popular and essential. The green database objective is to
reduce the negative effects of data management on the sur-
rounding ecosystem as much as possible while simultane-
ously preserving or enhancing the database’s performance
and availability. Green database techniques include the use of
energy-efficient hardware, the application of power manage-
ment methods, the use of cloud computing and virtualization
technologies, and so on.
Table 8, presents the studies conducted in the green
FIGURE 20. Country-wise distribution of green transportation systems
database area that solve various issues. However, due to
studies. relatively new and evolving fields, very limited studies have
been conducted in the area. Energy consumption benchmark
platform, GDB model and management for the stock mar-
ket, integration of GDB in software development, and GDB
assignment in green manufacturing are some of the topics that
have been explored by the researcher in the area of green
databases. Additionally, the green database also comprises
the use of sustainable data management strategies, such as
data compression and de-duplication, in order to reduce the
amount of energy that is necessary for the storing and retriev-
ing of data.

H. GREEN DATA MINING AND INFORMATION RETRIEVAL

Green data mining and information retrieval is the process
FIGURE 21. Year-wise distribution of green transportation systems
studies.
of gathering, storing, and analyzing data using ecologically
friendly practices. This entails using energy-efficient gear,
putting power management strategies into practice, as well
reducing CO2 and greenhouse gas emissions, thus minimiz- as employing renewable energy sources. Green data mining
ing and solving climate change issues. and information retrieval aim to reduce the environmental
Fig 20. illustrates the fact that various countries have per- impact of these activities while still delivering meaningful
formed green transportation-related research, where China insights from the data by implementing carbon footprint
dominates the area with a total study count of twelve, fol- management strategies into practice, as well as employing
lowed by the USA with four studies. Iran and Taiwan have renewable energy sources [208]. The sustainable disposal of
performed two studies each. On the other hand, other coun- electronic waste produced by data mining operations is also
tries have performed one study each. regarded as a component of green information retrieval and
In the area of the green transportation system, research data mining.
has been conducted over the years with increasing attention Table 9, presents the studies conducted in recent years
in recent times. From Fig 21., it can be observed that the to tackle green data mining and information retrieval area’s
least number of studies have been recorded in 2009, 2011, various issues. Cost-sensitive learning-based framework for
2012, 2014, 2016, and 2017 with one study each year, while data mining, the latency/power tradeoff model, CPU power
the year 2021 recorded the highest number of studies with management, the energy operational cost reduction model,
six publications. Similarly, 2018 had five studies, 2020 had data-aware and efficient data mining services and Web ser-
four, and 2013, 2015, and 2019 recorded two studies each. vice recommender systems, and the green data mining prin-
The results indicate that research in green transportation is ciple are some of the concepts that have been explored in
gradually gaining momentum, with a significant number of recent studies. Numerous strategies and methods for making
publications emerging in recent years, especially in 2018 and data mining more ecologically friendly have been studied, but
2021. Overall, the research indicates a global interest in various issues have not yet been properly addressed.
developing sustainable transportation systems to reduce car-
bon emissions and help mitigate climate change. I. GREEN MACHINE LEARNING
Green machine learning is the application of low-cost and
G. GREEN DATABASE efficient machine learning techniques to promote energy effi-
The phrase ‘‘green database’’ refers to the administration ciency and reduce carbon emissions in various industries.
and upkeep of databases that make use of eco-friendly This field of study aims to develop algorithms and models

VOLUME 11, 2023 87467

 S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

TABLE 7. Summary of green transportation system studies.

87468 VOLUME 11, 2023

S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

TABLE 7. (Continued.) Summary of green transportation system studies.

that can optimize energy consumption and reduce the envi- data preprocessing and models’ training, and incorporating
ronmental impact of machine learning systems. Research in environmental factors into the decision-making process of
this area includes developing energy-efficient architectures machine learning models. In recent years, green ML has
for neural networks, reducing the energy consumption of been applied in various industries, such as transportation,

VOLUME 11, 2023 87469

 S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

TABLE 8. Summary of green database studies.

agriculture, and manufacturing, to improve energy efficiency result from the use of green principles in online com-
and reduce carbon emissions. merce. Furthermore, incorporating sustainability principles
Due to a relatively new and evolving field, very lim- into e-commerce can have long-term positive effects.
ited studies have been conducted in the area, leaving Green e-government refers to the use of environmentally
most of its potential unexplored. Table 10, presents the sustainable practices by various organizations of government.
current studies conducted in the area of green machine This might include switching to paperless processes to cut
learning. Multi-information source optimization, green and waste, utilizing energy-efficient technologies in providing
energy-efficient hyperparameter optimization, knowledge- e-government services to lower emissions, and powering gov-
aware ML, an energy and carbon footprint analysis frame- ernment facilities using renewable energy sources. In order
work, and energy-efficient reinforcement learning are some to enhance decision-making and overall sustainability in
of the topics that have been explored by various researchers. public services, green e-government can also incorporate
Moreover, MAGIC, ADULT, COMPAS, and various other the use of data analysis and modeling. Implementing green
benchmark datasets have been utilized for the evaluation of e-government services will reduce the negative effects of gov-
the study’s mechanism. ernment activities on the environment and boost sustainable
development.
J. GREEN E-COMMERCE AND GREEN E-GOVERNMENT Table 11, presents the studies conducted on green
Green e-commerce is the practice of using environmen- e-commerce and green e-government in recent years.
tally responsible and sustainable methods in the world of To achieve sustainable e-commerce, various studies have
electronic commerce. Green e-commerce includes institut- been conducted on issues such as green supply chain and
ing recycling and waste reduction initiatives, employing risk management, feasibility assessment, assessing market
eco-friendly packaging materials, and running e-commerce readiness, supply chain coordination, and decision-making.
operations on renewable energy sources. Green e-commerce Due to the relatively new area concept, green e-commerce
also includes measures such as building energy-efficient studies mainly focus on performing feasibility analysis and
structures and establishing green transportation networks exploring public response and willingness to adopt green
to reduce its carbon footprint. Green e-commerce benefits e-commerce ecosystems. However, very few studies have
businesses by lowering energy and trash disposal expenses, been conducted on green e-government which mainly focuses
in addition to promoting environmental protection. A favor- on green government procurement practices across various
able brand image and increased customer loyalty may public enterprises and projects.

87470 VOLUME 11, 2023

S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

TABLE 9. Summary of green data mining and information retrieval studies.

K. GREEN EDUCATION tackling the current environmental issues on a worldwide

The term ‘‘green education’’ describes the incorporation of scale. Sustainable development requires a populace that is
environmental and sustainability concepts into the curricu- more ecologically aware and active, which may be accom-
lum and daily operations of educational institutions. It aims plished through green education. Additionally, it may aid
to encourage sustainable practices in the management of in the growth of the next generation of sustainability lead-
educational institutions and to educate students about envi- ers. Many educational institutions have begun to incorporate
ronmental challenges and their role in building a sustainable green education into their programs and daily operations
future. Environmental education may be included in current in recent years. To guarantee that every student has access
subjects, among other approaches to green education. It may to environmental education, additional work must be done
also include implementing sustainable principles into how to properly integrate green education into the educational
educational institutions run, such as lowering energy and system.
water use, encouraging recycling and waste minimization, Table 12, presents the current studies that have been con-
and using green building design. There are several advantages ducted in recent years in the area of green education. The
to green education. It may aid in the growth of students’ studies have been mainly focused on analyzing various issues
critical thinking and problem-solving abilities as well as their in green education using data collected from students, and
comprehension of the interconnection of social, economic, academics. Various studies have been conducted on analyzing
and environmental concerns. Additionally, it encourages chil- students’ knowledge, awareness, and practices toward green
dren to take action to preserve the environment by fostering a education and green computing. Some studies have found
feeling of environmental responsibility. Additionally, imple- that there is no interconnection between green computing
menting sustainable practices in educational institutions may knowledge and gender. However, female students have more
result in financial savings, better indoor air quality, and a positive attitudes toward green computing than male stu-
lower environmental impact. Green education is essential for dents. To promote and spread green education and e-learning,

VOLUME 11, 2023 87471

 S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

TABLE 10. Summary of green machine learning studies.

various energy-efficient green computing architectures, sys- growing crops in stacked layers are very cost and resource
tems, and infrastructures have been developed over the years. efficient for preserving land and water. Green agriculture’s
Furthermore, to implement sustainable education, various overall objective is to produce food in a manner that is
green university models have been developed and case studies ecologically sustainable and less harmful to the environment
have been analyzed. than conventional agricultural methods. Many studies have
been conducted in the area of environmentally friendly green
agriculture due to its direct impact on human lifestyle and the
L. GREEN AGRICULTURE economy.
The goal of green agricultural technology is to minimize Table 13 represents the studies conducted in the area of
the harmful effects of agricultural operations on the envi- green agriculture in the recent year. Policy development
ronment. This is accomplished by implementing measures and impact analysis, energy management and control sys-
for environmentally friendly agricultural production, such as tems, pricing determination mechanisms, farmers’ credit risk
employing renewable energy sources, energy-efficient smart assessment, and renewable energy source-based and IoT-
agricultural devices, utilizing resources more efficiently, inte- based efficient device development are some of the key
grating modern technologies into agricultural practices, dras- subjects that have been examined by various researchers
tically reducing the use of chemical fertilizers and pesticides, in recent years. Farmers’ and experts’ opinions, financial
and boosting the use of organic farming methods. Precision issues, sensor data, and government data have been pri-
agriculture, which makes use of tools like GPS and remote marily used to conduct the studies. However, various other
sensing to increase agricultural yields, can lessen the effect on issues are unexplored in the area of sustainable and green
the environment. Vertical farming mechanisms that involve agriculture.

87472 VOLUME 11, 2023

S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

TABLE 11. Summary of green e-commerce and green e-government studies.

VI. FINDING AND ANALYSIS less power and result in reducing energy consumption [45],
To understand the review studies’ full context, it is necessary [53], [57], [59], [63], [65] and decreasing the carbon foot-
to analyze and examine the studies’ various perspectives. print. The renewable energy sources utilization, such as wind
In the section proposed state-of-the-art research questions and solar power, to power data centers, is an important
reposed have been discussed after performing in-depth anal- advancement. Renewable energy sources usages not only
ysis. Furthermore, green computing area’s challenges and reduce dependence on fossil fuels but also decreases data
future research directions have been presented. center carbon footprint. Another key advancement is the use
of air and water-based cooling systems instead of traditional
A. QUESTION RESPONSE air conditioning systems, which use less energy and are
• RQ 1. What are the most recent advancements in the area more environmentally friendly [37]. Furthermore, the use of
of green computing research, and how do they stack up renewable resource-co-located distributed data centers that
against earlier efforts? are operated by coordinating among themselves can reduce
Green computing is a rapidly growing research area that energy consumption, reduce the need for heavy electricity
has made rapid progress in recent years and attracted lots of grid infrastructure, and reduce security concerns. In compari-
attention due to increasing concerns over the carbon footprint son to earlier efforts, these recent advancements in green data
and energy consumption of modern technologies. In the green center technologies have been much more impactful. Energy-
computing domain, many advancements have been made in efficient hardware and renewable energy sources have dras-
various areas. tically decreased data center carbon footprints. Furthermore,
In the area of GDC technologies, one of the most the implementation of air and water-based cooling systems
recent advancements is the use of energy-efficient hard- has increased energy efficiency and reduced data center
ware, such as processors and memory devices, that consume operational costs.

VOLUME 11, 2023 87473

 S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

TABLE 12. Summary of green education studies.

In the area of GCC, the most recent advancements include technologies has also become more widespread, allowing for
energy-efficient green cloud data centers implementation, the consolidation [81], [88], [105] of multiple workloads onto
virtual machine migration and management mechanisms, a single physical server, reducing energy consumption. Com-
various types of task schedulers, and the development of pared to earlier efforts, these advancements have significantly
more energy-efficient algorithms. The use of virtualization improved the sustainability of cloud computing.

87474 VOLUME 11, 2023

S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

TABLE 13. Summary of green agriculture studies.

In the area of green IoT, recent advancements have microcontrollers. Additionally, initiatives have been under-
focused on improving the energy efficiency of IoT devices taken to recycle and reuse IoT devices at the end of
and reducing the carbon footprint in production and usage. their life cycle. Compared to earlier efforts, these advance-
This includes the development of low-power communica- ments have significantly improved the sustainability of
tion protocols, algorithms, control and management mech- IoT. The development of low-power communication proto-
anisms, and energy-efficient hardware, such as sensors and cols, energy-efficient hardware, and algorithms has allowed

VOLUME 11, 2023 87475

 S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

IoT devices to operate with reduced energy consump- Furthermore, the development of green transportation poli-
tion [113], [114], [118], [128]. Furthermore, recycling and cies, ecological highway development, and production and
reusing IoT devices have also helped to reduce waste and the green transportation coordination plays a key role in achiev-
negative environmental impact. ing green transportation and has significantly improved the
Green procurement is the process of purchasing ser- efficiency and sustainability of the transportation sector,
vices and products that have a minimal environmental compared to earlier efforts.
impact. There have been considerable developments in green Recent advancements in the area of green database
procurement-related research in recent years. Companies are research have focused on reducing the energy con-
now considering the environmental impact of their purchases sumption of databases while maintaining service perfor-
and looking for more sustainable options. Furthermore, mance [201], [202]. The use of energy-efficient hardware
recent studies are analyzing the drivers and barriers of green components, such as solid-state drives (SSDs), which con-
procurement across various industries and projects, which sume less power compared to traditional hard disk drives
can be beneficial for understanding the current obstacles (HDDs), can provide services more quickly. Additionally,
that companies and organizations are facing towards imple- energy-efficient algorithms have been developed to reduce
menting green procurement. These findings have led to the the energy consumption of database operations, such as
development of new strategies, technologies, mechanisms, indexing, sorting, and querying. Furthermore, research has
and products that are environmentally friendly and more also been conducted on developing various green database
energy efficient. The use of these technologies helps reduce models for the stock market and optimizing the use of
the carbon footprint of companies, boost green procurement databases in the cloud, where resources can be shared and
implementation, and contribute to the overall sustainability of dynamically provisioned to reduce overall energy consump-
the planet. tion. These efforts have helped to improve the sustainability
In recent years, green education has seen significant of database systems and reduce their environmental impact.
advancements, particularly in the areas of curriculum devel- In the area of green data mining and information retrieval,
opment, e-learning system development, awareness analysis, recent advancements have focused on reducing the car-
and training programs. The focus has been on incorporating bon footprint of data centers and optimizing energy usage
environmentally sustainable practices into the design, devel- using energy-efficient data mining techniques. For instance,
opment, and use of technology [236]. The goal of green researchers have been developing algorithms that can min-
education is to educate individuals and organizations on the imize energy consumption during data processing and
importance of reducing the environmental impact of technol- storage [209]. Additionally, researchers have been devel-
ogy and how to implement sustainable practices [238]. Com- oping methods to reduce data redundancy and use more
pared to earlier efforts, these advancements have significantly energy-efficient information retrieval techniques to minimize
improved the availability of green computing education and the energy footprint.
increased the number of individuals and organizations that are In recent years, the area of green machine learning has
able to adopt environmentally sustainable green education. seen significant advancements in reducing the energy con-
Recent advancements in smart grid technology have sumption and carbon footprint of machine learning tech-
focused on improving the reliability, efficiency, and sustain- niques [218], [223]. One of the most notable efforts is the
ability of the electricity grid. This has been achieved through development of energy-efficient algorithms that require fewer
the use of advanced sensors, energy and resource manage- computational resources and datasets to train models, thus
ment, distribution control systems, communication network reducing energy consumption. Furthermore, various efficient
optimization, and data analytics to better monitor and control ML techniques have been deployed to analyze and solve
the flow of electricity. Smart grid technology also enables various issues. The development of knowledge-aware ML
renewable energy sources integration, such as wind and solar and the integration of ML in smart IoT devices can reduce
power, and energy control and management mechanisms into resource consumption and thus aid in achieving sustainable
the grid. Compared to earlier efforts, these advancements development.
have greatly increased the efficiency of the electricity grid, In the area of green e-commerce and e-government, recent
reducing energy waste and carbon emissions. advancements have focused on reducing the carbon footprint
Green transportation systems have seen significant of online transactions, supply chains, and electronic gover-
advancements in recent years, with a focus on reducing the nance processes by analyzing the current issue and devel-
carbon footprint of the transportation sector [198]. One of the oping modern technology-based solutions. The use of more
latest advancements is the increased use of electric vehicles energy-efficient servers, reducing the need for paper-based
(EVs), improvements in charging infrastructure and battery transactions, and implementing green procurement practices
technology, and plug-in hybrid electric vehicles (PHEVs), for electronic devices used in e-commerce and e-government
which have reduced dependency on fossil fuels. Another are most of the environment-friendly measures. These efforts
advancement is the development of smart traffic management have resulted in a reduction of energy consumption and a
systems, which use real-time data and analytics to opti- decrease in the amount of waste generated. Additionally,
mize traffic flow, reduce congestion, and lower emissions. many e-commerce companies are now offering carbon-offset

