Unit-3

GRID FRAMEWORK
3.1 Virtualization of IT systems

Virtualization is the latest in a long line of technical

innovations designed to increase the level of system abstraction
and enable IT users to harness ever-increasing levels of
computer performance.

At its simplest level, virtualization allows you to have

two or more computers, running two or more completely
different environments, on one piece of hardware. For example,
with virtualization, both Linux operating system and Microsoft
Windows operating system can be used on one server.

In slightly more technical terms, virtualization

essentially decouples users and applications from the specific
hardware characteristics of the systems they use to perform
computational tasks. This technology is likely to usher in an
entirely new wave of hardware and software innovation. For
example, and among other benefits, virtualization can simplify
system upgrades (and in some cases may eliminate the need for
such upgrades) by allowing users to capture the state of a virtual
machine (VM), and then transporting that state in its entirety
from an old to a new host system.

Virtualization is also designed to enable a generation of

more energy-efficient computing. Processor, memory, and
storage resources that today must be delivered in fixed amounts
determined by real hardware system configurations will be
delivered with finer granularity via dynamically tuned VMs.
What is a virtual machine (VM)?

In the simplest terms possible, a virtual machine (VM) is

a virtual representation of a physical computer.

Virtualization allows an organization to create multiple

virtual machines—each with their own operating system
(OS) and applications—on a single physical machine.
A virtual machine can’t interact directly with a
physical computer.

Instead, it needs a lightweight software layer called a

hypervisor to coordinate with the physical hardware
upon which it runs.
What is a hypervisor?
The hypervisor is essential to virtualization—it's a thin
software layer that allows multiple operating systems to run
alongside each other and share the same physical computing
resources.

These operating systems come as the aforementioned

virtual machines (VMs)—virtual representations of a physical
computer—and the hypervisor assigns each VM its own portion
of the underlying computing power, memory, and storage.
This prevents the VMs from interfering with each other
The Benefits of Virtualization
Up to 80 percent greater utilization of every server.

Reductions in hardware requirements by a ratio of 10:1

or better.

Capital and operations expenses cut by half, with annual

savings of more than $1,500 for each server virtualized.
Robust, affordable high availability.
How does Virtualization Help Green Computing?

Virtualization results in far more efficient use of

resources, including energy.
Virtualization's purpose in a simple way is virtualize and
make a single piece of hardware function as multiple
parts.
Different user interfaces isolate different parts of the
hardware, thereby making each one behave and function
as an individual.

Installing virtual infrastructure allows several operating

systems and applications to run on a lesser number of
servers, helping to reduce the overall energy used for the
data centre and for its cooling.

The energy saved per server would translate into

approximately 7000 Kilo Watt hours per year, which is a
tremendous potential for energy savings, Virtualization
is the best to practice green computing, especially data
centres.
3.1.1 Virtualisation can be classified into 3 categories,
namely:-

Desktop Virtualisation

Server Virtualisation

Storage Virtualisation

Figure 3.1 Virtualization Management

3.1.1.1 Desktop Virtualisation

 Desktop virtualization, often called client virtualization,
is a virtualization technology used to separate a
computer desktop environment from the physical
computer.

Desktop virtualization is considered a type of client-

server computing model because the "virtualized"
desktop is stored on a centralized, or remote, server and
not the physical machine being virtualized.

Desktop virtualization "virtualizes desktop computers"

and these virtual desktop environments are "served" to
users on the network.

Another benefit of desktop virtualization is that is

lets you remotely log in to access your desktop from
any location.

Essential documents on Desktop Virtualisation

Managing the Desktop estate: the low risk route to

desktop virtualisation

Ten Reasons to modernise the desktop

A Peer Survey: Desktop Virtualisation – Separating the

Hype from Reality

The Next Generation Virtual Desktop Solution for

Growing businesses

Best Practices for testing Desktop Virtualisation

Desktop Virtualisation: A buyer’s checklist

3.1.1.2 Server Virtualisation

Server virtualization is the partitioning of a physical

server into smaller virtual servers.

In server virtualization the resources of the server itself

are hidden, or masked, from users, and software is used
to divide the physical server into multiple virtual
environments, called virtual or private servers.

One common usage of this technology is in Web servers.

Virtual Web servers are a very popular way of providing
low-cost web hosting services.

Instead of requiring a separate computer for each

server, dozens of virtual servers can co-reside on the
same computer.

Essential documents on Server Virtualisation

The Business Value Of Virtualization

The Future of Virtualization

Virtualizing Business-Critical Applications

 Five Steps to Determine When to virtualise Your Servers

Benefitting from Server Virtualization - Beyond Initial

Workload Consolidation

Getting the most out of virtualisation

User Survey Analysis: Next Steps for Server

Virtualisation in the mid market.

3.1.1.3 Storage Virtualisation

Storage virtualization is the amalgamation of multiple

network storage devices into what appears to be a single storage
unit. Storage virtualization is often used in SAN (storage area
network), a high-speed sub-network of shared storage devices.

The management of storage devices can be tedious and

time-consuming. Storage virtualization helps the storage
administrator perform the tasks of backup, archiving, and
recovery more easily, and in less time, by disguising the actual
complexity of the SAN.
 Users can implement virtualization with software
applications or by using hardware and software hybrid
appliances.

The technology can be placed on different levels of a storage

area network.

