Classification of big data access control models : comparative study

Azhar Abdulmughni
Department of IT , Faculty of Computer and information technology, University of Sana'a
December, 2023
Abstract ___ In the era of big data, the rapid generation of delivery to authorized users to prevent denial of service.
large amounts of data from various sources poses a challenge [2] [3]
for access control. To address the needs of end users, Cloud computing is provided by specific providers
mechanisms for access control must be established to share like Amazon, Google, Salesforce.com through the
confidential information with authorized users. Currently, the internet. The front end is a new interface service, while
main concern in cloud computing is the security of access storage and maintenance are based on the terms of service
control, with ongoing research and implementation focused on providers. The front end and back end of cloud computing
ensuring secure and reliable file transfers. Existing solutions can be accessed globally through internet connection.
based on role, attribute, identification, and hierarchical access Data in the cloud is viewable and accessible through an
controls are insufficient to instill trust in cloud computing interface, with protection provided for confidential data.
servers. Therefore, future efforts should aim to implement a Access control is based on users and data, categorized into
trusted model for securing stored files. The emergence of public, private, and hybrid data. Private data includes
access control technology adds a barrier to the access of data confidential information and requires access control,
resources, avoids some illegal users' access to resources, and while public data does not. Hybrid data may contain a
reduces the problem of resource leakage to a certain extent. combination of data types, necessitating security
This paper classifies some access control technologies in the measures [3]
field of big data access control, including discretionary access The term Big Data means analyzing large amounts of
control technology, role-based access control technology, data with different formats, allowing for fast predictions
attribution-based access control technology, blockchain-based with low uncertainties and high value. [3] there are three
access control technology, etc. Then it summarizes the characteristics that define the nature of big data witch
application characteristics of access control technology in the called 3Vs: Volume, Variety, and Velocity.
field of big data, and finally looks forward to the development
-Volume referred to the vast amount of information
prospect of access control.
that is generated and collected from various sources such
Keywords: access control, Big data, Cloud computing, as social media and connected devices. For instance ,
Security social media platforms like YouTube and Twitter have an
I. Introduction extensive user base that produces a large amount of data
every second.
Cloud computing is a type of computing framework
that offers on-demand services using shared - Variety, encompasses the diverse types of data being
computational resources. These services have made our gathered, including text, sound, film, and sensor data.
lives more convenient, as they can convert resource- This data can be either structured or unstructured, posing
constrained mobile devices into powerful computers. challenges in terms of processing and analysis.
However, while cloud services have many benefits such -velocity denotes the speed at which data is being
as scalability and cost savings, there are also drawbacks generated and collected, making real-time processing and
in terms of centralization and security. It is important to analysis a complex task. This is especially significant for
address these security concerns, particularly unauthorized data used in time-sensitive decision-making processes
access to sensitive data, in order to fully realize the such as financial transactions and traffic management. [4]
potential of cloud computing.
Access control technology aims to prevent
Security for access control is the main concern in unauthorized users from intruding or destroying data,
cloud. Latest research and implementations in Cloud limiting access to protect user data. The main functions of
focus on secure file transfers. Existing solutions employ access control security are preventing unauthorized access
role, attribute, identification, and hierarchical access and allowing legal users to access and operate data. Data
controls. Encryption can prevent unauthorized access to access control in big data and cloud computing prevents
sensitive data. Challenges exist in practical illegal access, controls user access authority, and protects
implementation. Differentiated access is necessary based data integrity and confidentiality. [5]
on user roles. Data access control has evolved over the
past 30 years. Techniques like fine-grained access control first part of this paper a literature review of related
offer flexibility. Research is needed to address researches in the field of access control technologies with
vulnerability gaps. This paper discusses cloud storage and big data and cloud computing to illustrate the most
access control schemes. Three emerging fine-grained common models that been used and the advancing of
access control techniques are reviewed... [2] these models to overcome the increasing challenges of big
data security ,
Cloud service providers use different access control
models to achieve several objectives, including In the second part we discuss the major issues and
maintaining data confidentiality, integrity, and privacy, challenges that are facing the security of big data in cloud
preventing unauthorized access, and ensuring correct data computing environment including Access control as a
challenge , the third part introduces the platform and

XXX-X-XXXX-XXXX-X/XX/$XX.00 ©20XX IEEE

techniques which are used in processing and allowing the methodology concentrates on delineating and overseeing
access control methods in big data clusters and a brief tasks within an establishment, aiding in the determination
explaining of how to implement access control in Hadoop of user access rights according to their designated tasks.