87476 VOLUME 11, 2023

S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

programs for shipping, and e-government agencies are pro- data centers can reduce their carbon footprint and energy
moting the use of electronic signatures to reduce the need costs [68], [91]. Cloud computing’s successful implementa-
for physical document storage [225]. These advancements tions of the green computing approach involve energy-aware
demonstrate a commitment to sustainability and a recognition algorithms that can reduce energy consumption while main-
of the important role that technology can play in reducing the taining QoS, resource allocation and management systems,
environmental impact of e-commerce and governance. and efficient and low-cost VM migration. IoT includes energy
Green agriculture is a recent area of research in green management systems and design development, identifying
computing that aims to make the agriculture sector more sus- various industries’ green procurement barriers and drivers,
tainable and environmentally friendly. One of the most recent analyzing students’ awareness and attributes towards green
advancements in this area is the use of IoT-based solutions for computing, developing eco-friendly smart green universi-
precision agriculture. This involves using sensors and other ties, developing secure, energy-efficient algorithms, resource
IoT devices to monitor temperature, soil moisture levels, and and distribution management systems for smart grid, ana-
other factors that affect crop growth [253], [255]. This infor- lyzing and developing green transportation policies, smart
mation is then used to optimize irrigation and fertilization efficient automatic traffic management and ecological high-
practices, leading to improved yields and reduced resource way and railway development, energy, time, and space opti-
waste. Furthermore, integrating solar power in agricultural mization for green database model design and development,
devices such as water pumps, energy management systems, a cost-sensitive information retrieval system, efficient and
decision-making mechanisms, and policy development are data-aware data mining services, green hyperparameter opti-
some of the other recent achievements. mization for green ML, an automatic plant monitoring sys-
In conclusion, the field of green computing has seen tem, and solar power based water pump development are
significant advancements in recent years. From green data some of the major successful implementations of green com-
center technologies to green agriculture, researchers and puting. By implementing these approaches, organizations can
practitioners have been working towards creating more sus- reduce their environmental impact, minimize waste, save
tainable and energy-efficient systems and frameworks. These resources, and reduce energy costs.
advancements have resulted in increased energy efficiency, • RQ 3. Which areas of green computing are most under-
reduced carbon emissions, reduced resource requirements, developed, and what are the obstacles researchers in
and increased cost savings. In comparison to earlier efforts, those areas are facing?
the current research in green computing has a greater focus Green computing is an important area of research that seeks
on holistic solutions, taking into consideration the entire to develop sustainable and environmentally-friendly tech-
lifecycle of a system. Overall, green computing is becoming nologies in the field of computing. Despite the significant
a popular and important area of research and development advancements made in recent years, there are still areas that
as the need for sustainable technology solutions has been are underdeveloped and face various challenges. Some of the
increasing rapidly. areas that are lagging behind include green education, green
• RQ 2. What approaches have shown to be the most databases, green data mining and information retrieval, green
successful implementations in the areas of green com- machine learning, green e-commerce and e-government, and
puting? green agriculture.
Green computing is a process of using computers and The obstacles facing researchers in these areas range from
related resources in an environmentally friendly manner. Var- limited funding, a lack of infrastructure, and a lack of tech-
ious approaches such as energy-efficient computing, optimiz- nical knowledge to solve the complexity of the problems.
ing software, energy-efficient cooling systems, VM place- Green education research includes building eco-friendly edu-
ment, etc. are some of the successful implementations in cational infrastructure, energy efficiency e-learning devices,
the green computing area. The energy-efficient computing and applications, integrating green computing concepts in the
approach aims to minimize the computers, devices, soft- curriculum, and utilizing resources efficiently in the insti-
ware, and servers’ energy consumption by using energy- tution. One of the major challenges in conducting research
efficient algorithms, hardware design, and the integration of in this area is the cost and awareness associated with green
modern technologies. Optimizing software to reduce power educational infrastructure development, such as green uni-
consumption by using low-power processors, power-saving versities. Due to an evolving and new area in green com-
modes, and virtualization techniques to minimize the num- puting, a limited number of studies have been conducted
ber of active servers also showcases the efficacy and some in the area of green databases. Green database’s primary
of the successful implementation of green computing. The research goal is to develop efficient database models and
data center optimization approach focuses on optimizing algorithms for storing, searching, and efficiently exchang-
data centers, which are the backbone of many organiza- ing data. However, technical understanding and knowledge,
tions’ computing systems and are some of the successful funding and awareness of companies, and proper infrastruc-
implementations of green cloud computing in green data ture and testing mechanisms are some of the obstacles that
center areas. By implementing energy-efficient cooling sys- researchers face in the area of the green database. Green
tems, virtualizing servers, and reducing power consumption, data mining and information retrieval research centers around

VOLUME 11, 2023 87477

 S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

building energy and other resources efficient data mining of individual devices and optimize their power usage accord-
and information retrieval systems that can efficiently retrieve ingly, reducing energy consumption and extending the life of
data across various sectors. However, proper energy-efficient the device. In this sense, AI can be seen as a tool for reducing
framework unavailability and technical knowledge require- the environmental impact of technology. Furthermore, AI can
ment for optimizing data mining techniques are some of the be used to develop new green technologies, such as smart
major obstacles for researchers. Lack of technical knowledge, grids and green transportation systems. AI algorithms can
funding, and awareness are some of the major obstacles be used to analyze energy consumption patterns and identify
in developing and implementing energy-efficient and opti- areas where energy savings can be made, leading to the
mized green ML across various industries. Green agriculture development of new, more efficient systems. AI can also be
research includes building energy-efficient IoT devices for used to monitor the energy consumption of these systems and
farming, smart controlling and monitoring systems, integrat- optimize their energy usage over time, further reducing their
ing renewable energy in various activities of agriculture, environmental impact.
and developing supply chain change management. However, In recent years, ML significance in green computing fields
the lack of practical implementation by the farmer due to has grown. ML algorithms has been employed to opti-
various constraints such as cost, time, and technical knowl- mize the energy consumption of computer systems, enabling
edge are the biggest obstacles to motivating researchers to the reduction of energy waste and the creation of more
perform green agriculture-related research. Government and energy-efficient solutions [210]. One of the main ways
e-commerce companies’ lack of awareness, willingness, and machine learning has affected energy savings in computer
funding are some of the major obstacles to building a green systems is through the development of smarter energy man-
e-commerce ecosystem and promoting green computing in agement techniques. For instance, machine learning algo-
government organizations and activities. To overcome these rithms can be used to predict the energy consumption of
challenges, researchers need to collaborate with industry part- different parts of a computer system, which can then be used
ners, governments, and other stakeholders to create solutions to dynamically adjust the power consumption of these parts
that are not only sustainable but also economically feasible. to minimize energy waste [216]. Additionally, to optimize
Furthermore, the lack of standardization in the fields and the allocation of resources, such as computing power and
the lack of a unified approach to green computing research memory, to save resource consumption and the operation cost
also pose a challenge. Nevertheless, by addressing these chal- of computer systems ML algorithms can be employed. This
lenges, the underdeveloped areas of green computing have the ML model utilization is important in data centers, where large
potential to drive innovation and lead to a more sustainable amounts of energy are consumed by the computing infras-
future. tructure. By using machine learning algorithms to optimize
• RQ 4. How do recent developments in artificial intelli- resource allocation, data centers can reduce energy consump-
gence and machine learning affect the amount of energy tion and lower their carbon footprint [218]. Furthermore,
saved by computer systems? ML algorithms or models can be utilized to predict failures
Artificial intelligence (AI) and ML are rapidly evolving tech- in computer systems, allowing for proactive maintenance and
nologies that have the capability to greatly minimize the reducing these systems energy consumption over the long
amount of energy consumed by computer systems. Recent term.
advancements in AI and ML, such as deep learning algo- Recent developments in AI and ML have a notable impact
rithms and edge computing, are revolutionizing the way that on the amount of energy saved by computer systems. These
computers process and analyze data. These developments technologies have been integrated into various systems to
have the capability and potential to minimize the energy reduce energy consumption by automating routine tasks and
consumption of computer systems as well as improve their distributing and managing resources properly. AI-based sys-
overall performance and efficiency. By leveraging AI and tems can dynamically manage the power consumption of
ML techniques, computer systems can make more intelligent computer systems by adjusting the power usage based on the
decisions about how to allocate and manage their resources, workload and system requirements. Additionally, ML meth-
reducing the energy consumption required to perform a given ods could be used to optimize resource utilization of com-
task. puter systems, further reducing their energy consumption.
AI algorithms such as ML can be employed to optimize As a result, these advancements have the potential to play a
data centers’ and other computing systems’ energy usage. For crucial role in reducing the carbon footprint of the technology
instance, AI can be used to predict the workload demands industry and ensuring a more sustainable future.
of a data center and adjust its energy usage accordingly, • RQ 5. What areas of green computing research are the
reducing energy waste. AI could also be employed to monitor most promising?
individual devices’ power consumption and shut down those Green computing research has made significant progress in
that are not in use in a data center, further reducing energy recent years, and there are several areas that show significant
consumption. Additionally, AI algorithms can be employed promise with respect to energy efficiency and sustainabil-
to enhance the energy efficiency of computing devices [65]. ity. Some of the most promising green computing areas
AI can be utilized to analyze the energy consumption patterns include the IoT, data centers, cloud computing, and smart

87478 VOLUME 11, 2023

S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

grid systems. These areas have the potential to reduce the cost, and resource usage, which can affect sustainability and
energy consumption of computer systems, servers, devices, environment-friendly green computing development. Using
and software, which can reduce greenhouse gas emissions natural resource-based technology to power and cool data
and the overall carbon footprint. Additionally, these areas centers can reduce operational costs and reduce CO2 emis-
have the potential to provide new, more sustainable solutions sions. Distributed data center development and management
for managing and analyzing large amounts of data, as well can reduce the electricity grid infrastructure requirements,
as automating processes and reducing energy consumption in which can be beneficial for setting up data centers in devel-
the energy sector. oping countries.
IoT has become an increasingly important aspect of the In conclusion, green computing research continues to
technology industry, and its growth has been exponential expand as the need for more sustainable and eco-friendly
over the past few years. Green IoT is the concept of mak- technology becomes increasingly important. From areas such
ing IoT devices more energy-efficient and environmentally as green IoT, green data centers, green cloud computing, and
sustainable, and this area of green computing research has smart grids, the potential for breakthroughs in reducing the
shown great promise [118], [122], [123]. Energy-efficient IoT carbon footprint of technology is substantial. With continued
devices have a major effect on the technology industry’s total innovation and collaboration between academia and industry,
energy consumption reduction [114], [115]. Additionally, the the future of green computing looks bright, and the potential
increasing demand for smart homes and infrastructure is driv- to make a significant impact on the environment is real. It’s
ing the development of more energy-efficient IoT devices, up to all individuals to support and drive these efforts toward
further increasing the potential for energy savings [111]. a more sustainable future.
As the usage of IoT devices continues to grow, the need for • RQ 6. What will be the most difficult obstacles for those
more sustainable and energy-efficient technology becomes researchers who are now intending to do research on
increasingly important. One of the most promising areas of green computing?
green computing research is the green IoT, as it has the Conducting research in the field of green computing can be
potential to make a significant impact on reducing the energy challenging, as there are several obstacles that researchers
consumption of technology while also reducing its environ- may face. These obstacles can include a lack of standardiza-
mental impact. tion across the field, technical barriers, and limited funding.
Smart grids are considered one of the most promising areas Standardization across the field of green computing
in green computing due to smart grids’ ability to minimize research is one of the most challenging obstacles for
energy efficiency and reduce carbon emissions in the energy researchers [50], [119], [217]. There are a multitude of
sector [161], [163]. The smart grid technologies implemen- devices, technologies, and protocols used in the field of green
tation involves the integration of digital technologies with computing, making it difficult to ensure compatibility and
the traditional power grid system, allowing for real-time interoperability. Moreover, there is a lack of uniformity in
monitoring and control of energy generation, distribution, the methods used to evaluate the effectiveness and efficiency
and consumption. This leads to increased energy efficiency, of various green computing technologies and solutions. This
as power grids can dynamically respond to changes in energy makes it difficult for researchers to compare their results
demand, reducing the need for additional energy genera- and draw meaningful conclusions from their studies. In addi-
tion. Additionally, renewable energy sources usages such as tion, there is a lack of consensus on the best practices for
solar and wind power can be optimized through the use of measuring and reporting the energy consumption of com-
smart grid technologies, reducing dependence on fossil fuels puting systems and DC, making it difficult for researchers
and decreasing overall carbon emissions. Furthermore, the to accurately compare the energy savings of different solu-
deployment of smart grid technologies also enables the inte- tions. Furthermore, the rapidly evolving nature of technology
gration of electric vehicles and other consumer-facing tech- makes it difficult to standardize the technologies and devel-
nologies, contributing to a more sustainable energy future. opment process. With new technologies being developed and
With these advantages, it’s not surprising that smart grid deployed at an increasingly rapid pace, it can be challeng-
technology is considered one of the most promising areas in ing for researchers to keep up and ensure that their work
green computing research. remains relevant and up-to-date. Additionally, there are often
Due to the adaptation of modern technology, the use of trade-offs between energy efficiency and other performance
CC is increasing rapidly. Therefore, green cloud computing metrics, such as processing power and data throughput, mak-
is a promising research field that can improve the sustainable ing it challenging to determine the optimal balance between
development of modern CC-based technologies and services. energy efficiency and performance.
The usage of energy-efficient algorithms, task and job sched- One of the main technical obstacles faced by researchers in
ulers, and VM migration can maintain QoS and reduce energy green computing is the complexity of designing and imple-
in the CC process, thus promoting the sustainability of cloud menting energy-efficient systems and algorithms [83], [93].
computing. The complexity arises because energy consumption is a
The energy control and management system of the data key factor in the design and performance of computer sys-
center has a direct impact on energy usage, operational tems, and reducing energy consumption can often result in

VOLUME 11, 2023 87479

 S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

trade-offs with other performance metrics, such as speed, Despite these obstacles, the importance of green computing
accuracy, and reliability. Reducing the power consumption research is crucial for the future of technology and the envi-
of a computer system may result in slower processing times ronment. By overcoming these challenges, researchers can
or reduced accuracy, which can negatively impact the overall make significant contributions toward the creation of sustain-
effectiveness and usefulness of the system. Additionally, the able and energy-efficient technologies and infrastructures.
wide range of different devices, platforms, and operating • RQ 7. What role does green computing research and
systems that are used in computer systems can make it dif- development play in achieving the aims of sustainability
ficult to design energy-efficient systems that are compatible across the world?
with a broad range of different devices and configurations. Green computing research and development serves a key
Those issues make it difficult for researchers to develop and role in achieving the aims of sustainability across the
implement energy-efficient algorithms and systems that are world [66], [181], [245],. Due to the rapid technologi-
widely usable and scalable across different platforms. Fur- cal development and the increasing demand for computing
thermore, the high pace of technological advancement and resources, the energy consumption of the ICT sector has also
the introduction of new devices and platforms can quickly risen, leading to significant carbon emissions and environ-
render existing energy-efficient systems and algorithms obso- mental degradation [59], [143]. In this context, green com-
lete, requiring ongoing research and development efforts to puting provides a means to reduce the environmental impact
stay ahead of the curve. The limited availability of data and of the ICT sector while still meeting the growing demand
resources can also be a major obstacle for researchers in green for computing resources. This area of research and develop-
computing [41], [78], [122]. Collecting and analyzing large ment focus on designing and implementing energy-efficient
amounts of data on energy consumption and performance can technologies and practices that minimize the environmental
be time-consuming and resource-intensive, which can limit footprint of computing systems while maximizing their per-
the ability of researchers to fully explore and understand the formance and functionality [55], [57]. The ultimate goal of
impact of different energy-saving strategies. green computing is to create a more sustainable computing
Limited funding is a significant obstacle facing researchers infrastructure that can support the needs of both current and
in the field of green computing. Government agencies and future generations while preserving the environment and nat-
private sector organizations are increasingly interested in ural resources.
investing in research and development projects aimed at The role of research and development in green computing
reducing the environmental impact of technology. However, is critical to reaching the objectives of global sustainability.
the amount of funding available for these projects is often Sustainability refers to the ability of an ecosystem, society,
limited, which can make it difficult for researchers to pursue or economy to maintain its function and structure over time.
their work in an impactful manner and scale the developed In the context of computing, sustainability involves reducing
solution across multiple domains. The competition for fund- the environmental impact of computing systems while ensur-
ing can be intense, and the process of securing funding is ing their long-term viability and reliability. This requires a
often time-consuming and complicated. In addition to the focus on energy efficiency, material use, and waste reduction,
competition for funding, many researchers in the field of as well as on creating a sustainable computing infrastructure
green computing also face a lack of financial stability and that supports the development and use of environmentally
job security. Research is often carried out in academic insti- friendly technologies.
tutions or research labs, which typically do not offer the same One key area where green computing research and devel-
level of financial stability as industry positions. Researchers opment is having a significant impact is in the optimiza-
must be prepared to take on a significant amount of risk tion of energy use in computing systems [65], [77]. This
and uncertainty in order to pursue their work. This lack of involves developing hardware and software technologies that
financial stability can also make it difficult for researchers reduce the energy consumption of computing systems, both
to retain top talent, as they may be attracted to more sta- in terms of the amount of energy used by the systems
ble and secure positions in the industry. Moreover, limited themselves and the energy required to produce and dispose
funding can also restrict the ability of researchers to access of the systems [51], [66]. Researchers are developing new
the latest technology and resources needed to conduct their types of low-power processors and memory technologies,
work [139], [197], [254]. New technologies are emerging as well as more efficient power management techniques,
all the time, and researchers must stay up-to-date with these that can minimize the energy consumption of computing
developments in order to make progress in the field. How- systems [92], [113]. In addition, advances in virtualization
ever, the cost of new technologies and equipment can be and cloud computing are helping to reduce energy consump-
prohibitively high, and without adequate funding, researchers tion by enabling the sharing of computing resources across
may struggle to access the resources they need to advance multiple applications and users.
their work. Another key area where green computing research and
In closing, the researcher in the field of green computing is development is having an impact is in the reduction of e-waste
facing multiple challenges, including limited funding, a lack through the development of new technologies and e-waste
of standardization across the field, and technical obstacles. management strategies [241]. E-waste refers to the waste

87480 VOLUME 11, 2023

S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

generated by electronic devices, such as computers, smart- goals. One challenge is the limited power and processing
phones, and televisions, that have reached the end of their capabilities of small and low-cost IoT devices, making it
usable life. This waste can contain toxic materials, such as hard to integrate energy-saving features. Another challenge
mercury, lead, and cadmium, that can harm human health and is the absence of a standardized approach to green IoT, which
the environment if not properly disposed. Green computing can make it challenging for organizations to choose between
research and development is helping to reduce e-waste by different solutions. Additionally, the decentralized nature of
developing technologies that enable the reuse and recycling IoT networks can make it difficult to effectively monitor
of electronic components, as well as by promoting the design and manage energy usage. The complex interconnections
of devices that are more environmentally friendly and have a between IoT devices can also pose difficulties in identify-
longer lifespan. ing energy inefficiencies and implementing solutions. The
Therefore, it can be concluded that green computing fast-paced technological advancements in IoT also present a
research and development plays a critical role in achieving the challenge in keeping up with the latest green IoT technologies
aims of sustainability by reducing the environmental impact and best practices.
of computing systems and ensuring their long-term viability The challenges of implementing green procurement in var-
and reliability. Through the development of energy-efficient ious industries include the limited availability of eco-friendly
technologies, the reduction of e-waste, and the promotion of products, higher costs associated with these products, a lack
sustainable computing practices, green computing research of standardization and certification for green products, and
and development is helping to create a more environmentally resistance to changing traditional procurement practices from
friendly and sustainable computing industry that can support both suppliers and customers. Additionally, there is a lack of
the needs of future generations. awareness and education about the benefits of green procure-
ment, which makes it difficult for organizations to convince
B. CHALLENGES stakeholders to adopt this practice. Finally, there is a lack
In the process of reviewing studies, various challenges related of more robust green supply chain management systems that
to green computing have been observed and identified. ensure that the products are not only environmentally friendly
By overcoming these challenges, green computing can help but also ethically sourced. These challenges present a signifi-
reduce energy consumption, minimize waste, and mitigate the cant barrier to the widespread adoption of green procurement
environmental impact of the technology industry. across various industries.
The main challenges in the area of GDC include the One major challenge of green education is its lack of
high levels of energy consumption required to run and cool integration into traditional educational systems and curricula.
large amounts of equipment in a data center, as well as This leads to fragmented and inconsistent delivery of green
the heat generated by the equipment itself. Another chal- education, with some schools and universities offering com-
lenge is the limited availability of renewable energy sources, prehensive programs while others offer little or none. Another
which makes it difficult for data centers to operate in a challenge is the limited resources and funding available for
completely carbon-neutral manner. The rapid pace of tech- green education initiatives, which hampers the ability to
nological advancements in the data center industry presents effectively educate and engage students and the wider public.
challenges in managing equipment obsolescence and the Furthermore, various resistance to change or adaptation of
resulting electronic waste. The continuous improvement and green education procedures from some stakeholders, who
upgrades in data center equipment led to shorter lifespans may be skeptical about the value of green education or the
and increased disposal requirements, contributing to environ- need for a more sustainable future. Lastly, better collaboration
mental impacts. Finding sustainable solutions to minimize and coordination among educators, politicians, and industry
electronic waste and reduce the environmental effects of data leaders is necessary to ensure that green education is inte-
centers becomes a significant challenge. Implementing effec- grated into all sectors of society.
tive strategies for equipment reuse, recycling, and responsible The smart grid is a modernized electrical grid that inte-
disposal is crucial to addressing this challenge. grates digital technology to improve efficiency, reliability,
Green cloud computing challenges include standardization and sustainability. However, there are several challenges
and regulation to ensure that cloud computing is conducted associated with implementing a smart grid. One challenge is
in a responsible and sustainable manner. Evaluating the the high cost of upgrading existing infrastructure and deploy-
developed energy-efficient algorithm, task scheduling, and ing new technologies, which makes it difficult for some utility
resource management and control system in a standardized companies and governments to invest in the green transi-
way is another challenge that needs to be resolved. Integra- tion. Another challenge is the lack of standardization, which
tion of developed energy-efficient mechanisms across various makes it difficult to ensure interoperability and compatibility
devices and platforms is another challenge that needs to be between different smart grid components and systems. There
addressed to obtain efficient green cloud computing. is also a challenge to address security and privacy concerns
Green IoT seeks to reduce the environmental impact and associated with the increased use of digital technology in the
increase the energy efficiency of IoT devices and systems. grid, as well as to ensure the reliable operation of the grid
However, there are several challenges to achieving these in the face of potential cyber threats. Additionally, there is a

VOLUME 11, 2023 87481

 S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

challenge for greater consumer engagement and education to innovation in energy-efficient data mining and information
encourage the adoption of smart grid technologies and prac- retrieval methods and promote sustainability in this field.
tices. Finally, there is also a challenge for regulatory support The biggest challenge in the area of green machine learn-
and collaboration between government agencies, utilities, and ing is the massive amounts of energy and computational
other stakeholders to ensure that the transition to a smart grid resources required to train and run large machine learning
is coordinated and effective. models, which may result in a significant carbon footprint
There are several challenges associated with transitioning and contribute to climate change. Additionally, the lack of
to green transportation systems. One challenge is the high transparency in the supply chain of hardware used for train-
cost of developing and deploying alternative fuel vehicles and ing and inference can also result in unethical practices such
infrastructure and the lack of publicly available, usable green as e-waste and the exploitation of resources. Moreover, the
transportation infrastructure, which can be common barriers shortage of efficient hardware and scalable infrastructure for
to the widespread adoption of green transportation systems. running such models in a sustainable manner is a hindrance
Another challenge is the lack of public awareness and educa- to the widespread adoption of green machine learning.
tion about the benefits of green transportation, which makes Green e-commerce refers to the practice of using digital
it difficult to encourage the widespread adoption of these platforms for commerce in an environmentally sustainable
systems. The lack of widespread coordination and collabora- manner. The main challenge in this area is the large amounts
tion between government agencies, transportation providers, of energy and resources consumed by data centers and
and other stakeholders poses a significant challenge in other infrastructure required for e-commerce operations. This
integrating green transportation systems into existing infras- includes the production and transportation of goods, as well
tructure. Ensuring seamless integration requires aligning as the use of electronic devices for browsing and purchas-
policies, regulations, and incentives across multiple entities. ing products. Additionally, the challenge of electronic waste
Limited funding and resources for infrastructure upgrades (e-waste) is also a significant concern, as many electronic
and the adoption of greener transportation technologies devices used for e-commerce become obsolete rapidly due
present additional challenges. Overcoming resistance to to improvements in technology and are often not disposed
change and addressing potential conflicts of interest among of properly. Green e-government refers to the use of digi-
stakeholders are key obstacles to achieving widespread adop- tal technology in government operations and services in an
tion of green transportation solutions. environmentally sustainable manner. The main challenge in
Despite the increasing demand for green databases, there this area is the lack of infrastructure and resources in many
are several challenges associated with their implementa- countries to support digital government services. This can
tion and use. One challenge is the need for more efficient result in an increase in the use of paper-based processes,
and effective methods for managing and processing large which have a significant environmental impact. Additionally,
amounts of data, which is becoming increasingly complex many government websites and digital platforms are often
and diverse. Another challenge is the high cost of acquiring outdated and not user-friendly, which can lead to a lack of
and deploying green database technologies, which can be a adoption and utilization of digital government services.
barrier for some organizations. Additionally, there is a lack Green agriculture refers to farming practices that prioritize
of technical knowledge and awareness about the importance environmental sustainability. The main challenges in this area
of green databases and their role in promoting sustainability. include soil degradation and water scarcity due to the adapta-
Finally, there is a challenge for collaboration and standard- tion of traditional, inefficient, and not eco-friendly practices
ization across the industry to ensure that green databases rather than green agriculture practices. Furthermore, the lack
are compatible and interoperable with other systems and of technical knowledge of farmers, financial issues, and lack
technologies. of awareness are some of the major challenges to implement-
The high computational and energy costs associated with ing green agriculture practices. The excessive use of chemical
processing and analyzing vast volumes of data pose signifi- fertilizers and pesticides and the lack of awareness of the
cant challenges in the fields of green data mining and infor- adoption of organic and green farming are other critical chal-
mation retrieval. Finding efficient algorithms and techniques lenges in the process of green agriculture adaptation. Other
that can minimize resource consumption while maintaining challenges include the unavailability of cost-effective green
accurate results is crucial. Moreover, the lack of standardized agriculture technologies, green supply chain management,
methods and tools for evaluating and comparing the envi- and a lack of government subsidies for the adaptation of
ronmental impact of different data mining and information green agricultural practices. Climate change is also a major
retrieval approaches complicates the decision-making pro- challenge for green agriculture, as it can lead to unpredictable
cess for organizations striving for sustainability. Developing weather patterns, soil erosion, and crop failures.
comprehensive frameworks and metrics to assess the energy Finally, it can be summarized that green computing has
efficiency and environmental footprint of these techniques the potential to revolutionize various industries and create a
is essential for making informed choices. Addressing these more sustainable future. However, it is crucial to address the
challenges requires interdisciplinary collaboration between challenges associated with its implementation and encour-
researchers, industry experts, and policymakers to drive age widespread adoption. This can be achieved through

87482 VOLUME 11, 2023

S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

investments in research and development, standardization, VM migration techniques to improve resource utilization
education, and awareness-raising initiatives. By working and reduce energy consumption. Furthermore, the integration
together, industry leaders, policymakers, and individuals can of various energy-efficient algorithms, task schedulers, and
help bring about a greener future that balances economic VM migration techniques in smart, efficient green cloud
growth with environmental protection. computing architectures is one of the major future research
directions. Another important aspect of future research will
C. FUTURE RESEARCH DIRECTIONS be the development of cloud computing platforms that are
Green computing, which refers to the development of meth- environmentally friendly throughout their entire lifecycle,
ods that are environmentally sustainable for computing and from design and manufacturing to decommissioning and dis-
information technology, is a rapidly growing field of research. posal. Proper planning and implementation of end-of-life
Considering the growing worldwide demand for computing strategies can ensure the safe and responsible disposal of
resources and the increasing environmental impact of tra- electronic waste (e-waste) generated by these systems. This
ditional computing practices, there is a growing need for involves adopting sustainable practices for recycling and dis-
research in green computing. posing of hardware components, ensuring compliance with
Green data centers are gaining significance in the current environmental regulations, and promoting circular economy
scenario. Future research directions in this field will likely principles by promoting reuse and refurbishment.
focus on the development of more efficient and cost-effective IoT has the potential to significantly impact our daily lives
cooling systems, increased utilization of renewable energy and the environment. Future green IoT research may concen-
sources, improved power distribution systems, and the reduc- trate on the development of multiple devices and platform-
tion of hardware waste through extended product life cycles integrated sustainable, and energy-efficient IoT solutions.
and recycling programs. In addition to the aforementioned One area of research may be the development of low-power
aspects, there are several other areas of research that hold consumption devices and protocols that can operate for long
great potential in advancing the field of green computing. One periods of time on minimal energy. An additional area of
such area involves the development of intelligent algorithms focus in future research lies in the creation of IoT networks
that can optimize energy usage in various computing sys- that leverage renewable energy sources and strive to minimize
tems and infrastructures. These algorithms may dynamically energy waste. By integrating sustainable power generation
adjust resource allocation, workload scheduling, and power technologies, such as solar or wind energy, into IoT infras-
management to maximize energy efficiency and minimize tructure, researchers can significantly reduce reliance on fos-
wastage. Another promising avenue of exploration is the vir- sil fuels and contribute to a cleaner and more environmentally
tualization of data center resources. By virtualizing servers, friendly network.
storage, and networking infrastructure, organizations can Green procurement, or the practice of purchasing products
achieve higher resource utilization rates, reduce energy costs, and services that are environmentally sustainable, is gaining
and improve overall operational efficiency. This approach importance as organizations seek to reduce their carbon foot-
allows for the consolidation of workloads onto fewer phys- print and minimize their impact on the environment. In this
ical servers, leading to energy savings and a smaller carbon context, future research in green procurement will focus on
footprint. Furthermore, the creation of modular and scalable developing methodologies for evaluating the environmental
data center designs offers opportunities for more flexible impact of products and services, as well as tools and systems
and efficient deployment and maintenance. Modular designs to support sustainable purchasing decisions. Analyzing and
allow for the addition or removal of components based identifying green procurement practices, drivers, and barriers
on demand, enabling organizations to scale their data cen- in various industries is one of the major future research
ters in a cost-effective manner. Additionally, incorporating directions. Government green procurement adaptation at var-
energy-efficient cooling systems, power distribution mecha- ious levels and organizations, identifying green procurement
nisms, and renewable energy sources can further enhance the relationships with various factors, supplier collaboration, and
sustainability of data centers. green information systems are some of the future research
The future of green cloud computing is characterized opportunities. Green procurement presents another com-
by a strong focus on sustainability, reducing carbon emis- pelling avenue of research, focusing on the exploration of
sions and energy consumption, and optimizing the use of novel and sustainable green products and services crafted
resources. In this context, future research directions are from recycled materials or powered by renewable energy
likely to include the development of efficient algorithms for sources. By delving into this area, researchers can contribute
energy-aware resource allocation and management, the inte- to the development of environmentally friendly procurement
gration of renewable energy sources into cloud data centers, practices that prioritize the adoption of sustainable alterna-
and the investigation of new cooling technologies that reduce tives. Through the identification and evaluation of innovative
the environmental impact of data center operation. Another product designs and service offerings, this research endeavor
area of interest could be the study of innovative hardware seeks to enhance the availability and accessibility of green
and software architectures that can improve the energy effi- options in the market. By promoting the utilization of recy-
ciency of cloud computing systems, as well as the use of cled materials and renewable energy, researchers can foster a

VOLUME 11, 2023 87483

 S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

circular economy and facilitate the transition toward a more reducing the energy consumption of databases, such as opti-
sustainable and resource-efficient future. mizing query processing and data compression.
Green education, or the integration of environmental sus- Green data mining and information retrieval, which refers
tainability into educational curricula, is a growing field that to the development of environmentally sustainable methods
aims to educate and engage individuals in environmental for collecting, storing, and processing data, is becoming
stewardship. In this context, future research in green edu- an increasingly important area of research. In the future,
cation will focus on developing innovative and effective research in this field will focus on developing energy-efficient
methods for teaching environmental sustainability, as well as algorithms and systems for data mining and information
creating educational materials that are engaging and accessi- retrieval. This could include the exploration of alternative
ble to a wide range of learners. Another key focus area is the energy sources for powering data centers as well as the
investigation of virtual and augmented reality tools and their optimization of existing algorithms to reduce energy con-
potential to enhance educational experiences in the context of sumption.
sustainability. By integrating these immersive technologies, Green machine learning, which refers to the develop-
researchers can delve into the development of innovative ment of environmentally sustainable methods for training and
teaching methods that promote environmental awareness and deploying machine learning models, is a rapidly growing area
eco-conscious behavior. This research endeavor aims to eval- of research. In the future, research in green machine learn-
uate the effectiveness of virtual and augmented reality in ing will focus on developing energy-efficient algorithms and
delivering impactful and engaging green education, fostering systems for training and deploying machine learning mod-
a deeper understanding and appreciation of environmental els. This could include the exploration of alternative energy
issues among learners. By embracing technology as a tool sources for powering data centers as well as the optimization
for sustainable education, researchers can contribute to the of existing algorithms to reduce energy consumption.
advancement of eco-literacy and empower individuals to Green e-commerce and green e-government, which refer
make informed and sustainable choices in their lives. to the development of environmentally sustainable methods
The smart grid, which refers to the integration of digital for conducting business and governance online, are grow-
technology into the electrical power grid to improve effi- ing areas of research. In the future, research in the field
ciency, reliability, and sustainability, is an area of ongoing of green e-commerce may prioritize the development of
research and development. In the future, research in the field energy-efficient systems to enhance sustainability in online
of smart grids will focus on improving the overall resilience commercial activities. This includes exploring alternative
and security of the grid, as well as increasing the integra- energy sources for powering data centers that support e-
tion of renewable energy sources. This could include the commerce platforms. By integrating energy-conscious prac-
development of advanced control algorithms and monitoring tices and technologies, researchers aim to reduce the envi-
systems to improve the stability of the grid and the integration ronmental impact of online transactions and digital services.
of energy storage systems to store excess energy gener- Such research endeavors seek to optimize energy usage,
ated from renewable sources. Another important aspect of improve operational efficiency, and promote eco-friendly
future research will be the development of advanced demand practices within the e-commerce and e-government domains.
management systems that can dynamically adjust energy con- These efforts can contribute to a more sustainable and envi-
sumption as a result of supply and demand fluctuations. ronmentally conscious digital economy.
Green transportation systems, or the development of sus- Green agriculture, which refers to the development of
tainable and environmentally friendly modes of transporta- environmentally sustainable methods for farming and food
tion, is a rapidly growing field. In the future, research in production, is a growing area of research. In the future,
green transportation systems will focus on developing and research in green agriculture can focus on developing meth-
improving alternative fuel vehicles, such as electric and ods for minimizing the environmental impact of agriculture,
hydrogen-powered vehicles, and exploring the integration of such as the use of sustainable crop management techniques,
these vehicles into existing transportation networks. Another the adoption of renewable energy sources, and the reduc-
important aspect of future research will be the development tion of chemical inputs. Another important area of research
of sustainable transportation infrastructure, such as charging can be the development of new technologies and techniques
and refueling stations, as well as the integration of smart for precision agriculture, such as the use of drones and
transportation systems to optimize the use of these assets. satellite imagery to optimize crop production and reduce
Green databases, or databases designed to minimize their waste.
environmental impact, are becoming increasingly impor- In the realm of green computing, exploring the future
tant as organizations seek to reduce their carbon footprint inter-dependency among various areas holds significant
and minimize their impact on the environment. In the promise for advancing sustainability efforts. The develop-
future, research in green databases will focus on developing ment of energy-efficient green data centers can directly
energy-efficient database management systems and exploring impact the overall efficiency of green cloud computing infras-
alternative energy sources for powering data centers. Another tructure, while the integration of green IoT devices with
area of research will be the development of techniques for transportation systems contributes to intelligent manage-

87484 VOLUME 11, 2023

S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

ment and reduced energy consumption. Green procurement ensuring a holistic understanding of the field. By incorpo-
practices are crucial in driving the adoption of sustainable rating key variables directly linked to green computing, the
technologies across green computing, promoting eco-friendly survey enables the measurement of specific indicators reflect-
solutions throughout the supply chain. The convergence of ing progress in addressing environmental impact and sustain-
green education and agriculture leverages smart technologies ability. Comparative analysis of different green computing
and data-driven decision-making to foster environmentally initiatives allows for assessing the impact of interventions,
conscious practices. Synergies between green databases, data identifying trends and patterns indicative of progress towards
mining, and machine learning may enable the development sustainable practices. The study employs a combination of
of energy-efficient algorithms for large-scale environmental quantitative and qualitative measures to provide a nuanced
data analysis. Recognizing and exploring these interdepen- understanding of challenges and potential solutions. Rigor-
dencies may lead to integrated approaches that propel sus- ous validation techniques and expert feedback enhance the
tainable development in the digital era, including establishing reliability and credibility of the survey findings, contributing
linkages between green data centers and cloud computing, to the advancement of the field.
integrating green IoT with smart grids, promoting green pro- This review study conducted an extensive analysis of
curement, and converging education with agriculture. Inter- studies spanning across multiple years and originating from
action between transportation systems and green databases, various countries. As the year advances, the number and
along with the integration of machine learning and agricul- quality of the studies conducted have increased. In this
ture, may unlock new avenues for synergy, innovation, and research work, most of the selected studies are from 2020.
transformative impact, paving the way toward a greener and China is the leading country in almost every area of green
more sustainable future in green computing. computing research. However, the USA and India have con-
In conclusion, green computing continues to be an impor- ducted numerous studies and have been able to dominate
tant area of research as the demand for more sustainable and in some areas. Apart from these nations, several countries
eco-friendly technology increases. Future research directions have made significant contributions by conducting research
in green computing could include the development of more in diverse fields of green computing, demonstrating the
efficient algorithms for data processing, energy consumption global trend toward sustainable and eco-friendly develop-
optimization in cloud computing, the usage of renewable ment. Efforts towards achieving sustainable and efficient
energy sources for powering data centers, and the exploration technology in green computing have yielded notable progress
of new materials for creating greener electronics. Addition- in various areas, including green data centers, green cloud
ally, researchers could investigate the integration of AI and computing, green IoT, smart grids, and green transporta-
ML techniques to enable dynamic power management and tion. The key developments and achievements in the areas
resource allocation in computing systems. include energy-efficient control and management systems,
algorithms, protocols, task schedulers, VM migration mech-
VII. CONCLUSION anisms and algorithms, renewable energy integrated smart
The implementation of sustainable technologies across mul- systems, green architecture framework development, smart
tiple green computing areas to achieve sustainable and IoT-based devices, and findings of green drivers and barri-
eco-friendly development has been analyzed and examined ers. Conversely, several areas of green computing, includ-
in this study. Identifying sustainable green approaches across ing green education, green databases, green data mining
multiple areas of green computing is the primary focus of this and information retrieval, green machine learning, green
study. Therefore, numerous studies have been obtained from e-commerce and e-government, and green agriculture, remain
various academic search engines through the use of multi- underdeveloped due to a range of impediments, such as
ple keywords. The primary studies obtained from academic insufficient funding, inadequate infrastructure, limited tech-
search engines have been further filtered using a structural nical expertise and knowledge, a lack of awareness, and
and defined approach. The assessment of the abstract, rigor- the intricacy of the issues. Energy consumption optimiza-
ous evaluation of the dataset or testing mechanisms, selec- tion and efficiency have been the primary focus of much
tion through inclusion and exclusion criteria, and meticulous of the research conducted in the field, with a multitude of
scrutiny of the methodological quality are among the primary objectives and goals being pursued. The integration and appli-
factors that determine the final selection of studies. Through cation of cutting-edge technologies such as AI and ML have
a structural and extreme quality-driven approach, from the demonstrated significant promise in promoting sustainable
initial 800 studies, 216 studies have been selected and dis- development in the field of environmentally-friendly comput-
tributed across twelve green computing areas. This study has ing. The utilization of AI and ML algorithms can enhance
analyzed the final selected studies by summarizing and exam- energy management and system control, leading to enhanced
ining various developed approaches, analyses, and designs. efficiency and decreased environmental consequences.
The proposed survey methodology in the context of green However, in the area of green computing, researchers face
computing offers a comprehensive approach to demonstrate significant barriers that impede their progress. One such
the resolution of the research problem. It involves exten- challenge is the lack of standardization, which hinders the
sive study collection from diverse sources and perspectives, establishment of consistent practices and frameworks across