ecosystem ,the fourth part illustrates classifying of the Through the amalgamation of RBAC and task
most common access control models to be implemented methodology, the proposed model offers a more detailed
in cloud big data applications , The fourth part presents a and adaptable access control system for big data
classification of the most important access control models environments. The model is crafted to tackle the security
used with big data applications in cloud computing and a obstacles linked with the escalating speed and intricacy of
comparison between these models according to several big data, guaranteeing the protection of confidential data
standards and requirements. Last part concludes the from both governmental and non-governmental entities.
results of the study and best practices.
This paper [7] presents a new ABE scheme called
II. Literature review attribute-based encryption with attribute hierarchies and
provide comparison operations between RBAC and
According to the characteristics of big data and cloud ABAC .
computing most of the literature discussed mainly data
stream analysis and Internet of Things applications, by The paper [8] provides a comprehensive analysis of
analyzing related access control enforcement techniques. access control technologies, including RBAC, ABAC,
Researchers in domestic circles have proposed various FGAC, PPAC, SMPC, HE, PRE, ABE, and key
access control methods for big data and cloud computing management techniques, in the context of secure and
environments based on their unique requirements. These private big data analytics in the cloud environment.
methods largely involve modifying and broadening discusses the benefits of RBAC, such as improved
traditional access control technology. Although research security, sscalability, and ease of administration, making
on access control for Big Data platforms is advancing, a it a suitable access control model for big data analytics in
unifying access control framework that combines the cloud.
generality and efficiency has not yet been proposed due to Komar et al [9] presents a framework for access
the complex nature of the managed data, resulting in ad- control in big data as service (BDaaS )using blockchain
hoc solutions for specific platforms and limited and modified CP-ABE algorithms. It explores the use of
applicability of language-centric approaches. [3] [6] ABE algorithms, specifically CP-ABE, for access control.
paper [6] Introducing big data cloud computing and The research paper develops the access control algorithm
access control technology through user roles and entity from CP-ABE and applies it on the blockchain platform.
attributes. And provide a comparison of existing access The framework utilizes smart contracts on the blockchain
control technology with user roles and entity attributes in for managing access information, ensuring a
the context of big data cloud computing. Proposing an decentralized and secure access control environment. The
access control model based on big data cloud computing. use of CP-ABE and blockchain in the framework
Explanation of various aspects of big data cloud improves security and data protection in BDaaS. The
computing access control technology, But The paper does paper contributes to access control field by offering a
not provide a comprehensive evaluation or comparison of specific approach using blockchain and CP-ABE for
the proposed access control model with existing access secure data access in BDaaS platforms.
control technologies in terms of performance, scalability, Table (1) summarizes the literature review of this
or security. paper
Tal .et al [5] proposed framework to implement
attribute-based access control (ABAC) In the processing III. big data security
of big data with various sensitivities. The authors have big Data has evolved into a form of production factor
conducted an analysis and implemented a prototype that exerts a significant influence. The advancement of
application of ABAC for processing large datasets in swift processing and analytical technologies has enabled
Amazon Web Services. They utilized open-source the swift extraction of valuable information to serve as a
versions of Apache Hadoop, Ranger, and Atlas. In their basis for decision-making. While Big Data presents
study, they have rigorously analyzed cybersecurity opportunities for both businesses and society through the
aspects related to the implementation of ABAC policies exploration of diverse and extensive data sets, the issue of
in Hadoop. This includes the development of a synthetic information security emerges as a crucial aspect.
dataset containing information at different sensitivity Insufficient security measures and encryption solutions
levels, reflecting healthcare and interconnected social could potentially transform Big Data into a major
media data realistically. The outcome of their work is a challenge. A key obstacle associated with Big Data
framework designed to enhance the security of big data, pertains to the storage and processing of vast amounts of
effectively addressing significant cybersecurity issues. information. Many organizations employ various
The paper [4] presents a model that integrates Role- technologies such as NoSQL, Hadoop, Spark, as well as
based access control (RBAC) with task methodology for other analytical tools, artificial intelligence, and Machine
the management of big data security. This model is Learning to uncover the necessary insights. Security
intended to improve the authorization capabilities of the remains a paramount concern within the realm of Big
Hadoop ecosystem, particularly the Apache Ranger and Data. Some enterprises implement additional security
Apache Sentry frameworks. RBAC is a commonly measures like identity and access control, encryption, and
utilized access control strategy that assigns roles to users data segregation to enhance the security posture. The
and allocates permissions based on these roles. The task preservation of information confidentiality revolves
 around safeguarding data from unintended, illicit, or identity theft, account and system compromise, legal or
unauthorized access, disclosure, or theft. Data reputational harm, and other severe repercussions. [7]
characterized by high confidentiality requirements is
deemed as confidential and must be safeguarded to avert

Reference AC model Contribution of the study

K. Vijayalakshmi 2020 [2] Multi models Analyzed different types of Access control
And provide a comparison
Tal et al 2023 [5] Attribute-based access The authors analyze and implement a prototype
control (ABAC) application of ABAC to large dataset processing in
Amazon Web Services, using open-source versions of
Apache Hadoop, Ranger, and Atlas
Pandey .et al2023 [4] collaboration of Role- introduce a model for new access control with the
based Access Control with collaboration of Role-based Access Control with task
task methodology methodology. And to illustrate the framework, they
have a use case in IoT
Wang et al 2023 [6] attribute encryption The paper proposes an access control model based
access control and multiuser on big data cloud computing, based on the basic
access control RBAC and ABAC models,
Oussama Mounnan et al 2019 [8] Blockchain based Access control the paper contributes to enhancing security, privacy, and
decentralized access control in Big Data environments
through the use of Blockchain technology.
Yan zhu et al 2015 [7] attribute encryption access Presents a new ABE scheme called attribute-based
control encryption with attribute hierarchies (ABE-AH) that
efficiently implements comparison operations between
attribute values on a poset derived from an attribute
lattice.
Arun et al 2023 [8] RBAC analysis of access control technologies, including RBAC,
ABAC, FGAC, PPAC, SMPC, HE, PRE, ABE, and key
management techniques,
Komar et al 2022 [9] Blockchan Santosh Kumar Sharma1, Dr Ajay Pratap2* and Dr
ABE Harsh Dev3
Table 1summary of litratrure review
A- Security and Privacy requirements in Big Data: allocation and memory management algorithms. Big data
challenges are prominent in industries like telecoms, web
1-Confidentiality: that limits access to privileged marketing, retail, and government activities, leading to a
individuals only. Breaching confidentiality occurs competitive advantage for companies capable of
through intentional or unintentional disclosure. Various analyzing such data. Data mining techniques can aid in
mechanisms are used to protect information malware detection in cloud environments. Security
confidentiality challenges in cloud computing are divided into network,
2-. Integrity ensures data are whole and uncorrupted, user authentication, data, and general issues. Network
facing risks of corruption, damage, or disruption. challenges involve protocols, security, distributed nodes,
data, and communication. User authentication challenges
include encryption, authentication methods,
3- Availability ensures authorized users can access administrative rights, application authentication, and
information in a usable format without obstruction. It does logging. Data challenges focus on integrity, availability,
not guarantee access to all users, but availability when protection, and distribution. General challenges
needed by authorized individuals. [7] encompass traditional security tools and technology
utilization. [5]
B- Big data Security challenges:
The support provided by Big Data platforms for
Cloud computing faces security challenges due to storage and analysis of large and diverse datasets is
various technologies involved like networks, databases, unmatched by traditional data management systems.