VOLUME 11, 2023 87485

 S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

studies. Technical obstacles further complicate the imple- [5] Z. Li and Y. Lin, ‘‘Energy-saving study of green data center based
mentation of green computing solutions, as complex tech- on the natural cold source,’’ in Proc. 6th Int. Conf. Inf. Manag.,
Innov. Manag. Ind. Eng., vol. 3, Nov. 2013, pp. 355–358, doi:
nologies and infrastructure necessitate specialized knowledge 10.1109/ICIII.2013.6703591.
and resources. Limited funding adds to the constraints, [6] J. Stoess, C. Lang, and F. Bellosa, ‘‘Energy management for
as substantial financial support is often required for equip- hypervisor-based virtual machines, in Proc. USENIX, 2007,
pp. 1–14.
ment, data collection, and analysis. These collective barriers
[7] A. Gandhi, M. Harchol-Balter, R. Das, C. Lefurgy, and J. Kephart,
contribute to the complexities and challenges encountered by ‘‘Power capping via forced idleness,’’ in Proc. Workshop Energy Efficient
researchers in the field of green computing. Design, 2009, pp. 1–6.
The outcome of this research study has a substantial impact [8] H. P. Staff, ‘‘HP power capping and dynamic power capping for ProLiant
servers,’’ HP, Tech. Rep. C090303TB, 2009.
on the field of green computing. By investigating and analyz- [9] X. Wang, M. Chen, C. Lefurgy, and T. W. Keller, ‘‘SHIP: Scal-
ing various aspects related to green computing, this research able hierarchical power control for large-scale data centers,’’ in Proc.
contributes to the existing body of knowledge in the field. 18th Int. Conf. Parallel Architectures Compilation Techn., Sep. 2009,
pp. 91–100.
The findings of this study provide valuable insights and
[10] A. Jha, G. Gupta, S. Rau, P. Meshram, and N. Labhane, ‘‘Green comput-
advancements in the development of energy-efficient com- ing approach in ICT components,’’ Vidyabharati Int. Interdiscipl. Res. J.,
puting technologies, sustainable computing practices, and vol. 2022, pp. 564–569, Sep. 2022.
environmentally friendly IT solutions. [11] Green ICT 1.0 Taxonomy (Making IT More sustainable) | ICT4Green
by Donato Toppeta. Accessed Feb. 1, 2023. [Online]. Available:
The significant information obtained, examined, and pro- https://ict4green.wordpress.com/2010/01/15/green-ict-1-0-taxonomy/
vided in this review study regarding green computing is cur- [12] X. Gao, A. Curtis, and B. Wong, ‘‘It’s not easy being green,’’ ACM
rent and reliable. The precise keywords have been utilized to SIGCOMM Comput. Commun. Rev., vol. 42, no. 1, pp. 211–222,
Aug. 2012.
search for appropriate content and aid in fulfilling the objec-
[13] C. Ren, D. Wang, B. Urgaonkar, and A. Sivasubramaniam, ‘‘Carbon-
tives of the study. While conducting research and selecting aware energy capacity planning for datacenters,’’ in Proc. IEEE 20th Int.
specific keywords, there is a potential risk of overlooking sig- Symp. Model., Anal. Simul. Comput. Telecommun. Syst., Washington, DC,
nificant sources that may not be captured by the chosen terms. USA, Aug. 2012, pp. 391–400.
[14] L.-D. Radu, ‘‘Determinants of green ICT adoption in organizations: A
This limitation could result in the omission of relevant studies theoretical perspective,’’ Sustainability, vol. 8, no. 8, p. 731, Jul. 2016,
from the analysis. Additionally, during the data extraction and doi: 10.3390/su8080731.
analysis process, it is possible to miss certain data points, [15] S. B. Othman, F. A. Almalki, C. Chakraborty, and H. Sakli, ‘‘Privacy-
preserving aware data aggregation for IoT-based healthcare with green
which may affect the comprehensiveness and accuracy of the
computing technologies,’’ Comput. Electr. Eng., vol. 101, Jul. 2022,
findings. These considerations highlight the need for caution Art. no. 108025, doi: 10.1016/j.compeleceng.2022.108025.
and ongoing refinement in the research methodology to mit- [16] A. Srivastava, A. Singh, S. G. Joseph, M. Rajkumar, Y. D. Borole,
igate the potential impact of these limitations. and H. K. Singh, ‘‘WSN-IoT clustering for secure data transmission
in e-health sector using green computing strategy,’’ in Proc. 9th Int.
Due to the increasing significance of green computing, Conf. Cyber IT Service Manag. (CITSM), Sep. 2021, pp. 1–8, doi:
numerous studies are currently being conducted in this field. 10.1109/CITSM52892.2021.9588977.
Hence, in order to remain current, a consistent examination [17] A. Q. Gill, D. Bunker, and P. Seltsikas, ‘‘An empirical analysis of
cloud, mobile, social and green computing: Financial services IT
of research pertaining to diverse areas of green comput- strategy and enterprise architecture,’’ in Proc. IEEE 9th Int. Conf.
ing may be conducted. Future investigations in the field of Dependable, Autonomic Secure Comput., Dec. 2011, pp. 697–704, doi:
green computing may entail the identification and resolution 10.1109/DASC.2011.122.
[18] M. Dhaini, M. Jaber, A. Fakhereldine, S. Hamdan, and R. A. Haraty,
of area-specific hindrances and impediments. Forthcoming
‘‘Green computing approaches—A survey,’’ Informatica, vol. 45, no. 1,
research endeavors may encompass an examination of the pp. 1–12, Mar. 2021, doi: 10.31449/inf.v45i1.2998.
interrelatedness among various areas of green computing. [19] O. Castillo and P. Melin, ‘‘Review on the interactions of green com-
Finally, in order to achieve a sustainable and environmentally puting and computational intelligence techniques and their applications
to real-world problems,’’ J. Smart Environ. Green Comput., vol. 1, p. 2,
conscious future, it is crucial to implement eco-friendly alter- Jan. 2021, doi: 10.20517/jsegc.2021.01.
natives and engage in various awareness-raising initiatives. [20] I. Hamizi, Z. Kholmatova, and G. Succi, ‘‘Exploring research hypothe-
ses in green computing,’’ J. Smart Environ. Green Comput., vol. 1,
pp. 110–120, Jan. 2021, doi: 10.20517/jsegc.2021.05.
REFERENCES [21] A. Khelifi, O. Aziz, M. S. Farooq, A. Abid, and F. Bukhari, ‘‘Social
[1] S. Murugesan, ‘‘Harnessing green IT: Principles and practices,’’ IT Prof., and economic contribution of 5G and blockchain with green comput-
vol. 10, no. 1, pp. 24–33, 2008, doi: 10.1109/MITP.2008.10. ing: Taxonomy, challenges, and opportunities,’’ IEEE Access, vol. 9,
[2] P. Malviya and S. Singh, ‘‘A study about green computing,’’ Int. pp. 69082–69099, 2021, doi: 10.1109/ACCESS.2021.3075642.
J. Adv. Res. Comput. Sci. Softw. Eng., vol. 3, no. 6, pp. 790–794, [22] C. Geetha, ‘‘A review on applications of green computing with bots for
Jun. 2013. [Online]. Available: http://www.ijarcsse.com/docs/ environmental sustainability,’’ J. Data Acquisition Process., vol. 38, no.
papers/Volume_3/6_June2 013/V3I6-0320.pdf 2, pp. 196–203, 2023. [Online]. Available: https://sjcjycl.cn/article/view-
[3] B. A. Weerts, D. Gallaher, R. Weaver, and O. VanGeet, ‘‘Green data center 2023/02_196.php
cooling: Achieving 90% reduction: Airside economization and unique [23] S. Sagar and N. Pradhan, ‘‘A review: Recent trends in green comput-
indirect evaporative cooling,’’ in Proc. IEEE Green Technol. Conf., Tulsa, ing,’’ in Green Computing in Smart Cities: Simulation and Techniques,
OK, USA, Apr. 2012, pp. 1–6, doi: 10.1109/GREEN.2012.6200950. B. Balusamy, N. Chilamkurti, and S. Kadry, Eds. Cham, Switzerland:
[4] Anup R. Nimje, V. T. Gaikwad, and H. N. Datir, ‘‘Green cloud Springer, 2021, pp. 19–34, doi: 10.1007/978-3-030-48141-4_2.
computing: A virtualized security framework for green cloud [24] A. M. Aslam and M. Kaur, ‘‘A review on energy efficient techniques
computing,’’ Int. J. Adv. Res. Comput. Sci. Softw. Eng., vol. 3, no. 4, in green cloud open research challenges and issues,’’ Int. J. Sci. Res.
pp. 642–646, Apr. 2013. [Online]. Available: http://www.ijarcsse. Comput. Sci. Eng., vol. 6, no. 3, pp. 44–50, Jun. 2018, doi: 10.26438/ijs-
com/docs/papers/Volume_3/4_Apr. 2013/V3I4-0125.pdf rcse/v6i3.4450.

87486 VOLUME 11, 2023

S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

[25] L.-D. Radu, ‘‘Green cloud computing: A literature survey,’’ Symmetry, [43] V. Thayananthan, O. Abdulkader, K. Jambi, and A. M. Bamahdi,
vol. 9, no. 12, p. 295, Nov. 2017, doi: 10.3390/sym9120295. ‘‘Analysis of cybersecurity based on Li-Fi in green data storage envi-
[26] H. Yuan, C.-C.-J. Kuo, and I. Ahmad, ‘‘Energy efficiency in data ronments,’’ in Proc. IEEE 4th Int. Conf. Cyber Secur. Cloud Com-
centers and cloud-based multimedia services: An overview and put. (CSCloud), New York, NY, USA, Jun. 2017, pp. 327–332, doi:
future directions,’’ in Proc. Int. Conf. Green Comput., Chicago, 10.1109/CSCloud.2017.32.
IL, USA, Aug. 2010, pp. 375–382, doi: 10.1109/GREENCOMP.2010. [44] T. Chen, A. G. Marques, and G. B. Giannakis, ‘‘Space-time scheduling
5598292. for green data center networks,’’ in Proc. 50th Asilomar Conf. Signals,
[27] S. Benhamaid, A. Bouabdallah, and H. Lakhlef, ‘‘Recent advances in Syst. Comput., Pacific Grove, CA, USA, Nov. 2016, pp. 795–799, doi:
energy management for green-IoT: An up-to-date and comprehensive 10.1109/ACSSC.2016.7869155.
survey,’’ J. Netw. Comput. Appl., vol. 198, Feb. 2022, Art. no. 103257, [45] E. Baccour, S. Foufou, and R. Hamila, ‘‘A guaranteed performance of a
doi: 10.1016/j.jnca.2021.103257. green data center based on the contribution of vital nodes,’’ in Proc. IEEE
[28] S. G. Paul, A. Saha, A. A. Biswas, M. S. Zulfiker, M. S. Arefin, Globecom Workshops (GC Wkshps), Washington, DC, USA, Dec. 2016,
M. M. Rahman, and A. W. Reza, ‘‘Combating COVID-19 using pp. 1–6, doi: 10.1109/GLOCOMW.2016.7848812.
machine learning and deep learning: Applications, challenges, and [46] M. Ghamkhari and H. Mohsenian-Rad, ‘‘Profit maximization and power
future perspectives,’’ Array, vol. 17, Mar. 2023, Art. no. 100271, doi: management of green data centers supporting multiple SLAs,’’ in Proc.
10.1016/j.array.2022.100271. Int. Conf. Comput., Netw. Commun. (ICNC), San Diego, CA, USA,
[29] M. S. Mekala and P. Viswanathan, ‘‘A survey: Energy-efficient sensor Jan. 2013, pp. 465–469, doi: 10.1109/ICCNC.2013.6504129.
and VM selection approaches in green computing for X-IoT applica- [47] A. Kiani and N. Ansari, ‘‘Profit maximization for geographically dis-
tions,’’ Int. J. Comput. Appl., vol. 42, no. 3, pp. 290–305, Apr. 2020, doi: persed green data centers,’’ IEEE Trans. Smart Grid, vol. 9, no. 2, pp.
10.1080/1206212X.2018.1558511. 703–711, Mar. 2018, doi: 10.1109/TSG.2016.2562565.
[30] H. Lee, Y. Choi, T. Van Nguyen, Y. Hai, J. Kim, M. Bahja, and [48] M. Haddad, G. Da Costa, J.-M. Nicod, M.-C. Péra, J.-M. Pierson,
H. Hocaoǧlu, ‘‘COVID19 led virtualization: Green data center V. Rehn-Sonigo, P. Stolf, and C. Varnier, ‘‘Combined IT and power
for information systems research,’’ Inf. Syst. Manag., vol. 37, supply infrastructure sizing for standalone green data centers,’’ Sus-
no. 4, pp. 272–276, Oct. 2020, doi: 10.1080/10580530.2020. tain. Comput., Informat. Syst., vol. 30, Jun. 2021, Art. no. 100505, doi:
1818901. 10.1016/j.suscom.2020.100505.
[31] I. Masudin, S. Z. Umamy, C. N. Al-Imron, and D. P. Restuputri, ‘‘Green [49] L. Wang, F. Zhang, A. V. Vasilakos, C. Hou, and Z. Liu, ‘‘Joint vir-
procurement implementation through supplier selection: A bibliometric tual machine assignment and traffic engineering for green data center
review,’’ Cogent Eng., vol. 9, no. 1, Dec. 2022, Art. no. 2119686, doi: networks,’’ ACM SIGMETRICS Perform. Eval. Rev., vol. 41, no. 3,
10.1080/23311916.2022.2119686. pp. 107–112, Jan. 2014, doi: 10.1145/2567529.2567560.
[32] M. Massaoudi, H. Abu-Rub, S. S. Refaat, I. Chihi, and F. S. Oueslati, [50] Y. Wu, M. Tornatore, S. Ferdousi, and B. Mukherjee, ‘‘Green
‘‘Deep learning in smart grid technology: A review of recent advance- data center placement in optical cloud networks,’’ IEEE Trans.
ments and future prospects,’’ IEEE Access, vol. 9, pp. 54558–54578, Green Commun. Netw., vol. 1, no. 3, pp. 347–357, Sep. 2017, doi:
2021, doi: 10.1109/ACCESS.2021.3071269. 10.1109/TGCN.2017.2709327.
[33] P. Brereton, B. A. Kitchenham, D. Budgen, M. Turner, and M. Khalil, [51] N. H. Thanh, B. D. Cuong, T. D. Thien, P. N. Nam, N. Q. Thu,
‘‘Lessons from applying the systematic literature review process within T. T. Huong, and T. M. Nam, ‘‘ECODANE: A customizable hybrid
the software engineering domain,’’ J. Syst. Softw., vol. 80, no. 4, testbed for green data center networks,’’ in Proc. Int. Conf. Adv. Technol.
pp. 571–583, Apr. 2007, doi: 10.1016/j.jss.2006.07.009. Commun. (ATC ), Ho Chi Minh, Vietnam, Oct. 2013, pp. 312–317, doi:
[34] L. A. Barroso and U. Hölzle, ‘‘The case for energy-proportional 10.1109/ATC.2013.6698128.
computing,’’ Computer, vol. 40, no. 12, pp. 33–37, Dec. 2007, doi: [52] H. Yeganeh, A. Salahi, and M. A. Pourmina, ‘‘A novel cost opti-
10.1109/MC.2007.443. mization method for mobile cloud computing by capacity planning
[35] S. Mittal, ‘‘Power management techniques for data centers: A survey,’’ of green data center with dynamic pricing,’’ Can. J. Electr. Comput.
2014, arXiv:1404.6681. Eng., vol. 42, no. 1, pp. 41–51, Winter. 2019, doi: 10.1109/CJECE.2019.
[36] T. C. Patil and S. P. Duttagupta, ‘‘Hybrid self—Sustainable green power 2890833.
generation system for powering green data center,’’ in Proc. Int. Conf. [53] Y. Jin, Y. Wen, Q. Chen, and Z. Zhu, ‘‘An empirical investigation
Control, Instrum., Energy Commun. (CIEC), Calcutta, India, Jan. 2014, of the impact of server virtualization on energy efficiency for green
pp. 331–334, doi: 10.1109/CIEC.2014.6959104. data center,’’ Comput. J., vol. 56, no. 8, pp. 977–990, Aug. 2013, doi:
[37] K. Elahee and S. Jugoo, ‘‘Ocean thermal energy for air-conditioning: 10.1093/comjnl/bxt017.
Case study of a green data center,’’ Energy Sour., A, Recovery, Uti- [54] C. Sonu and A. Kothari, ‘‘Green data center using Spearmans ranking
lization, Environ. Effects, vol. 35, no. 7, pp. 679–684, Apr. 2013, doi: algorithm,’’ Int. J. Comput. Sci. Inf. Technol., vol. 6, no. 2, pp. 1672–1676,
10.1080/15567036.2010.504941. 2015.
[38] L. Liu, H. Wang, X. Liu, X. Jin, W. B. He, Q. B. Wang, and Y. Chen, [55] T. M. Nam, N. H. Thanh, and D. A. Tuan, ‘‘Green data center using
‘‘GreenCloud: A new architecture for green data center,’’ in Proc. centralized power-management of network and servers,’’ in Proc. Int.
6th Int. Conf. Ind. Session Autonomic Comput. Commun. Ind. Ses- Conf. Electron., Inf., Commun. (ICEIC), Danang, Vietnam, Jan. 2016,
sion, Barcelona, Spain, Jun. 2009, pp. 29–38, doi: 10.1145/1555312. pp. 1–4, doi: 10.1109/ELINFOCOM.2016.7562926.
1555319. [56] S. Zafar, S. Chaudhry, and S. Kiran, ‘‘Adaptive TrimTree: Green data
[39] T. Wang, Y. Xia, J. Muppala, M. Hamdi, and S. Foufou, ‘‘A center networks through resource consolidation, selective connectedness
general framework for performance guaranteed green data cen- and energy proportional computing,’’ Energies, vol. 9, no. 10, p. 797,
ter networking,’’ in Proc. IEEE Global Commun. Conf., Austin, Oct. 2016, doi: 10.3390/en9100797.
TX, USA, Dec. 2014, pp. 2510–2515, doi: 10.1109/GLOCOM.2014. [57] S. Yan, S. Xiao, Y. Chen, Y. Cui, and J. Liu, ‘‘GreenWay: Joint VM
7037185. placement and topology adaption for green data center networking,’’ in
[40] S. K. S. Gupta, A. Banerjee, Z. Abbasi, G. Varsamopoulos, M. Jonas, Proc. 26th Int. Conf. Comput. Commun. Netw. (ICCCN), Vancouver, BC,
J. Ferguson, R. R. Gilbert, and T. Mukherjee, ‘‘GDCSim: A simula- Canada, Jul. 2017, pp. 1–9, doi: 10.1109/ICCCN.2017.8038372.
tor for green data center design and analysis,’’ ACM Trans. Model. [58] M. Kushwaha, A. Singh, and B. L. Raina, ‘‘Categorization of metrics for
Comput. Simul., vol. 24, no. 1, pp. 1–27, Jan. 2014, doi: 10.1145/ improving efficiency of green data centers,’’ in Proc. Int. Conf. Com-
2553083. put. Intell. Knowl. Economy (ICCIKE), Dubai, United Arab Emirates,
[41] N. A. Emran, N. Abdullah, and M. N. M. Isa, ‘‘Storage space optimisation Dec. 2019, pp. 56–59, doi: 10.1109/ICCIKE47802.2019.9004437.
for green data center,’’ Proc. Eng., vol. 53, pp. 483–490, Jan. 2013, doi: [59] L. Xu, C. Li, L. Li, Y. Liu, Z. Yang, and Y. Liu, ‘‘A virtual data cen-
10.1016/j.proeng.2013.02.062. ter deployment model based on the green cloud computing,’’ in Proc.
[42] S. Bird, A. Achuthan, O. Ait Maatallah, W. Hu, K. Janoyan, A. Kwasinski, IEEE/ACIS 13th Int. Conf. Comput. Inf. Sci. (ICIS), Taiyuan, China,
J. Matthews, D. Mayhew, J. Owen, and P. Marzocca, ‘‘Distributed (green) Jun. 2014, pp. 235–240, doi: 10.1109/ICIS.2014.6912140.
data centers: A new concept for energy, computing, and telecommunica- [60] H. Li, J. Liu, and Q. Zhou, ‘‘Research on energy-saving virtual machine
tions,’’ Energy for Sustain. Develop., vol. 19, pp. 83–91, Apr. 2014, doi: migration algorithm for green data center,’’ IET Control Theory Appl.,
10.1016/j.esd.2013.12.006. early access, pp. 1–10, 2022, doi: 10.1049/cth2.12401.