virtualization, and more, making security issues These new systems offer flexibility, efficacy, and superior
applicable. Network security is crucial for cloud systems, performance and scalability. However, Big Data systems
and virtualization in cloud computing raises security lack the same level of data protection features. While
concerns, especially in mapping virtual machines securely traditional systems have various data protection
to physical ones. Data security involves encryption, frameworks, Big Data platforms mostly have basic access
enforcing data sharing policies, and securing resource control mechanisms. This poses a serious threat as there
is unrestricted access to large volumes of data from processed in parallel using Map and Reduce functions.
multiple sources, including sensitive and private This approach can handle unstructured and semi--
information. The advanced analysis capabilities of Big structured data. [3] mplementing access control within a
Data platforms can be exploited to derive correlations MapReduce setting necessitates the assurance that
between sensitive and personal data. For example, fitness exclusively authorized individuals possess entry to the
apps on mobile and wearable devices can profile user data and resources. This objective can be accomplished
lifestyles and infer their inclination to diseases through the by integrating access control protocols, such as role-based
analysis of movement data, heartbeats, and weight. access control (RBAC) or attribute-based access control
Despite the undeniable benefits of Big Data analytics, the (ABAC), within the MapReduce framework. These
absence of standard data protection tools makes these models of access control serve the purpose of outlining
services vulnerable to attackers. [3] and enforcing access regulations grounded on user roles,
attributes, or alternative criteria. Through the integration
Figure 3 provides an elucidation of several prominent of access control within the MapReduce framework,
security obstacles encountered in the realm of big data, entities are able to safeguard confidential information and
encompassing encryption, key management, access ascertain that solely authorized individuals are capable of
control, distributed environments, and numerous others.
executing particular operations on the data. [12]
Of all these concerns, access controls hold the utmost
significance in ensuring the authentication of data [3]. 3- NoSQL datastores
NoSQL datastores are flexible, scalable, and efficient
systems for Big Data. They can be classified into three
IV. Platforms of Big Data processing classes: key value, wide column, and document-oriented
The emergence of big data has resulted in the creation datastores. Key-value datastores are like big hash tables
of new technologies and methods for gathering, storing, and are used for efficient look-up operations. Wide
and analyzing data, such as Hadoop, Spark, and NoSQL column stores are good for blogging platforms and
databases. These tools allow organizations to extract content management systems. Document-oriented
valuable insights from their data, like detecting patterns, datastores use hierarchical records and are used for event
trends, fraud, and enhancing operational efficiency. Big logging systems and content management systems. Fine
data has significant implications for research, particularly grained access control in NoSQL datastores is still in the
in medicine, biology, and social science, enabling early stage with only a few frameworks proposed so far.
researchers to analyze large genetic datasets for drug [3] NoSQL datastores, like MongoDB and Cassandra,
targets and study social media data for public opinion on offer flexible storage for big data but have different access
specific topics. [4] control from relational databases. NoSQL databases have
varying access control mechanisms, like MongoDB's user
1- Apache Hadoop authentication and RBAC. RBAC in NoSQL lets admins
Hadoop is an open-source software, facilitates the define roles and privileges for users to control data access.
distributed processing of large datasets while ensuring Some NoSQL databases support ABAC for access
high availability and addressing failure at the application decisions based on user attributes. Access control in
layer; its additional components include Hadoop Map NoSQL varies by system, so organizations need to
reduce, Hadoop common, Hadoop distributed file system configure and manage settings for data security. NoSQL
(HDFS), and Apache YARN, while Apache Ranger and datastores offer customizable access control for specific
Apache Sentry serve as open-source programs within the security needs in big data environments. [6]
Hadoop environment, providing authentication methods V. Access control technology
and access control to ensure data security and prevent data
breaches. [4] Access control is a software module or function used
to protect resources or prevent illegal access. The access
Hadoop-based service are hosted by cloud service control model is formed with security policies that
providers such as Amazon, Google, and Microsoft. [5] determine authorized and unauthorized users. An efficient
Hadoop is a Java-based framework for processing big access control model prevents illegal data access.
data in distributed computing. It's part of Apache project. Examples of unauthorized access include a terrorist
Hadoop cluster uses Master/Slave structure for processing reading military records, a hacker using a duplicate bank
data. [6]basic big data model consists of a Master Slaves account, a hospital receptionist reading patients' medical
system. The Master Node receives data from providers reports, and a server attack denying data access to
and distributes tasks to slaves. The Master Node provides authorized clients. [2] Access control is also crucial for
three important tasks to the slave node: load forwarding, compliance with regulations and standards such as the
data forwarding, and analysis collection. The slave node General Data Protection Regulation (GDPR) and HIPAA.
processes the load from the master node for big data. This These regulations frequently mandate organizations to
architecture is susceptible to attack due to a failed enforce stringent access controls to safeguard sensitive
firewall. Big data requires a proper access control data. [3] fig1 illustrate the access control mechanism in
mechanism. [2] general.
2- Map reduce systems : Big data expansion leads to increased data flow in
cloud computing. Access requests in this environment are
MapReduce is a method for analyzing large data sets. complex. Access control faces challenges due to massive
It works by dividing the data into smaller parts and data access. [5] Access control is an essential aspect of
distributing them to different computers. The data is then big data security that helps organizations and researchers
 Figure 1 access control mechanism [3]
to protect sensitive information and comply with relevant policy violations for timely response. Blockchain-
regulations. Implementing strict access controls, such as based access control technology provides inherent
role-based access, attribute-based access control, multi- auditability by recording all events on the
factor authentication, and data-level access control, can blockchain. This allows for easy tracking,
help organizations and researchers to effectively secure monitoring, and auditing of access control events,
their big data and protect it from unauthorized access. It ensuring compliance with regulations. [12]
ensures that only authorized users can access and view
sensitive information [4] . This is particularly important 4- Auditability: access control system must
for big data because it often contains sensitive and effectively track and log access attempts for big data
confidential information, such as personal and financial security analysis, Auditing and logging are essential
data, that can be exploited by cybercriminals if it falls into for monitoring access to resources, creating a record
the wrong hands [1] for accountability and security. Organizations use
auditing to detect unauthorized activities and
a- Access control requirements prevent security breaches. Logging records user
actions for forensic investigations and compliance
In this section a review of the key Requirements of audits. Auditing and logging aid in meeting
access control mechanism for Big Data in cloud regulatory requirements for data protection. A
platforms. comprehensive audit trail helps organizations
1- Fine-grained access control (FGAC) The access enhance access control policies and security
control of data in servers is managed by software, measures by identifying access behavior patterns
which checks authorization for accessing specific and trends. [1]
data. However, the software is not entirely reliable 5- Efficiency : Efficient policy compliance
due to the risk of information leakage by those who mechanisms are necessary. Two main approaches
have access to the server or cloud. To address this for enforcing FGAC in traditional relational DBMSs
issue, fine-grained access control is implemented, are view-based and query rewriting. It is important
where different individuals have different rights to to determine how suitable these approaches are for
access files based on their identity. This type of the Big Data scenario and how they can be
access control is particularly useful in encrypted customized or extended. [3]
attribute searching and personal health records, as it
ensures efficient and secure data access [2] Access b- Access control models
control rules in Big Data analytics platforms need to
be bound to data at very specific levels, but new There are several types of access control methods that
enforcement mechanisms are required due to the organizations and researchers can use to protect big
heterogeneity and lack of schema in Big Data. [3] data, including :

2- Context – management :The support for context- 1- Discretionary Access Control Model (DAC)
based access constraints allows for highly DAC is a basic access control model where the owner