VOLUME 11, 2023 87487

 S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

[61] M. Jonas, R. R. Gilbert, J. Ferguson, G. Varsamopoulos, and [79] L. Xu, K. Wang, Z. Ouyang, and X. Qi, ‘‘An improved binary PSO-based
S. K. S. Gupta, ‘‘A transient model for data center thermal prediction,’’ in task scheduling algorithm in green cloud computing,’’ in Proc. 9th Int.
Proc. Int. Green Comput. Conf. (IGCC), San Jose, CA, USA, Jun. 2012, Conf. Commun. Netw. China, Maoming, China, Aug. 2014, pp. 126–131,
pp. 1–10, doi: 10.1109/IGCC.2012.6322262. doi: 10.1109/CHINACOM.2014.7054272.
[62] D. Li, Y. Shang, and C. Chen, ‘‘Software defined green data center [80] V. R. Reguri, S. Kogatam, and M. Moh, ‘‘Energy efficient
network with exclusive routing,’’ in Proc. IEEE INFOCOM Conf. Com- traffic-aware virtual machine migration in green cloud data
put. Commun., Toronto, ON, Canada, Apr. 2014, pp. 1743–1751, doi: centers,’’ in Proc. IEEE 2nd Int. Conf. Big Data Secur. Cloud
10.1109/INFOCOM.2014.6848112. (BigDataSecurity), Int. Conf. High Perform. Smart Comput. (HPSC),
[63] H. Lei, R. Wang, T. Zhang, Y. Liu, and Y. Zha, ‘‘A multi-objective IEEE Int. Conf. Intell. Data Secur. (IDS), New York, NY, USA,
co-evolutionary algorithm for energy-efficient scheduling on a green Apr. 2016, pp. 268–273, doi: 10.1109/BigDataSecurity-HPSC-IDS.
data center,’’ Comput. Oper. Res., vol. 75, pp. 103–117, Nov. 2016, doi: 2016.55.
10.1016/j.cor.2016.05.014. [81] Md. K. Hossain, M. Rahman, A. Hossain, S. Y. Rahman, and
[64] P. Guo, S. Wang, Y. Lei, and J. Li, ‘‘Numerical simulation of solar M. M. Islam, ‘‘Active & idle virtual machine migration algorithm—
chimney-based direct airside free cooling system for green data cen- A new ant colony optimization approach to consolidate virtual
ters,’’ J. Building Eng., vol. 32, Nov. 2020, Art. no. 101793, doi: machines and ensure green cloud computing,’’ in Proc. Emerg. Tech-
10.1016/j.jobe.2020.101793. nol. Comput., Commun. Electron. (ETCCE), Dec. 2020, pp. 1–6, doi:
[65] Q. Zhou, J. Lou, and Y. Jiang, ‘‘Optimization of energy consump- 10.1109/ETCCE51779.2020.9350915.
tion of green data center in e-commerce,’’ Sustain. Comput., Informat. [82] M. H. Fathi and L. M. Khanli, ‘‘Consolidating VMs in green
Syst., vol. 23, pp. 103–110, Sep. 2019, doi: 10.1016/j.suscom.2019. cloud computing using harmony search algorithm,’’ in Proc. 1st Int.
07.008. Conf. Internet E-Business, Singapore, Apr. 2018, pp. 146–151, doi:
[66] A. Rahaman, K. N. Noor, T. A. Abir, S. Rana, and M. Ali, ‘‘Design 10.1145/3230348.3230369.
and analysis of sustainable green data center with hybrid energy [83] J. I. Torrens, D. Mehta, V. Zavrel, D. Grimes, T. Scherer, R. Birke,
sources,’’ J. Power Energy Eng., vol. 9, no. 7, pp. 76–88, 2021, doi: L. Chen, S. Rea, L. Lopez, E. Pages, and D. Pesch, ‘‘Integrated energy
10.4236/jpee.2021.97006. efficient data centre management for green cloud computing—The FP7
[67] M. Uddin, A. Shah, and A. Rehman, ‘‘Metrics for computing performance GENiC project experience,’’ in Proc. 6th Int. Conf. Cloud Comput. Ser-
of data center for instigating energy efficient data centers,’’ J. Sci. Ind. vices Sci., 2016, pp. 375–386, doi: 10.5220/0005928003750386.
Res., vol. 73, pp. 11–15, Jan. 2014. [84] W. Yue and Q. Chen, ‘‘Dynamic placement of virtual machines with
[68] W. Xia, Y. Wen, K.-C. Toh, and Y.-W. Wong, ‘‘Toward green data both deterministic and stochastic demands for green cloud comput-
centers as an interruptible load for grid stabilization in Singapore,’’ ing,’’ Math. Problems Eng., vol. 2014, pp. 1–11, Jan. 2014, doi:
IEEE Commun. Mag., vol. 53, no. 11, pp. 192–198, Nov. 2015, doi: 10.1155/2014/613719.
10.1109/MCOM.2015.7321990. [85] S. K. Garg, C. S. Yeo, and R. Buyya, ‘‘Green cloud framework for improv-
[69] X. Hu, P. Li, and Y. Sun, ‘‘Minimizing energy cost for green data ing carbon efficiency of clouds,’’ in Euro-Par 2011 Parallel Processing
center by exploring heterogeneous energy resource,’’ J. Modern Power (Lecture Notes in Computer Science), vol. 6852, E. Jeannot, R. Namyst,
Syst. Clean Energy, vol. 9, no. 1, pp. 148–159, Jan. 2021, doi: and J. Roman, Eds. Berlin, Germany: Springer, 2011, pp. 491–502, doi:
10.35833/MPCE.2019.000052. 10.1007/978-3-642-23400-2_45.
[70] F.-H. Tseng, C.-Y. Chen, L.-D. Chou, H.-C. Chao, and J.-W. Niu, [86] M. Guazzone, C. Anglano, and M. Canonico, ‘‘Exploiting VM migration
‘‘Service-oriented virtual machine placement optimization for green data for the automated power and performance management of green cloud
center,’’ Mobile Netw. Appl., vol. 20, no. 5, pp. 556–566, Oct. 2015, doi: computing systems,’’ in Energy Efficient Data Centers (Lecture Notes
10.1007/s11036-015-0600-9. in Computer Science), vol. 7396, J. Huusko, H. De Meer, S. Klingert,
[71] H. Yuan, J. Bi, M. Zhou, and A. C. Ammari, ‘‘Time-aware multi- and A. Somov, Eds. Berlin, Germany: Springer, 2012, pp. 81–92, doi:
application task scheduling with guaranteed delay constraints in green 10.1007/978-3-642-33645-4_8.
data center,’’ IEEE Trans. Autom. Sci. Eng., vol. 15, no. 3, pp. 1138–1151, [87] M. N. Hulkury and M. R. Doomun, ‘‘Integrated green cloud comput-
Jul. 2018, doi: 10.1109/TASE.2017.2741965. ing architecture,’’ in Proc. Int. Conf. Adv. Comput. Sci. Appl. Tech-
[72] L. Ganesh, H. Weatherspoon, T. Marian, and K. Birman, ‘‘Inte- nol. (ACSAT), Kuala Lumpur, Malaysia, Nov. 2012, pp. 269–274, doi:
grated approach to data center power management,’’ IEEE Trans. 10.1109/ACSAT.2012.16.
Comput., vol. 62, no. 6, pp. 1086–1096, Jun. 2013, doi: 10.1109/TC. [88] F. Farahnakian, T. Pahikkala, P. Liljeberg, J. Plosila, and H. Tenhunen,
2013.32. ‘‘Utilization prediction aware VM consolidation approach for green cloud
[73] T. Yang, Y. Hou, Y. C. Lee, H. Ji, and A. Y. Zomaya, ‘‘Power con- computing,’’ in Proc. IEEE 8th Int. Conf. Cloud Comput., New York City,
trol framework for green data centers,’’ IEEE Trans. Cloud Comput., NY, USA, Jun. 2015, pp. 381–388, doi: 10.1109/CLOUD.2015.58.
vol. 10, no. 4, pp. 2876–2886, Oct. 2022, doi: 10.1109/TCC.2020. [89] G. Kaur and S. Midha, ‘‘A preemptive priority based job scheduling
3022789. algorithm in green cloud computing,’’ in Proc. 6th Int. Conf.-Cloud Syst.
[74] V. Bindhu and M. Joe, ‘‘Green cloud computing solution for opera- Big Data Eng., Noida, India, Jan. 2016, pp. 152–156, doi: 10.1109/CON-
tional cost efficiency and environmental impact reduction,’’ J. ISMAC, FLUENCE.2016.7508105.
vol. 1, no. 2, pp. 40–48, Jun. 2019, doi: 10.36548/jismac.2019. [90] A. Alarifi, K. Dubey, M. Amoon, T. Altameem, F. E. A. El-Samie,
2.005. A. Altameem, S. C. Sharma, and A. A. Nasr, ‘‘Energy-efficient
[75] W. Shu, W. Wang, and Y. Wang, ‘‘A novel energy-efficient resource hybrid framework for green cloud computing,’’ IEEE Access, vol. 8,
allocation algorithm based on immune clonal optimization for pp. 115356–115369, 2020, doi: 10.1109/ACCESS.2020.3002184.
green cloud computing,’’ EURASIP J. Wireless Commun. Netw., [91] M. Anan and N. Nasser, ‘‘SLA-based optimization of energy effi-
vol. 2014, no. 1, p. 64, Dec. 2014, doi: 10.1186/1687-1499- ciency for green cloud computing,’’ in Proc. IEEE Global Commun.
2014-64. Conf. (GLOBECOM), San Diego, CA, USA, Dec. 2015, pp. 1–6, doi:
[76] S. R. Hussein, Y. Alkabani, and H. K. Mohamed, ‘‘Green cloud comput- 10.1109/GLOCOM.2015.7417712.
ing: Datacenters power management policies and algorithms,’’ in Proc. [92] T. T. Huu and C.-K. Tham, ‘‘An auction-based resource allocation
9th Int. Conf. Comput. Eng. Syst. (ICCES), Cairo, Egypt, Dec. 2014, model for green cloud computing,’’ in Proc. IEEE Int. Conf. Cloud
pp. 421–426, doi: 10.1109/ICCES.2014.7030998. Eng. (ICE), Redwood City, CA, USA, Mar. 2013, pp. 269–278, doi:
[77] Y. Liu, W. Shu, and C. Zhang, ‘‘A parallel task scheduling optimiza- 10.1109/IC2E.2013.21.
tion algorithm based on clonal operator in green cloud computing,’’ J. [93] F. Cao and M. M. Zhu, ‘‘Energy efficient workflow job scheduling for
Commun., vol. 11, no. 2, pp. 185–191, 2016, doi: 10.12720/jcm.11.2. green cloud,’’ in Proc. IEEE Int. Symp. Parallel Distrib. Process., Work-
185-191. shops Phd Forum, Cambridge, MA, USA, May 2013, pp. 2218–2221, doi:
[78] A. A. A. Ari, I. Damakoa, C. Titouna, N. Labraoui, and A. Gueroui, 10.1109/IPDPSW.2013.19.
‘‘Efficient and scalable ACO-based task scheduling for green cloud [94] W. Shu, K. Cai, and N. N. Xiong, ‘‘Research on strong agile response
computing environment,’’ in Proc. IEEE Int. Conf. Smart Cloud (Smart- task scheduling optimization enhancement with optimal resource usage
Cloud), New York, NY, USA, Nov. 2017, pp. 66–71, doi: 10.1109/Smart- in green cloud computing,’’ Future Gener. Comput. Syst., vol. 124,
Cloud.2017.17. pp. 12–20, Nov. 2021, doi: 10.1016/j.future.2021.05.012.