customized access control forms. These constraints manages all object access, deciding which users can
can limit access based on time periods or access and to what extent. The owner has full control,
geographical locations. When access control granting access privileges to users as desired. The aim is
decisions are derived from contexts, authorizations to prevent unauthorized access through Access Control
are granted if conditions referring to the Lists (ACL) for each object .DAC offers flexibility by
environment of the access request are met. [3] allowing object owners to customize security
3- Scalability: The capability to monitor access mechanisms, with easily updatable policies. It simplifies
control events and changes ensures transparency and usability and reduces administrative complexity in
accountability. Recording and analyzing access distributed environments. While efficient for a small
activities creates a detailed audit trail. Auditability number of users and data, DAC is easier to manage than
aids in identifying unauthorized access attempts and other access controls and is widely used in networking
 operating systems like UNIX, WINDOWS, and Linux. attributes, and a user's private key is associated with a set
However, DAC has limitations in modern computing of attributes that they possess. To decrypt the data, the
environments due to increasing security requirements and user's private key must match the attributes associated
access control complexity. It may fail to protect shared with the encrypted data. ABE provides fine-grained
resources in high communication scenarios, potentially access control, allowing different users to access different
compromising data confidentiality, integrity, and parts of the encrypted data based on their attributes. This
availability if a mistake is made. Additionally, the makes ABE particularly useful in scenarios where access
requirement for an owner in DAC may not align with control needs to be flexible and dynamic, such as in
contemporary computing technologies, leading to lower supply chain management systems. [9]
security levels compared to other access controls.
Moreover, DAC is stateless, lacking awareness of events
post-permission granting. [2]
2- Role Based Access Control (RBAC)
is an access control model that provides a structured
approach to managing user permissions within an
organization. In RBAC, access rights are assigned to
roles, and users are assigned to specific roles based on
their job responsibilities and functions. This model
simplifies access management by allowing
administrators to define roles and their associated
permissions, rather than assigning permissions directly Figure 2 attribute encryption access control [5]
to individual users. RBAC offers several benefits, 6- Blockchain access control:
including improved security, scalability, and ease of
Blockchain-based access control uses blockchain tech
administration. It helps organizations enforce the
for managing access rights to resources. Access
principle of least privilege, where users are granted only
permissions are on a distributed ledger for transparency.
the permissions necessary to perform their tasks. RBAC
has been widely adopted in various domains, including Smart contracts on the blockchain enforce access policies.
cloud computing, where it plays a crucial role in ensuring Blockchain's decentralization enhances security and
secure access control for big data analytics in the cloud privacy. Access events are recorded on the blockchain for
environment [10] auditability. [12]The blockchain efficiently transfers
information between parties. A block is created by one
3- Attribute-based access control (ABAC) party in a transaction. Many computers verify the block
before it is added to a chain. Falsifying a record in the
The overall approach for ABAC is defined in the chain is extremely difficult. Bitcoin uses this system for
National Institute of Standards Special Publication (NIST transactions. The blockchain is maintained by a peer-to-
SP) 800-162 as the ABAC trust chain [5] method that peer network of interconnected nodes. Each node
enables organizations to regulate data access by processes input to produce an output. The network divides
considering the attributes of the user, such as their job work among equally privileged peers. The finance
title, location, or clearance level. By employing this industry has been transformed by combining a peer-to-
approach, organizations can establish precise access peer network with a payment system. [7]
controls that are customized to particular data sets or
specific use cases. [3] VI. Comparison between access control
4- Multi-factor authentication (MFA) models
necessitates users to provide multiple forms of
authentication, such as a password and a fingerprint or a The DAC model, being the most simplistic, entails
token, in order to access data. This enhances the difficulty access control by the owner for operations such as
for cybercriminals to obtain unauthorized access to reading, writing, and modifying. Unsuitable for big data
sensitive data. [3] In the setting of big data cloud contexts due to limitations in scalability and lack of fine-
computing, access frequency is high, with multiple grained control in intricate environments with numerous
tenants accessing data resources simultaneously. To users and data elements. So that’s make it not suitable for
prevent access control failure, different applications and big data and cloud computing environment
computing resources are limited, involving multi-user
access control technology. Integration of RBAC model RBAC, known for its efficiency in managing access
and multi-tenant access technology enhances access based on well-defined user roles and their associated
control in big data cloud computing, forming a needs, is considered appropriate. It is particularly well-
management system to improve tenants' access control suited for big data scenarios featuring structured roles,
over data. [5] offering efficient management of access in systems with
a high volume of users.
5- Attribute based Encryption Access Control
ABAC relies on dynamic attributes related to the user,
Attribute-based encryption (ABE) is a cryptographic data, and environment to make access decisions.
technique that allows access to encrypted data based on Designed for complex big data landscapes, this model
specific attributes or characteristics of the user. ABE uses provides detailed control based on multiple factors,
a combination of public and private keys to encrypt and making it a suitable choice for environments with diverse
decrypt data. [9]In ABE, data is encrypted with a set of
 DAC RBAC ABAC ABE Blockchain

Fine-grained Low Medium High High Medium

access control
Context Low Limited High N/A High
management
Scalability Low Medium High Medium Medium
Auditability Low Medium High N/A High
Efficiency High High Medium Low Medium
Integrity Low Medium High High Medium
Application Simplest Ideal for A good for specific an
with big data option, but not complex big choice for big scenarios emerging
environment recommended data data with requiring technology
for big data environments well-defined fine- with potential
due to with diverse user roles and grained for big data
scalability and access needs access access security, but
security and dynamic requirements. control managing
limitations. contexts through complex rules
data can be
encryption. challenging.
access requirements, although handling attributes can by merging different models. The selection of a hybrid access
pose challenges. control model for big data relies on specific needs such as
scalability, flexibility, security, and compatibility. These
The Blockchain Access Control model employs a models present a promising approach to access control in big
decentralized and tamper-proof ledger to enforce access data, balancing security, scalability, and flexibility. [12]
control rules. While holding potential for enhancing
security and auditability in big data settings, the VII. Conclusion
complexity of managing intricate access rules on a
blockchain may present challenges. Table (2) shows a The realm of big data introduces distinctive security
comparison between these models based on the challenges owing to its vast volume, variety, and velocity.
requirement of big data access control Conventional access control frameworks encounter
difficulties in adequately regulating access to extensive arrays
Choosing the best model of heterogeneous data in dynamic settings. The present study
The development of big data cloud computing emphasizes delved into a range of access control models applicable to
the significance of access control for system security in this ensuring security in big data contexts, encompassing RBAC,
environment. Traditional access control methods do not ABAC, DAC, ABE, and Blockchain access control. Upon
easily apply to big data cloud storage due to its dynamic examination, it was evident that RBAC strikes a harmonious
nature. Research in this area is still in early stages and lacks equilibrium between efficacy and governance for large-scale
consideration for dynamic access control features. Big data data by delineating clear-cut roles. ABAC surfaced as the
environments require real-time and dynamic adjustments to most versatile and adjustable model suited for intricate
access control based on changing attributes and requirements scenarios in big data with manifold access requirements and
of objects and subjects. [6] [1] fluctuating contexts. Nevertheless, its intricate attribute
management demands meticulous attention. Despite being
In general RBAC stands out as a suitable choice for big the most straightforward model, DAC's drawbacks in terms
data environments characterized by clearly defined user roles. of scalability and meticulous control render it ill-suited for the
ABAC emerges as the preferred option for intricate big data majority of big data applications. ABE furnishes robust