87488 VOLUME 11, 2023

S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

[95] B. Kaur and A. Kaur, ‘‘An efficient approach for green cloud com- [114] O. Said, Z. Al-Makhadmeh, and A. Tolba, ‘‘EMS: An energy man-
puting using genetic algorithm,’’ in Proc. 1st Int. Conf. Next Gener. agement scheme for green IoT environments,’’ IEEE Access, vol. 8,
Comput. Technol. (NGCT), Dehradun, India, Sep. 2015, pp. 10–15, doi: pp. 44983–44998, 2020, doi: 10.1109/ACCESS.2020.2976641.
10.1109/NGCT.2015.7375073. [115] Y. Ren, X. Zhang, and G. Lu, ‘‘The wireless solution to realize green IoT:
[96] K. L. Giridas and A. S. Nargunam, ‘‘Optimal resource allocation tech- Cellular networks with energy efficient and energy harvesting schemes,’’
nique (ORAT) for green cloud computing,’’ Int. J. Comput. Appl., vol. 55, Energies, vol. 13, no. 22, p. 5875, Nov. 2020, doi: 10.3390/en13225875.
no. 5, pp. 20–26, Oct. 2012, doi: 10.5120/8752-2648. [116] P. Gupta, V. K. Jain, and S. Aggarwal, ‘‘Exploring relationship between
[97] W. Itani, C. Ghali, A. Kayssi, A. Chehab, and I. Elhajj, ‘‘G-route: employees well-being and green IoT using structural equation modeling,’’
An energy-aware service routing protocol for green cloud comput- IJAST, vol. 29, pp. 2590–2600, May 2020.
ing,’’ Cluster Comput., vol. 18, no. 2, pp. 889–908, Jun. 2015, doi: [117] S. Qiao, Q. Zhang, Q. Zhang, F. Guo, and W. Li, ‘‘Hybrid
10.1007/s10586-015-0443-y. seismic-electrical data acquisition station based on cloud technology
[98] Y. Lu and N. Sun, ‘‘An effective task scheduling algorithm based and green IoT,’’ IEEE Access, vol. 8, pp. 31026–31033, 2020, doi:
on dynamic energy management and efficient resource utilization in 10.1109/ACCESS.2020.2966510.
green cloud computing environment,’’ Cluster Comput., vol. 22, no. 1, [118] S. F. Abedin, Md. G. R. Alam, R. Haw, and C. S. Hong, ‘‘A system model
pp. 513–520, Jan. 2019, doi: 10.1007/s10586-017-1272-y. for energy efficient green-IoT network,’’ in Proc. Int. Conf. Inf. Netw.
(ICOIN), Jan. 2015, pp. 177–182, doi: 10.1109/ICOIN.2015.7057878.
[99] H. M. Lee, Y.-S. Jeong, and H. J. Jang, ‘‘Performance analysis based
resource allocation for green cloud computing,’’ J. Supercomput., vol. 69, [119] S. Zeb, Q. Abbas, S. A. Hassan, A. Mahmood, R. Mumtaz,
no. 3, pp. 1013–1026, Sep. 2014, doi: 10.1007/s11227-013-1020-x. S. M. H. Zaidi, S. A. R. Zaidi, and M. Gidlund, ‘‘NOMA enhanced
backscatter communication for green IoT networks,’’ in Proc. 16th Int.
[100] S. K. Mishra, D. Puthal, B. Sahoo, S. K. Jena, and M. S. Obaidat,
Symp. Wireless Commun. Syst. (ISWCS), Oulu, Finland, Aug. 2019,
‘‘An adaptive task allocation technique for green cloud comput-
pp. 640–644, doi: 10.1109/ISWCS.2019.8877102.
ing,’’ J. Supercomput., vol. 74, no. 1, pp. 370–385, Jan. 2018, doi:
[120] Z. Na, X. Wang, J. Shi, C. Liu, Y. Liu, and Z. Gao, ‘‘Joint resource
10.1007/s11227-017-2133-4.
allocation for cognitive OFDM-NOMA systems with energy harvesting
[101] R. Mandal, M. K. Mondal, S. Banerjee, and U. Biswas, ‘‘An approach in green IoT,’’ Ad Hoc Netw., vol. 107, Oct. 2020, Art. no. 102221, doi:
toward design and development of an energy-aware VM selection policy 10.1016/j.adhoc.2020.102221.
with improved SLA violation in the domain of green cloud comput-
[121] D. Deng, J. Xia, L. Fan, and X. Li, ‘‘Link selection in buffer-aided coop-
ing,’’ J. Supercomput., vol. 76, no. 9, pp. 7374–7393, Sep. 2020, doi:
erative networks for green IoT,’’ IEEE Access, vol. 8, pp. 30763–30771,
10.1007/s11227-020-03165-6.
2020, doi: 10.1109/ACCESS.2020.2972698.
[102] A. H. Ismail, N. A. El-Bahnasawy, and H. F. A. Hamed, ‘‘AGCM: Active [122] H. Kaur, S. K. Sood, and M. Bhatia, ‘‘Cloud-assisted green IoT-enabled
queue management-based green cloud model for mobile edge comput- comprehensive framework for wildfire monitoring,’’ Cluster Comput.,
ing,’’ Wireless Pers. Commun., vol. 105, no. 3, pp. 765–785, Apr. 2019, vol. 23, no. 2, pp. 1149–1162, Jun. 2020, doi: 10.1007/s10586-019-
doi: 10.1007/s11277-019-06119-1. 02981-7.
[103] A. Mohammadzadeh, M. Masdari, F. S. Gharehchopogh, and A. Jafarian, [123] A. S. H. Abdul-Qawy and T. Srinivasulu, ‘‘SEES: A scalable and energy-
‘‘Improved chaotic binary grey wolf optimization algorithm for workflow efficient scheme for green IoT-based heterogeneous wireless nodes,’’
scheduling in green cloud computing,’’ Evol. Intell., vol. 14, no. 4, J. Ambient Intell. Humanized Comput., vol. 10, no. 4, pp. 1571–1596,
pp. 1997–2025, Dec. 2021, doi: 10.1007/s12065-020-00479-5. Apr. 2019, doi: 10.1007/s12652-018-0758-7.
[104] R. Beik, ‘‘Green cloud computing: An energy-aware layer in soft- [124] R. Sukjaimuk, Q. Nguyen, and T. Sato, ‘‘A smart congestion con-
ware architecture,’’ in Proc. Spring Congr. Eng. Technol., Xi’an, China, trol mechanism for the green IoT sensor-enabled information-centric
May 2012, pp. 1–4, doi: 10.1109/SCET.2012.6341950. networking,’’ Sensors, vol. 18, no. 9, p. 2889, Aug. 2018, doi:
[105] F. Farahnakian, A. Ashraf, T. Pahikkala, P. Liljeberg, J. Plosila, I. Porres, 10.3390/s18092889.
and H. Tenhunen, ‘‘Using ant colony system to consolidate VMs for [125] G. Chen, J. Wu, W. Yang, A. K. Bashir, G. Li, and M. Hammoudeh,
green cloud computing,’’ IEEE Trans. Services Comput., vol. 8, no. 2, ‘‘Leveraging graph convolutional-LSTM for energy-efficient caching in
pp. 187–198, Mar. 2015, doi: 10.1109/TSC.2014.2382555. blockchain-based green IoT,’’ IEEE Trans. Green Commun. Netw., vol. 5,
[106] R. Shaw, E. Howley, and E. Barrett, ‘‘A predictive anti-correlated vir- no. 3, pp. 1154–1164, Sep. 2021, doi: 10.1109/TGCN.2021.3069395.
tual machine placement algorithm for green cloud computing,’’ in Proc. [126] H. A. B. Salameh, M. B. Irshaid, A. Al Ajlouni, and M. Aloqaily,
IEEE/ACM 11th Int. Conf. Utility Cloud Comput. (UCC), Dec. 2018, ‘‘Energy-efficient cross-layer spectrum sharing in CR green IoT net-
pp. 267–276, doi: 10.1109/UCC.2018.00035. works,’’ IEEE Trans. Green Commun. Netw., vol. 5, no. 3, pp. 1091–1100,
[107] L. A. Rocha and E. Cardozo, ‘‘A hybrid optimization model for green Sep. 2021, doi: 10.1109/TGCN.2021.3076695.
cloud computing,’’ in Proc. IEEE/ACM 7th Int. Conf. Utility Cloud Com- [127] J. Li, Y. Liu, Z. Zhang, J. Ren, and N. Zhao, ‘‘Towards green IoT network-
put., London, U.K., Dec. 2014, pp. 11–20, doi: 10.1109/UCC.2014.9. ing: Performance optimization of network coding based communication
[108] P. Fraga-Lamas, S. I. Lopes, and T. M. Fernández-Caramés, ‘‘Green and reliable storage,’’ IEEE Access, vol. 5, pp. 8780–8791, 2017, doi:
IoT and edge AI as key technological enablers for a sustainable digital 10.1109/ACCESS.2017.2706328.
transition towards a smart circular economy: An Industry 5.0 use case,’’ [128] L. Xu, M. Qin, Q. Yang, and K. Kwak, ‘‘Deep reinforcement learn-
Sensors, vol. 21, no. 17, p. 5745, Aug. 2021, doi: 10.3390/s21175745. ing for dynamic access control with battery prediction for mobile-edge
computing in green IoT networks,’’ in Proc. 11th Int. Conf. Wireless
[109] J. Bai, Z. Zeng, K. M. Abualnaja, and N. N. Xiong, ‘‘ADCC: An effec-
Commun. Signal Process. (WCSP), Xi’an, China, Oct. 2019, pp. 1–6, doi:
tive adaptive duty cycle control scheme for real time big data in green
10.1109/WCSP.2019.8927946.
IoT,’’ Alexandria Eng. J., vol. 61, no. 8, pp. 5959–5975, Aug. 2022, doi:
[129] H. Dai, H. Bian, C. Li, and B. Wang, ‘‘UAV-aided wireless commu-
10.1016/j.aej.2021.11.026.
nication design with energy constraint in space-air-ground integrated
[110] Q. Liu, S. Sun, H. Wang, and S. Zhang, ‘‘6G green IoT network: Joint green IoT networks,’’ IEEE Access, vol. 8, pp. 86251–86261, 2020, doi:
design of intelligent reflective surface and ambient backscatter com- 10.1109/ACCESS.2020.2992466.
munication,’’ Wireless Commun. Mobile Comput., vol. 2021, pp. 1–10, [130] S. B. Joseph, E. G. Dada, and M. S. Abdullahi, ‘‘Development of Internet
Jun. 2021, doi: 10.1155/2021/9912265. of Things (IoT) based energy consumption monitoring and device control
[111] A. Sant, L. Garg, P. Xuereb, and C. Chakraborty, ‘‘A novel green IoT- system,’’ NIPES J. Sci. Technol. Res., vol. 2, no. 3, p. 85, Aug. 2020, doi:
based pay-as-you-go smart parking system,’’ Comput., Mater. Continua, 10.37933/nipes/2.3.2020.9.
vol. 67, no. 3, pp. 3523–3544, 2021, doi: 10.32604/cmc.2021.015265. [131] B. Chowdhury and M. U. Chowdhury, ‘‘RFID-based real-time smart
[112] A. Heidari, M. A. J. Jamali, N. J. Navimipour, and S. Akbarpour, waste management system,’’ in Proc. Australas. Telecommun. Netw.
‘‘Deep Q-learning technique for offloading offline/online computation in Appl. Conf., Christchurch, New Zealand, Dec. 2007, pp. 175–180, doi:
blockchain-enabled green IoT-edge scenarios,’’ Appl. Sci., vol. 12, no. 16, 10.1109/ATNAC.2007.4665232.
p. 8232, Aug. 2022, doi: 10.3390/app12168232. [132] C. Blome, D. Hollos, and A. Paulraj, ‘‘Green procurement and green
[113] R. K. Lenka, A. K. Rath, and S. Sharma, ‘‘Building reliable rout- supplier development: Antecedents and effects on supplier perfor-
ing infrastructure for green IoT network,’’ IEEE Access, vol. 7, mance,’’ Int. J. Prod. Res., vol. 52, no. 1, pp. 32–49, Jan. 2014, doi:
pp. 129892–129909, 2019, doi: 10.1109/ACCESS.2019.2939883. 10.1080/00207543.2013.825748.

VOLUME 11, 2023 87489

 S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

[133] M. M. Islam, A. Turki, M. W. Murad, and A. Karim, ‘‘Do sustainable pro- [152] V. K. Sood, D. Fischer, J. M. Eklund, and T. Brown, ‘‘Developing a
curement practices improve organizational performance?’’ Sustainability, communication infrastructure for the smart grid,’’ in Proc. IEEE Electr.
vol. 9, no. 12, p. 2281, Dec. 2017, doi: 10.3390/su9122281. Power Energy Conf. (EPEC), Montreal, QC, Canada, Oct. 2009, pp. 1–7,
[134] S. Bag, ‘‘Identification of green procurement drivers and their doi: 10.1109/EPEC.2009.5420809.
interrelationship using total interpretive structural modelling,’’ Vis., [153] W. Bein, B. B. Madan, D. Bein, and D. Nyknahad, ‘‘Algorithmic
J. Bus. Perspective, vol. 21, no. 2, pp. 129–142, Jun. 2017, doi: approaches for a dependable smart grid,’’ in Information Technolog: New
10.1177/0972262917700990. Generations (Advances in Intelligent Systems and Computing), vol. 448,
[135] B. K. AlNuaimi, S. K. Singh, and B. Harney, ‘‘Unpacking the role of S. Latifi, Ed. Cham, Switzerland: Springer, 2016, pp. 677–687, doi:
innovation capability: Exploring the impact of leadership style on green 10.1007/978-3-319-32467-8_59.
procurement via a natural resource-based perspective,’’ J. Bus. Res., [154] B. Hu and H. Gharavi, ‘‘Greedy backpressure routing for smart
vol. 134, pp. 78–88, Sep. 2021, doi: 10.1016/j.jbusres.2021.05.026. grid sensor networks,’’ in Proc. IEEE 11th Consum. Commun. Netw.
[136] H. Song, K. Yu, and S. Zhang, ‘‘Green procurement, stakeholder satis- Conf. (CCNC), Las Vegas, NV, USA, Jan. 2014, pp. 32–37, doi:
faction and operational performance,’’ Int. J. Logistics Manag., vol. 28, 10.1109/CCNC.2014.6866544.
no. 4, pp. 1054–1077, Nov. 2017, doi: 10.1108/IJLM-12-2015-0234. [155] Q. D. La, Y. W. E. Chan, and B.-H. Soong, ‘‘Dynamic market for
[137] S. B. Rane and S. V. Thakker, ‘‘Green procurement process model based distributed energy resources in the smart grid,’’ in Proc. IEEE 11th
on blockchain-IoT integrated architecture for a sustainable business,’’ Consum. Commun. Netw. Conf. (CCNC), Las Vegas, NV, USA, Jan. 2014,
Manag. Environ. Qual., Int. J., vol. 31, no. 3, pp. 741–763, Dec. 2019, pp. 38–43, doi: 10.1109/CCNC.2014.6866545.
doi: 10.1108/MEQ-06-2019-0136. [156] Y. Li, J. C. Zhang, and M. Dong, ‘‘Smart grid in radio access networks
[138] D. J. Guo, P. B. Sarpong, H. A. Antwi, and I. A. Mensah, ‘‘Evalua- (SG-RAN): Smart energy management at cell-sites,’’ in Proc. IEEE 11th
tion of green procurement practices among mining companies’ hospitals Consum. Commun. Netw. Conf. (CCNC), Las Vegas, NV, USA, Jan. 2014,
in Ghana: A qualitative analysis,’’ Environ. Health Insights, vol. 14, pp. 44–49, doi: 10.1109/CCNC.2014.6866546.
Jan. 2020, Art. no. 117863021984311, doi: 10.1177/1178630219843115. [157] M. N. Faqiry and S. Das, ‘‘A budget balanced energy distribution mecha-
[139] A. Galeazzo, N. Ortiz-de-Mandojana, and J. Delgado-Ceballos, ‘‘Green nism among consumers and prosumers in microgrid,’’ in Proc. IEEE Int.
procurement and financial performance in the tourism industry: Conf. Internet Things (iThings), IEEE Green Comput. Commun. (Green-
The moderating role of tourists’ green purchasing behaviour,’’ Cur- Com), IEEE Cyber, Phys. Social Comput. (CPSCom), IEEE Smart Data
rent Issues Tourism, vol. 24, no. 5, pp. 700–716, Mar. 2021, doi: (SmartData), Dec. 2016, pp. 516–520, doi: 10.1109/iThings-GreenCom-
10.1080/13683500.2020.1734546. CPSCom-SmartData.2016.119.
[140] Z. Hu, C. Rao, Y. Zheng, and D. Huang, ‘‘Optimization decision of [158] K. A. Ahmed, Z. Aung, and D. Svetinovic, ‘‘Smart grid wireless net-
supplier selection in green procurement under the mode of low carbon work security requirements analysis,’’ in Proc. IEEE Int. Conf. Green
economy,’’ Int. J. Comput. Intell. Syst., vol. 8, no. 3, p. 407, 2015, doi: Comput. Commun. IEEE Internet Things IEEE Cyber, Phys. Social Com-
10.1080/18756891.2015.1017375. put., Beijing, China, Aug. 2013, pp. 871–878, doi: 10.1109/GreenCom-
[141] Y. Dou, J. Sarkis, and C. Bai, ‘‘Government green procurement: iThings-CPSCom.2013.153.
A fuzzy-DEMATEL analysis of barriers,’’ in Supply Chain Management [159] S. Misra, A. Mondal, S. Banik, M. Khatua, S. Bera, and M. S. Obaidat,
Under Fuzziness (Studies in Fuzziness and Soft Computing), vol. 313, ‘‘Residential energy management in smart grid: A Markov decision
C. Kahraman and B. Oztaysi, Eds. Berlin, Germany: Springer, 2014, process-based approach,’’ in Proc. IEEE Int. Conf. Green Comput. Com-
pp. 567–589, doi: 10.1007/978-3-642-53939-8_24. mun., IEEE Internet Things IEEE Cyber, Phys. Social Comput., Beijing,
[142] S. W. Chan, S. S. Tiwari, M. F. Ahmad, I. Zaman, and W. L. Sia, China, Aug. 2013, pp. 1152–1157, doi: 10.1109/GreenCom-iThings-
‘‘Green procurement practices and barriers in furniture manufacturing CPSCom.2013.200.
companies,’’ Int. J. Supply Chain Manag., vol. 7, no. 6, pp. 431–438, [160] L. Zhang, S. Tang, Y. Jiang, and Z. Ma, ‘‘Robust and efficient authen-
2018. tication protocol based on elliptic curve cryptography for smart grids,’’
[143] T. Abdallah, A. Diabat, and D. Simchi-Levi, ‘‘A carbon sensitive supply in Proc. IEEE Int. Conf. Green Comput. Commun., IEEE Internet
chain network problem with green procurement,’’ in Proc. 40th Int. Things IEEE Cyber, Phys. Social Comput., Beijing, China, Aug. 2013,
Conf. Comput. Indutrial Eng., Awaji City, Japan, Jul. 2010, pp. 1–6, doi: pp. 2089–2093, doi: 10.1109/GreenCom-iThings-CPSCom.2013.392.
10.1109/ICCIE.2010.5668278. [161] X. Miao and X. Chen, ‘‘Research on quantum-bit error correction coding
[144] M. A. Salam, ‘‘An empirical investigation of the determinants of adop- for smart grid substation,’’ in Proc. IEEE Int. Conf. Green Comput.
tion of green procurement for successful green supply chain manage- Commun. IEEE Internet Things IEEE Cyber, Phys. Social Comput.,
ment,’’ in Proc. 4th IEEE Int. Conf. Manag. Innov. Technol., Sep. 2008, Beijing, China, Aug. 2013, pp. 2145–2147, doi: 10.1109/GreenCom-
pp. 1038–1043, doi: 10.1109/ICMIT.2008.4654511. iThings-CPSCom.2013.404.
[145] S. Bag, ‘‘Buyer-supplier relationship and trust theory in innovative green [162] P. Kulkarni, S. Gormus, and Z. Fan, ‘‘Tree balancing in smart grid
procurement practices,’’ Int. J. Logistics Syst. Manag., vol. 30, no. 1, advanced metering infrastructure mesh networks,’’ in Proc. IEEE
p. 113, 2018, doi: 10.1504/IJLSM.2018.10012469. Int. Conf. Green Comput. Commun., Besancon, France, Nov. 2012,
[146] L. Shen, Z. Zhang, and X. Zhang, ‘‘Key factors affecting green pro- pp. 109–115, doi: 10.1109/GreenCom.2012.27.
curement in real estate development: A China study,’’ J. Cleaner Prod., [163] N. Matta, R. Ranhim-Amoud, L. Merghem-Boulahia, and A. Jrad,
vol. 153, pp. 372–383, Jun. 2017, doi: 10.1016/j.jclepro.2016.02.021. ‘‘A wireless sensor network for substation monitoring and control in the
[147] J. K. W. Wong, J. K. S. Chan, and M. J. Wadu, ‘‘Facilitating effective smart grid,’’ in Proc. IEEE Int. Conf. Green Comput. Commun., Besan-
green procurement in construction projects: An empirical study of the con, France, Nov. 2012, pp. 203–209, doi: 10.1109/GreenCom.2012.39.
enablers,’’ J. Cleaner Prod., vol. 135, pp. 859–871, Nov. 2016, doi: [164] X. Lu, Z. Guan, X. Zhou, X. Du, L. Wu, and M. Guizani, ‘‘A secure and
10.1016/j.jclepro.2016.07.001. efficient renewable energy trading scheme based on blockchain in smart
[148] H. Fang, B. Wang, and W. Song, ‘‘Analyzing the interrelationships among grid,’’ in Proc. IEEE 21st Int. Conf. High Perform. Comput. Commun.,
barriers to green procurement in photovoltaic industry: An integrated IEEE 17th Int. Conf. Smart City, IEEE 5th Int. Conf. Data Sci. Syst.
method,’’ J. Cleaner Prod., vol. 249, Mar. 2020, Art. no. 119408, doi: (HPCC/SmartCity/DSS), Zhangjiajie, China, Aug. 2019, pp. 1839–1844,
10.1016/j.jclepro.2019.119408. doi: 10.1109/HPCC/SmartCity/DSS.2019.00253.
[149] J. Liu, Y. Liu, and L. Yang, ‘‘Uncovering the influence mechanism [165] M. A. Zaveri, S. K. Pandey, and J. S. Kumar, ‘‘Collaborative service
between top management support and green procurement: The effect of oriented smart grid using the Internet of Things,’’ in Proc. Int. Conf.
green training,’’ J. Cleaner Prod., vol. 251, Apr. 2020, Art. no. 119674, Commun. Signal Process. (ICCSP), Apr. 2016, pp. 1716–1722, doi:
doi: 10.1016/j.jclepro.2019.119674. 10.1109/ICCSP.2016.7754459.
[150] T. Simcoe and M. W. Toffel, ‘‘Government green procurement spillovers: [166] P. Ravindran, K. R. Das, and A. S. Mohan, ‘‘Flexible demand response
Evidence from municipal building policies in California,’’ J. Envi- in smart grid based automatic generation control,’’ in Proc. Int. Conf.
ron. Econ. Manag., vol. 68, no. 3, pp. 411–434, Nov. 2014, doi: Green Comput. Commun. Electr. Eng. (ICGCCEE), Coimbatore, India,
10.1016/j.jeem.2014.09.001. Mar. 2014, pp. 1–6, doi: 10.1109/ICGCCEE.2014.6922456.
[151] S. H. Khahro, A. H. Memon, N. A. Memon, A. Arsal, and T. H. Ali, [167] Z. Zhou, M. Xie, T. Zhu, W. Xu, P. Yi, Z. Huang, Q. Zhang, and S. Xiao,
‘‘Modeling the factors enhancing the implementation of green procure- ‘‘EEP2P: An energy-efficient and economy-efficient P2P network proto-
ment in the Pakistani construction industry,’’ Sustainability, vol. 13, col,’’ in Proc. Int. Green Comput. Conf., DALLAS, TX, USA, Nov. 2014,
no. 13, p. 7248, Jun. 2021, doi: 10.3390/su13137248. pp. 1–6, doi: 10.1109/IGCC.2014.7039171.