landscapes with diverse access needs. Blockchain access encryption-centric access control but may exhibit reduced
efficiency and necessitate specific use cases. The domain of
control, although promising for enhancing big data security, Blockchain access control introduces appealing attributes like
poses challenges in managing complex rules effectively. It is immutability and auditability; however, the management of
control, although promising for enhancing big data security, intricate rules on a blockchain can pose challenges. In
poses challenges in managing complex rules effectively. It is essence, the selection of an optimal access control model
essential to remember that the selection of the most suitable hinges on the particular security prerequisites, scalability
model hinges on the specific security requirements of the big imperatives, and access intricacies prevalent within the
data . sphere of big data. Through a comprehensive grasp of the
merits and demerits of each model, establishments can deploy
Many researches presented hybrid access control models a robust and effective access control strategy to safeguard
to overcome the vulnerabilities of traditional models , Hybrid their invaluable big data assets. This investigation lays the
access control models integrate various technologies to foundation for further inquiry into hybrid methodologies that
enhance access control in big data settings. They utilize amalgamate the strengths of diverse models to cater to the
different technologies to meet the evolving requirements in perpetually evolving security landscape of big data.
big data. Hybrid models can incorporate DAC, RBAC, and
ABAC to offer a detailed and scalable access control solution. VIII. References
These models enable dynamic and adaptable access control
[1] N. K. Odugua and A. Rajeshb, "A Fine-Grained
Access Control Survey For The Secure Big Data
Access," Turkish Journal of Computer and
Mathematics Education, pp. 1-7, 28 april 2021.
[2] K. V. a. V. Jayalakshmi, "Shared Access Control
Models for Big".
[3] Pietro Colombo and Elena Ferrari, "Access control
technologies for Big Data management systems:
literature review and future trend," Springer open
access, p. 14, 24 jan 2019.
[4] S. Pandey and S. Maurya, "Big Data Security
management through Task Role Based Access Control
Mechanism," IEEE, pp. 1-6, 13 mar 2023.
[5] A. M. Tal and . C. C. Zou, "A Framework for
Attribute-Based Access Control in Processing Big
Data with Multiple Sensitivities," Appl.Sci., 2023.
[6] F. Wang, H. Wang and C. Xingji , "Research on access
control technology of big data cloud computing,"
IEEE, 2023.
[7] Y. Zhu and . D. Huang , "From RBAC to ABAC:
Constructing Flexible Data Access Control for Cloud
Storage Services," IEEE, 2015.
[8] A. Arun, Rajan and Vetriselvi., "Systematic Survey:
Secure and Privacy-Preserving Big Data Analytics in
Cloud," Journal of Computer Information Systems,,
pp. 1-21, 2023.
[9] S. . K. Sharma, A. Pratap and Har, "Design of Access
Control Framework for Big Dataas-a-Service (BDaaS)
Platform," . Ymer, 21(07):1350-1368, 2022.
[10] O. Mounnan, A. Abou El Kalam and L. El Haourani,
"Decentralized Access Control Infrastructure Using
Blochchain for Big data," IEEE, 2019.
[11] V. N. Inukollu, S. Arsi and S. R. Ravuri, "Security
issues associated with big data in cloud computing,"
International Journal of Network Security & Its
Applications (IJNSA), p. 8, may 2014.
[12] H. Shittu and M. Nabil, "Smart Supply Chain
Management with Attribute-Based Encryption Access
Control," IEEE 13th Annual Computing and
Communication Workshop and Conference (CCWC),
2023.
[13] Yinxia Zhuang, "Research on big data access control
mechanism," International Journal of Computational
Science and Engineering, 26(2):192-192, 2023.
[14] . U. U. Uchibeke, S. H. Kassani* and R. Deters,
"Blockchain access control Ecosystem for Big Data
security," 2018.
 .

Conclusion