87490 VOLUME 11, 2023

S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

[168] V. Suhendra, Y. Wu, H. Saputra, Z. Zhao, and F. Bao, ‘‘Lightweight key [186] W. Shu, G. Zhang, M.-Y. Wu, and J.-L. Lu, ‘‘A social-network-enabled
management protocols for smart grids,’’ in Proc. IEEE Int. Conf. Internet green transportation system,’’ in Proc. Int. Conf. Connected Vehicles
Things (iThings), IEEE Green Comput. Commun. (GreenCom), IEEE Expo (ICCVE), Las Vegas, NV, USA, Dec. 2013, pp. 425–430, doi:
Cyber, Phys. Social Comput. (CPSCom), IEEE Smart Data (SmartData), 10.1109/ICCVE.2013.6799830.
Dec. 2016, pp. 345–348, doi: 10.1109/iThings-GreenCom-CPSCom- [187] H. S. Hayajneh, M. Lainfiesta, and X. Zhang, ‘‘Three birds one stone:
SmartData.2016.82. A solution to maximize renewable generation, incentivize battery deploy-
[169] A. Alsharif, A. Shafee, M. Nabil, M. Mahmoud, and W. Alasmary, ment, and promote green transportation,’’ in Proc. IEEE Power Energy
‘‘A multi-authority attribute-based signcryption scheme with efficient Soc. Innov. Smart Grid Technol. Conf. (ISGT), Washington, DC, USA,
revocation for smart grid downlink communication,’’ in Proc. Int. Conf. Feb. 2020, pp. 1–5, doi: 10.1109/ISGT45199.2020.9087710.
Internet Things (iThings), IEEE Green Comput. Commun. (Green- [188] J. Asl-Najafi, S. Yaghoubi, and A. Azaron, ‘‘Coordination of a bi-level
Com), IEEE Cyber, Phys. Social Comput. (CPSCom), IEEE Smart closed-loop supply chain considering economic and green transporta-
Data (SmartData), Atlanta, GA, USA, Jul. 2019, pp. 1025–1032, doi: tion modes,’’ Int. J. Ind. Eng. Prod. Res., vol. 29, no. 4, pp. 429–441,
10.1109/iThings/GreenCom/CPSCom/SmartData.2019.00178. 2018.
[170] I. F. Siddiqui, N. M. F. Qureshi, B. S. Chowdhry, and M. A. Uqaili, [189] Z. Jiang, Y. Chen, X. Li, and B. Li, ‘‘A heuristic optimization approach
‘‘Edge-Node-Aware adaptive data processing framework for smart grid,’’ for multi-vehicle and one-cargo green transportation scheduling in ship-
Wireless Pers. Commun., vol. 106, no. 1, pp. 179–189, May 2019, doi: building,’’ Adv. Eng. Informat., vol. 49, Aug. 2021, Art. no. 101306, doi:
10.1007/s11277-019-06264-7. 10.1016/j.aei.2021.101306.
[171] B. Schäfer, M. Matthiae, M. Timme, and D. Witthaut, ‘‘Decentral smart [190] K. L. Soon, J. M.-Y. Lim, and R. Parthiban, ‘‘Coordinated traffic light
grid control,’’ New J. Phys., vol. 17, no. 5, May 2015, Art. no. 059502, control in cooperative green vehicle routing for pheromone-based multi-
doi: 10.1088/1367-2630/17/5/059502. agent systems,’’ Appl. Soft Comput., vol. 81, Aug. 2019, Art. no. 105486,
[172] S. Yu, L. Mu, and B. Ji, ‘‘On green transport and low carbon transport,’’ doi: 10.1016/j.asoc.2019.105486.
in Proc. ICTE. Chengdu, China: American Society of Civil Engineers, [191] A. J. C. Trappey, C. Trappey, C. T. Hsiao, J. J. R. Ou, S. J. Li, and
Jul. 2011, pp. 3061–3066, doi: 10.1061/41184(419)505. K. W. P. Chen, ‘‘An evaluation model for low carbon island policy:
[173] C.-W. Yang and Y.-L. Ho, ‘‘Assessing carbon reduction effects toward The case of Taiwan’s green transportation policy,’’ Energy Policy, vol. 45,
the mode shift of green transportation system: Carbon reduction effects pp. 510–515, Jun. 2012, doi: 10.1016/j.enpol.2012.02.063.
of GREEN TRANSPORT,’’ J. Adv. Transp., vol. 50, no. 5, pp. 669–682, [192] C. Lin, K. L. Choy, G. T. S. Ho, and T. W. Ng, ‘‘A genetic algorithm-
Aug. 2016, doi: 10.1002/atr.1367. based optimization model for supporting green transportation opera-
[174] I. Beniušienė and A. Jankauskienė, ‘‘Green transportation in the Siauliai tions,’’ Exp. Syst. Appl., vol. 41, no. 7, pp. 3284–3296, Jun. 2014, doi:
regional companies,’’ Rural Sustainability Res., vol. 46, no. 341, 10.1016/j.eswa.2013.11.032.
pp. 31–38, Dec. 2021, doi: 10.2478/plua-2021-0015. [193] T.-H. Chang, J.-S. Tseng, T.-H. Hsieh, Y.-T. Hsu, and Y.-C. Lu, ‘‘Green
[175] J. Froehlich, T. Dillahunt, P. Klasnja, J. Mankoff, S. Consolvo, transportation implementation through distance-based road pricing,’’
B. Harrison, and J. A. Landay, ‘‘UbiGreen: Investigating a mobile tool Transp. Res. A, Policy Pract., vol. 111, pp. 53–64, May 2018, doi:
for tracking and supporting green transportation habits,’’ in Proc. SIGCHI 10.1016/j.tra.2018.02.015.
Conf. Human Factors Comput. Syst., Boston, MA, USA, Apr. 2009, [194] Z. Guo, D. Zhang, H. Liu, Z. He, and L. Shi, ‘‘Green trans-
pp. 1043–1052, doi: 10.1145/1518701.1518861. portation scheduling with pickup time and transport mode selec-
[176] S. Smaldone, C. Tonde, V. K. Ananthanarayanan, A. Elgammal, tions using a novel multi-objective memetic optimization approach,’’
and L. Iftode, ‘‘The cyber-physical bike: A step towards safer green Transp. Res. D, Transp. Environ., vol. 60, pp. 137–152, May 2018, doi:
transportation,’’ in Proc. 12th Workshop Mobile Comput. Syst. Appl., 10.1016/j.trd.2016.02.003.
Mar. 2011, pp. 56–61, doi: 10.1145/2184489.2184502. [195] S. Wang, J. Wang, and F. Yang, ‘‘From willingness to action: Do push-
[177] X. Hu, V. C. M. Leung, K. G. Li, E. Kong, H. Zhang, N. S. Surendrakumar, pull-mooring factors matter for shifting to green transportation?’’ Transp.
and P. TalebiFard, ‘‘Social drive: A crowdsourcing-based vehicular social Res. D, Transp. Environ., vol. 79, Feb. 2020, Art. no. 102242, doi:
networking system for green transportation,’’ in Proc. 3rd ACM Int. 10.1016/j.trd.2020.102242.
Symp. Design Anal. Intell. Veh. Netw. Appl., Barcelona, Spain, Nov. 2013, [196] A. A. Shojaie and H. Raoofpanah, ‘‘Solving a two-objective green trans-
pp. 85–92, doi: 10.1145/2512921.2512924. portation problem by using meta-heuristic methods under uncertain fuzzy
[178] D. Ku, J. Kim, Y. Yu, S. Kim, S. Lee, and S. Lee, ‘‘Assessment of eco- approach,’’ J. Intell. Fuzzy Syst., vol. 34, no. 1, pp. 1–10, Jan. 2018, doi:
friendly effects on green transportation demand management,’’ Chem. 10.3233/JIFS-161584.
Eng. Trans., vol. 89, pp. 121–126, Dec. 2021, doi: 10.3303/CET2189021. [197] Z. Wong, A. Chen, C. Shen, and D. Wu, ‘‘Fiscal policy and the
[179] C.-I. Hsu and H.-M. Wang, ‘‘Strategies for green transportation while development of green transportation infrastructure: The case of China’s
preserving mobility and accessibility: A case study of Taipei city,’’ high-speed railways,’’ Econ. Change Restructuring, vol. 55, no. 4,
J. Urban Plann. Dev., vol. 142, no. 1, Mar. 2016, Art. no. 04015008, doi: pp. 2179–2213, Nov. 2022, doi: 10.1007/s10644-021-09381-1.
10.1061/(ASCE)UP.1943-5444.0000286. [198] F. Guo, Q. Liu, D. Liu, and Z. Guo, ‘‘On production and green transporta-
[180] C. Xin, L. Wang, B. Liu, Y.-H. Yuan, and S.-B. Tsai, ‘‘An empirical tion coordination in a sustainable global supply chain,’’ Sustainability,
study for green transportation scheme of municipal solid waste based on vol. 9, no. 11, p. 2071, Nov. 2017, doi: 10.3390/su9112071.
complex data model analysis,’’ Math. Problems Eng., vol. 2021, pp. 1–17, [199] M. Lu, R. Xie, P. Chen, Y. Zou, and J. Tang, ‘‘Green transportation and
Apr. 2021, doi: 10.1155/2021/6614312. logistics performance: An improved composite index,’’ Sustainability,
[181] Q. Wang and H. Sun, ‘‘Traffic structure optimization in historic vol. 11, no. 10, p. 2976, May 2019, doi: 10.3390/su11102976.
districts based on green transportation and sustainable development [200] J. Fu and J. Chen, ‘‘A green transportation planning approach for coal
concept,’’ Adv. Civil Eng., vol. 2019, pp. 1–18, Jan. 2019, doi: heavy-haul railway system by simultaneously optimizing energy con-
10.1155/2019/9196263. sumption and capacity utilization,’’ Sustainability, vol. 13, no. 8, p. 4173,
[182] K. Ghosh and K. S. S. Musti, ‘‘Integration of SLAM with GIS to model Apr. 2021, doi: 10.3390/su13084173.
sustainable urban transportation system: A smart city perspective,’’ in [201] Y. Jin, B. Xing, and P. Jin, ‘‘Towards a benchmark platform for measuring
Proc. 12th Int. Conf. Comput. Intell. Commun. Netw. (CICN), Bhimtal, the energy consumption of database systems,’’ in Proc. Interdiscipl. Res.
India, Sep. 2020, pp. 261–267, doi: 10.1109/CICN49253.2020.9242571. Theory Technol., Nov. 2013, pp. 385–389, doi: 10.14257/astl.2013.29.79.
[183] A. Lv, D. Xin, L. Zhang, L. Shi, and R. Geng, ‘‘Research on creating [202] M. Mohankumar and M. A. Kumar, ‘‘Green database design model in
ecological highway for serving construction of powerful transporta- software development life cycle phase,’’ Indian J. Sci. Technol., vol. 9,
tion country,’’ in Proc. E3S Web Conf., vol. 145, 2020, p. 02046, doi: no. 30, pp. 1–9, Aug. 2016, doi: 10.17485/ijst/2016/v9i30/93118.
10.1051/e3sconf/202014502046. [203] K. K. Singh, P. Dimri, and M. Rawat, ‘‘Green database model for
[184] N. Chowdhury, C. Hossain, M. Longo, and W. Yaïci, ‘‘Optimization of stock market: A case study of Indian stock market,’’ in Proc. 5th Int.
solar energy system for the electric vehicle at University Campus in Conf. Confluence Next Gener. Inf. Technol. Summit (Confluence), Noida,
Dhaka, Bangladesh,’’ Energies, vol. 11, no. 9, p. 2433, Sep. 2018, doi: India, Sep. 2014, pp. 848–853, doi: 10.1109/CONFLUENCE.2014.
10.3390/en11092433. 6949306.
[185] P. Song and C. Zhong, ‘‘Fuzzy comprehensive evaluation research of goal [204] K. K. Singh, P. Dimri, and J. N. Singh, ‘‘Green data base management
attainment of green transportation development policy—Take Dongguan system for the intermediaries of Indian stock market,’’ in Proc. Conf. IT
‘motorcycle ban’ policy as an example,’’ J. Transp. Technol., vol. 5, no. 1, Bus., Ind. Government (CSIBIG), Indore, India, Mar. 2014, pp. 1–5, doi:
pp. 46–54, 2015, doi: 10.4236/jtts.2015.51005. 10.1109/CSIBIG.2014.7056996.

VOLUME 11, 2023 87491

 S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

[205] N. C. Pa, F. Karim, and S. Hassan, ‘‘Dashboard system for [224] M. Kozlowski, R. McConville, R. Santos-Rodriguez, and R. Piechocki,
measuring green software design,’’ in Proc. 3rd Int. Conf. ‘‘Energy efficiency in reinforcement learning for wireless sensor net-
Sci. Inf. Technol. (ICSITech), Oct. 2017, pp. 325–329, doi: works,’’ 2018, arXiv:1812.02538.
10.1109/ICSITech.2017.8257133. [225] Y. Wang, R. Fan, L. Shen, and M. Jin, ‘‘Decisions and coordination
[206] Z. G. Jiang, H. Zhang, and M. Xiao, ‘‘Web-based process database sup- of green e-commerce supply chain considering green manufacturer’s
port system for green manufacturing,’’ Appl. Mech. Mater., vols. 10–12, fairness concerns,’’ Int. J. Prod. Res., vol. 58, no. 24, pp. 7471–7489,
pp. 94–98, Dec. 2007, doi: 10.4028/www.scientific.net/AMM. Dec. 2020, doi: 10.1080/00207543.2020.1765040.
10-12.94. [226] N. Singh and O. Sahu, ‘‘Feasibility assessment for e-commerce: A data
[207] M. Feng and J. Staum, ‘‘Green simulation with database Monte Carlo,’’ collection from developing country (Ethiopia),’’ MethodsX, vol. 9,
ACM Trans. Model. Comput. Simul., vol. 31, no. 1, pp. 1–26, Jan. 2021, Jan. 2022, Art. no. 101639, doi: 10.1016/j.mex.2022.101639.
doi: 10.1145/3429336. [227] P. Dutta, P. Suryawanshi, P. Gujarathi, and A. Dutta, ‘‘Managing risk for
[208] J. Schneider, M. Basalla, and S. Seidel, ‘‘Principles of green data min- e-commerce supply chains: An empirical study,’’ IFAC-PapersOnLine,
ing,’’ in Proc. Annu. Hawaii Int. Conf. Syst. Sci., 2019, pp. 1–10, doi: vol. 52, no. 13, pp. 349–354, 2019, doi: 10.1016/j.ifacol.2019.11.143.
10.24251/HICSS.2019.250. [228] D. H. Keserwani and D. H. Rastogi, ‘‘Green e-commerce: An assessment
[209] R. Alabdulrahman, H. Viktor, and E. Paquet, ‘‘An active learning of market readiness & movement to a new competitive landscape,’’
approach for ensemble-based data stream mining,’’ in Proc. 8th Int. J. Contemp. Issues Bus. Government, vol. 27, no. 3, pp. 1437–1441,
Joint Conf. Knowl. Discovery, Knowl. Eng. Knowl. Manag., 2016, Apr. 2021, doi: 10.47750/cibg.2021.27.03.193.
pp. 275–282, doi: 10.5220/0006047402750282. [229] Y. Wang, Z. Yu, L. Shen, R. Fan, and R. Tang, ‘‘Decisions and coor-
[210] K. Balaskas, G. Zervakis, K. Siozios, M. B. Tahoori, and J. Henkel, dination in e-commerce supply chain under logistics outsourcing and
‘‘Approximate decision trees for machine learning classification on tiny altruistic preferences,’’ Mathematics, vol. 9, no. 3, p. 253, Jan. 2021, doi:
printed circuits,’’ 2022, arXiv:2203.08011. 10.3390/math9030253.
[211] A. Freire, C. Macdonald, N. Tonellotto, I. Ounis, and F. Cacheda, [230] S. Buniamin, N. Ahmad, F. H. A. Rauf, N. H. Johari, and A. A. Rashid,
‘‘A self-adapting latency/power tradeoff model for replicated search ‘‘Green government procurement practices (GGP) in Malaysian public
engines,’’ in Proc. 7th ACM Int. Conf. Web Search Data Mining, enterprises,’’ Proc. Econ. Finance, vol. 35, pp. 27–34, Jan. 2016, doi:
New York, NY, USA, Feb. 2014, pp. 13–22, doi: 10.1145/2556195. 10.1016/S2212-5671(16)00006-X.
2556246. [231] N. Khalil, A. A. M. Bohari, S. M. Shamsudin, A. F. A. Rashid, and
[212] M. Catena, C. Macdonald, and N. Tonellotto, ‘‘Load-sensitive CPU power H. N. Husin, ‘‘Key approaches of life-cycle cost in green government
management for web search engines,’’ in Proc. 38th Int. ACM SIGIR Conf. procurement (GGP) for green projects,’’ Planning Malaysia, vol. 19,
Res. Develop. Inf. Retr., Santiago, Chile, Aug. 2015, pp. 751–754, doi: pp. 1–12, Jul. 2021, doi: 10.21837/PM.v19i16.949.
10.1145/2766462.2767809. [232] L. Liu and Q. Peng, ‘‘Evolutionary game analysis of enterprise green
[213] R. Blanco, M. Catena, and N. Tonellotto, ‘‘Exploiting green energy to innovation and green financing in platform supply chain,’’ Sustainability,
reduce the operational costs of multi-center web search engines,’’ in vol. 14, no. 13, p. 7807, Jun. 2022, doi: 10.3390/su14137807.
Proc. 25th Int. Conf. World Wide Web, Apr. 2016, pp. 1237–1247, doi: [233] X. Hou, Y. Ma, Y. Wu, and W. Wang, ‘‘Implementing green education of
10.1145/2872427.2883021. urban families: An action research project in Beijing, China,’’ Action Res.,
[214] Z. Al-Zanbouri and C. Ding, ‘‘Recommending energy-efficient data vol. 18, no. 1, pp. 19–47, Mar. 2020, doi: 10.1177/1476750319889385.
mining services with data as contextual factors,’’ in Proc. IEEE Int. [234] T. A. Shittu, A. I. Gambari, and A. O. Thomas, ‘‘Survey of education,
Conf. Services Comput. (SCC), Milan, Italy, Jul. 2019, pp. 14–18, doi: engineering, and information technology students knowledge of green
10.1109/SCC.2019.00016. computing in Nigerian University,’’ J. Educ. Learn. (EduLearn), vol. 10,
[215] Z. Al-Zanbouri and C. Ding, ‘‘Data-aware web service recommender no. 1, pp. 70–77, Feb. 2016, doi: 10.11591/edulearn.v10i1.3185.
system for energy-efficient data mining services,’’ in Proc. IEEE 11th [235] B. A. Birchi, ‘‘Assessing university students’ attitude toward green com-
Conf. Service-Oriented Comput. Appl., Nov. 2018, pp. 57–64, doi: puting practices,’’ in Proc. Int. Conf. Future Comput. Technol., Singapore,
10.1109/SOCA.2018.00015. Mar. 2015, pp. 27–33, doi: 10.17758/UR.U0315210.
[216] A. Deylamsalehi, D. A. P. Davis, P. Afsharlar, M. Bahrami, W.-P. Chen, [236] A. Ullah, N. Mohd Nawi, A. Shahzad, S. N. Khan, and M. Aamir, ‘‘An e-
and V. M. Vokkarane, ‘‘Using machine learning to balance energy cost learning system in Malaysia based on green computing and energy level,’’
and emissions in optical networks,’’ J. Opt. Commun. Netw., vol. 10, JOIV, Int. J. Informat. Visualizat., vol. 1, nos. 4–2, p. 184, Nov. 2017, doi:
no. 10, pp. 72–83, Oct. 2018, doi: 10.1364/JOCN.10.000D72. 10.30630/joiv.1.4-2.63.
[217] A. Candelieri, R. Perego, and F. Archetti, ‘‘Green machine learning via [237] T. B. Tunku Ahmad, A. Bello, and M. S. Nordin, ‘‘Exploring Malaysian
augmented Gaussian processes and multi-information source optimiza- University students’ awareness of green computing,’’ GSTF Int. J. Educ.,
tion,’’ Soft Comput., vol. 25, no. 19, pp. 12591–12603, Oct. 2021, doi: vol. 1, no. 2, pp. 1–11, Nov. 2013, doi: 10.5176/2345-7163_1.2.34.
10.1007/s00500-021-05684-7. [238] K. Palanivel and S. Kuppuswami, ‘‘A cloud-oriented green computing
[218] A. Candelieri, A. Ponti, and F. Archetti, ‘‘Fair and green hyperparam- architecture for e-learning applications,’’ Int. J. Recent Innov. Trends
eter optimization via multi-objective and multiple information source Comput. Commun., vol. 2, no. 11, pp. 3775–3783, Nov. 2014. [Online].
Bayesian optimization,’’ 2022, arXiv:2205.08835. Available: https://ijritcc.org/index.php/ijritcc/article/view/3556
[219] H. Lu, X. Liu, S. Zhou, R. Turner, P. Bai, R. E. Koot, M. Chasmai, [239] A. Ibrahim, ‘‘Understanding the factors affecting the adoption of green
L. Schobs, and H. Xu, ‘‘PyKale: Knowledge-aware machine learning computing in the Gulf Universities,’’ Int. J. Adv. Comput. Sci. Appl., vol. 9,
from multiple sources in Python,’’ in Proc. 31st ACM Int. Conf. Inf. no. 3, pp. 304–3011, 2018, doi: 10.14569/IJACSA.2018.090342.
Knowl. Manag., Atlanta, GA, USA, Oct. 2022, pp. 4274–4278, doi: [240] S. Hanief, L. G. S. Kartika, N. L. P. Srinadi, and K. R. Y. Negara, ‘‘A pro-
10.1145/3511808.3557676. posed model of green computing adoption in Indonesian higher educa-
[220] S. Savazzi, V. Rampa, S. Kianoush, and M. Bennis, ‘‘An energy and tion,’’ in Proc. 6th Int. Conf. Cyber IT Service Manag. (CITSM), Parapat,
carbon footprint analysis of distributed and federated learning,’’ IEEE Indonesia, Aug. 2018, pp. 1–6, doi: 10.1109/CITSM.2018.8674275.
Trans. Green Commun. Netw., vol. 7, no. 1, pp. 248–264, Mar. 2023, doi: [241] S. Agrawal, R. Biswas, and A. Nath, ‘‘Virtual desktop infrastructure
10.1109/TGCN.2022.3186439. in higher education institution: Energy efficiency as an application of
[221] S. Busaeed, R. Mehmood, I. Katib, and J. M. Corchado, ‘‘LidSonic green computing,’’ in Proc. 4th Int. Conf. Commun. Syst. Netw. Technol.,
for visually impaired: Green machine learning-based assistive smart Bhopal, India, Apr. 2014, pp. 601–605, doi: 10.1109/CSNT.2014.250.
glasses with smart app and Arduino,’’ Electronics, vol. 11, no. 7, p. 1076, [242] M. Srinivasan, ‘‘GreenEduComp: Low cost green computing system
Mar. 2022, doi: 10.3390/electronics11071076. for education in Rural India: A scheme for sustainable development
[222] M. Rehmani, S. Aslam, S. Tito, S. Soltic, P. Nieuwoudt, N. Pandey, through education,’’ in Proc. IEEE Global Humanitarian Technol.
and M. Ahmed, ‘‘Power profile and thresholding assisted multi-label Conf. (GHTC), San Jose, CA, USA, Oct. 2013, pp. 102–107, doi:
NILM classification,’’ Energies, vol. 14, no. 22, p. 7609, Nov. 2021, doi: 10.1109/GHTC.2013.6713663.
10.3390/en14227609. [243] W. Zhao and Y. Zou, ‘‘Green university initiatives in China: A case of
[223] A. Candelieri, F. Archetti, A. Ponti, and R. Perego, ‘‘Energy efficient Tsinghua University,’’ Int. J. Sustainability Higher Educ., vol. 16, no. 4,
hyperparameters tuning through augmented Gaussian processes and pp. 491–506, Jul. 2015, doi: 10.1108/IJSHE-02-2014-0021.
multi-information source optimization,’’ in Proc. 7th Int. Conf. Soft Com- [244] Y. Geng, K. Liu, B. Xue, and T. Fujita, ‘‘Creating a ‘green university’
put. Mach. Intell. (ISCMI), Stockholm, Sweden, Nov. 2020, pp. 34–38, in China: A case of Shenyang University,’’ J. Cleaner Prod., vol. 61,
doi: 10.1109/ISCMI51676.2020.9311599. pp. 13–19, Dec. 2013, doi: 10.1016/j.jclepro.2012.07.013.

87492 VOLUME 11, 2023

S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

[245] Y. Cai, ‘‘Integrating sustainability into undergraduate computing educa- ARPA SAHA received the degree in computer sci-
tion,’’ in Proc. 41st ACM Tech. Symp. Comput. Sci. Educ., Mar. 2010, ence and engineering (CSE) from Daffodil Inter-
pp. 524–528, doi: 10.1145/1734263.1734439. national University, Dhaka, Bangladesh. She has
[246] V. Protonotarios, M. Ungur, H. Ebner, and N. Manouselis, ‘‘Green gained the required skills and expertise to make
education using open educational resources (OER): Setting up a her research successful. Her research interests
green OER repository,’’ in Proc. 9th Multidisciplinary Symp. Design include machine learning, deep learning, artifi-
Eval. Digit. Content Educ. (SPDECE). Alicante, Spain: Univer-
cial intelligence, image processing, data mining,
sidad de Alicante, Jun. 2012, pp. 135–147. [Online]. Available:
natural language processing, health informatics,
http://hdl.handle.net/10045/35609
[247] W. Wang, K. Li, Y. Liu, J. Lian, and S. Hong, ‘‘A system dynamics model and computer vision. She has demonstrated her
analysis for policy impacts on green agriculture development: A case passion for these fields through her impressive
of the Sichuan Tibetan area,’’ J. Cleaner Prod., vol. 371, Oct. 2022, research accomplishments, including the publication of research articles in
Art. no. 133562, doi: 10.1016/j.jclepro.2022.133562. the respected Q1 journals.
[248] J. Qiu, L. Yang, X. Chen, and X. Wu, ‘‘Optimal configuration
of a negative carbon emission energy system for green agricul-
ture,’’ IFAC-PapersOnLine, vol. 55, no. 9, pp. 501–506, 2022, doi:
10.1016/j.ifacol.2022.07.087.
[249] J. Zhang, S. Bai, and N. Xu, ‘‘Optimal pricing of green agriculture
products in a marketing-mix program,’’ Int. J. Sustain. Develop. Planning,
vol. 15, no. 7, pp. 1001–1006, Nov. 2020, doi: 10.18280/ijsdp.150704.
[250] M. A. Hossain, M. S. Hassan, S. Ahmmed, and M. S. Islam, MOHAMMAD SHAMSUL AREFIN (Senior
‘‘Solar pump irrigation system for green agriculture,’’ Agric. Eng. Int. Member, IEEE) received the Doctor of Engi-
CIGR J., vol. 16, no. 4, pp. 1–15, Dec. 2014. [Online]. Available: neering degree in information engineering from
https://cigrjournal.org/index.php/Ejounral/article/view/2836 Hiroshima University, Japan, with the support of
[251] M. Wang and K. Zhang, ‘‘Improving agricultural green supply chain man- the scholarship of MEXT, Japan. He is currently
agement by a novel integrated fuzzy-delphi and grey-WINGS model,’’ the Dean of the ECE Faculty, Chittagong Uni-
Agriculture, vol. 12, no. 10, p. 1512, Sep. 2022, doi: 10.3390/agricul- versity of Engineering and Technology (CUET),
ture12101512.
Chattogram, Bangladesh. Previously, he has been
[252] S. Amaruzaman, B. Leimona, M. van Noordwijk, and B. Lusiana,
‘‘Discourses on the performance gap of agriculture in a green econ- affiliated with the Department of Computer Sci-
omy: A Q-methodology study in Indonesia,’’ Int. J. Biodiversity Sci., ence and Engineering (CSE), Daffodil Interna-
Ecosystem Services Manag., vol. 13, no. 1, pp. 233–247, Jan. 2017, doi: tional University, Dhaka, Bangladesh, and a Lien Professor with CUET,
10.1080/21513732.2017.1331264. where he was the Head of the Department of Computer Science and Engi-
[253] Y. Lu, X. Li, J. Zhong, and Y. Xiong, ‘‘Research on the innovation of neering. As a part of his doctoral research, he was with the IBM Yamato
strategic business model in green agricultural products based on Internet Software Laboratory, Japan. He has published more than 110 refereed pub-
of Things (IoT),’’ in Proc. 2nd Int. Conf. E-Business Inf. Syst. Secur., lications in international journals, book series, and conference proceedings.
May 2010, pp. 1–3, doi: 10.1109/EBISS.2010.5473338. His research interests include privacy-preserving data publishing and mining,
[254] H. Wang, S. Zhong, J. Guo, and Y. Fu, ‘‘Factors affecting green agri- distributed and cloud computing, big data management, multilingual data
cultural production financing behavior in Heilongjiang family farms: management, semantic web, object-oriented system development, and IT for
A structural equation modeling approach,’’ Frontiers Psychol., vol. 12, agriculture and the environment. He is a member of ACM and a fellow of
Sep. 2021, Art. no. 692140, doi: 10.3389/fpsyg.2021.692140.
IEB and BCS. He visited Japan, Indonesia, Malaysia, Bhutan, Singapore,
[255] H. A. M. Tran, H. Q. T. Ngo, T. P. Nguyen, and H. Nguyen, ‘‘Design
of green agriculture system using Internet of Things and image pro- South Korea, Egypt, India, Saudi Arabia, and China for different professional
cessing techniques,’’ in Proc. 4th Int. Conf. Green Technol. Sustain. and social activities. He is also the Organizing Chair of BIM 2021, the
Develop. (GTSD), Ho Chi Minh City, Vietnam, Nov. 2018, pp. 28–32, TPC Chair of ECCE 2017, the Organizing Co-Chair of ECCE 2019, and
doi: 10.1109/GTSD.2018.8595663. the Organizing Chair of BDML 2020.
[256] X. Long, Y. Luo, C. Wu, and J. Zhang, ‘‘The influencing factors of CO2
emission intensity of Chinese agriculture from 1997 to 2014,’’ Environ.
Sci. Pollut. Res., vol. 25, no. 13, pp. 13093–13101, May 2018, doi:
10.1007/s11356-018-1549-6.
[257] Y. Zhu and J. Chen, ‘‘Small-scale farmers’ preference heterogeneity for
green agriculture policy incentives identified by choice experiment,’’ Sus-
tainability, vol. 14, no. 10, p. 5770, May 2022, doi: 10.3390/su14105770.
[258] Y. Xia, H. Long, Z. Li, and J. Wang, ‘‘Farmers’ credit risk assessment TOUHID BHUIYAN received the Ph.D. degree
based on sustainable supply chain finance for green agriculture,’’ Sustain- in information security focusing on trust man-
ability, vol. 14, no. 19, p. 12836, Oct. 2022, doi: 10.3390/su141912836. agement for intelligent recommendations from
the Queensland University of Technology (QUT),
Australia. He is currently the Head of the Depart-
ment of Computer Science and Engineering, Daf-
fodil International University (DIU), Bangladesh.
He is also a certified ethical hacker. He has
received the Cyber Security: Cyber Risk and
SHOWMICK GUHA PAUL received the degree Resilience Certificate from the University of
in computer science and engineering (CSE) Oxford. He was a recipient of the Australian Postgraduate Award (APA) and
from Daffodil International University, Dhaka, the Deputy Vice-Chancellor’s Initiative Scholarship from QUT. He was the
Bangladesh. He is also actively involved in Director of the Cyber Security Centre, DIU, where he was also the Head and a
research in machine learning and data science. Professor with the Software Engineering Department. His research interests
He is also a member of several research and include cyber security, intelligent recommendations, social networks, trust
programming organizations. He has obtained aca- management, big data analytics, e-health, and e-learning. Before joining
demic and research experience by working as a with DIU, he worked for several institutions, including Monash University,
research assistant and has published Q1 journals The University of Western Australia, QUT, University of Western Sydney,
articles. His research interests include machine and Central College Sydney. He has more than 114 research publications
learning, deep learning, health informatics, artificial intelligence, computer in renowned national and international journals, books, and conference
vision, and green computing and their application on various field. proceedings.

VOLUME 11, 2023 87493

 S. G. Paul et al.: Comprehensive Review of Green Computing: Past, Present, and Future Research

AL AMIN BISWAS received the Bachelor of NAIF M. ALOTAIBI received the Ph.D. degree
Science and Master of Science degrees from the in statistics from Salford University, Manchester,
Department of Computer Science and Engineer- U.K., in 2018. He is currently an Associate Pro-
ing, Jahangirnagar University, Dhaka, Bangladesh. fessor with the Faculty of Science, School of
He is currently a Lecturer with the Computer Mathematics and Statistics, Imam Mohammad
Science and Engineering Department, Banga- Ibn Saud Islamic University (IMSIU), Riyadh,
bandhu Sheikh Mujibur Rahman University Saudi Arabia. He has already published many
(BSMRU), Kishoreganj, Bangladesh. Prior to join- research articles in different international journals.
ing BSMRU, he was a Lecturer (Senior Scale) with His research interests include applied statistics,
the Computer Science and Engineering Depart- mathematical statistics, reliability and mainte-
ment, Daffodil International University (DIU), Dhaka. His research interests nance, statistical modeling, R programming, multivariate analysis, and the
include machine learning, deep learning, data mining, and computer vision. applications of statistics in sports.
He has more than 30 research papers published in international conferences
and journals. He has served as the Session Chair for the International Confer-
ence on Computer Communication and Informatics (ICCCI 2021). He also
served as the Co-Convener for the International Conference on Machine
Intelligence and Data Science Application (MIDAS 2021). He reviewed
SALEM A. ALYAMI (Member, IEEE) received the
research papers for many international IEEE, Springer, ACM conferences,
Ph.D. degree in bio-statistics from Monash Uni-
and Elsevier journals. He received the Best Paper Award twice for his
versity, Australia, in 2017. He has been an Assis-
Outstanding Research Works from two springer conferences.
tant Professor with the School of Mathematics and
Statistics, IMAMU, Riyadh, Saudi Arabia, since
2017, contributing/leading several grants in bio-
statistics projects. Recently, he has been appointed
as the Dean of the Deanship of Scientific Research,
AHMED WASIF REZA (Member, IEEE) received
IMAMU. His research interests include Bayesian
the B.Sc. degree (Hons.) in computer science and
networks, neural networks, Bayesian statistics,
engineering from Khulna University, Bangladesh,
MCMC methods, and the applications of statistics in biology and medicine.
the Master of Engineering Science (M.Eng.Sc.)
degree from Multimedia University, Malaysia, and
the Doctor of Philosophy (Ph.D.) degree from the
University of Malaya, Malaysia. He is currently a
Professor with the Department of Computer Sci-
ence and Engineering (CSE) and the additional MOHAMMAD ALI MONI received the Ph.D.
Director of the Institutional Quality Assurance degree in artificial intelligence and digital health
Cell (IQAC), East West University, Bangladesh. He was also appointed as data science from the University of Cambridge,
the Chairperson of the CSE Department. Previously, he was attached with the U.K. He worked in different world top universities,
Department of Electrical Engineering, Faculty of Engineering, University of including Oxford University, Cambridge Univer-
Malaya, Malaysia, for almost eight years. He is also serving as a member sity, Sydney University, UNSW Sydney, and The
for the Evaluation Team (ET) for Accreditation of different engineering University of Queensland. He is currently a pro-
programs of various universities, appointed by the Board of Accreditation fessor in artificial intelligence and data science.
for Engineering and Technical Education (BAETE), Bangladesh. He also He continued to focus on the same teaching and
has vast experience in supervising Ph.D., master’s, and undergraduate stu- research areas during the last 15 years and has
dents. He has been working in the field of radio frequency identification led and managed significant research programs, such as using big health
(RFID), wireless communications, biomedical image processing, bioimag- data to develop machine learning, deep learning and translational data sci-
ing, bioinformatics, optimization, artificial intelligence/machine learning, ence models, and software tools to aid diagnosis and prediction of disease
deep learning, the robotics and Internet of Things, green computing/IT outcomes, particularly for hard-to-manage complex and chronic diseases.
and sustainable development, brain–computer interface (BCI), biomedical He has published more than 300 top-tire journal articles and received more
signal processing, and electromagnetics, for almost 18 years, both in indus- than 12300 citations with H-index 51. His research interests include devel-
trial exposure and academically research valued work. He has authored or oping statistical models, machine learning and deep learning algorithms,
coauthored several Science Citation Index (SCI)/Web of Science (WoS) and software tools utilizing multimodal data, especially medical images,
and Scopus-indexed journals and conference papers (more than 200 papers; neuroimaging, EEG, ECG, bioinformatics, and secondary usage of routinely
H-index: 23; and more than citations: 2100). He is a professional member of collected data for different conditions.
the IEEE Computer Society.

87494 VOLUME 11, 2023