Essential documents on Storage Virtualisation

Learn about Storage virtualisation, its benefits and what

it can mean for your business and storage infrastructure

Evaluating Storage Technologies For Virtual Server

Environments

Storage Virtualisation- what to know and what to look

for

Server and Storage Virtualization: A Complete Solution

3.2 Role of Electric Utilities

IT equipment has become the third largest source of

power demand in the commercial sector, accounting for more
than 10% of an organization’s energy use.

In fact, IT and related communications technologies now

account for more than 2% of all global CO2 emissions.

To improving the energy efficiency and environmental

friendliness of your IT Department can be a real challenge.

The easiest way to reduce energy costs is to implement

an easy and cost-effective PC power management solution.
The Challenge of Power Management

A typical PC consumes between 400 kWh (kilowatt

hours) and 600 kWh of electricity each year, depending
on the brand, how it is equipped (e.g. LCD or CRT
monitor), and how hard the CPU is working.

The average cost of a kilowatt hour of electricity in the

United States averaged $.0898, and ranged from $.0533
to $.2536.

Assuming all computers in a 10,000 machine

environment are left turned on all of the time these rates
would result in an annual cost for electricity of
somewhere between $1,867,632 and $13,329,216.

If you consider that 50% of all PC’s are left on both

overnight and on weekends (70% of the total hours each
week), it’s reasonable to expect that actively
implementing effective power management policies
across the organization can result in savings of up to
35% (i.e. 50% of 70%) on your PC-related electrical bill.

Given the heat generated by a PC, reducing the number

of hours that each PC is turned on each day will reduce
the cost of electricity for air conditioning in warmer
months.

3.2.1 Role of Telecommuting, Teleconferencing and

Teleporting

Teleworking technologies are variously implemented for

green computing initiatives and many advantages include lower
greenhouse gas emissions related to travel, greater worker
satisfaction and, as a result of lower overhead office costs,
increased profit margins. Teleconferencing and telepresence
technologies are often implemented in green computing
initiatives.

The ways in which telecommuting is eco-friendly.

It reduces carbon emissions. Whether it’s by planes,

trains or automobiles, traveling in to work has a negative
impact on the environment. Cutting the commute, thus
prevents an excessive amount of carbon emissions to go
into the air.

It reduces electricity. When people work in an office,

almost everything is powered by electricity. From lights
to computers and printers—even the coffee machine—
everything consumes large quantities of electricity.

Working at home allow reduce electricity consumption

to what it really takes to make home office run.

It reduces paper printing.

It makes you take better care of your equipment. Think

about how often you actually shut down your computer
at work.

In addition to flexible schedule and saving time and

money, telecommuting also greatly helps the
environment.

Why Telecommuting is a Green Way to Work

Studies have shown that remote workforces contribute to

sustainable, environmentally-friendly workplaces by reducing
congestion, lowering fuel consumption, minimizing
construction, lessening pollution emissions, reducing the strain
on transportation systems, and improving air quality.

Example: Dell and Xerox have experienced this first-hand.

The average U.S. commuter produces an estimated 380

lbs. of CO2 each year during rush hour, but switching to full-time
telecommuting reduces each person’s work-related carbon
footprint by 98 percent.

According to Global Workplace Analytics, if workers in the

U.S. who held telework-compatible jobs (50 percent) and
wanted to (79 percent) worked from home just two days a week,
the U.S. as a whole would:

Gas Use: Save nearly 52 million gallons of gas—the

greenhouse gas equivalent of taking approximately
88,000 vehicles off the road per year.

Oil: Save over 2.6 million barrels of oil, valued at over

$264 million.

Roads: Reduce wear and tear on highways by over 1

billion miles a year.

The advantages are many-

Increased worker satisfaction,

Reduction of greenhouse gas emissions related to travel,

and

Increased profit margins as a result of lower overhead

costs for office space, heat, lighting, etc.
Implementation

Blackle

Fit-PC: a tiny PC that draws only 5w

Zonbu Computer

The Asus Eee PC and other ultra portables

Blackle

Blackle is a search-engine site powered by Google

Search.

Blackle came into being based on the concept that when

a computer screen is white, presenting an empty word
page or the Google home page, your computer consumes
74W.When the screen is black it consumes only 59W.
Based on this theory if everyone switched from Google
to Blackle, mother earth would save 750MW each year.

Fit-PC:

A tiny PC that draws only 5w

Fit-PC is the size of a paperback and absolutely silent,

yet fit enough to run Windows XP or Linux.

Fit-PC is designed to fit where a standard PC is too

bulky, noisy and power hungry.

Fit-PC draws only 5 Watts, consuming in a day less

power than a traditional PC consumes in 1 hour.
Zonbu Computer

The Zonbu is a new, very energy efficient PC.

The Zonbu consumes just one third of the power of a

typical light bulb.

The device runs the Linux operating system using a 1.2

gigahertz processor and 512 meg of RAM. It also
contains no moving parts, and does even contain a fan.

The Asus Eee PC and other ultra portables

The "ultra-portable" class of personal computers is

characterized by a small size, fairly low power CPU,
compact screen, low cost and innovations such as using
flash memory for storage rather than hard drives with
spinning plotters.

These factors combine to enable them to run more

efficiently and use less power than a standard form
factor laptop. The Asus Eee PC is one example of
anultraportable.

3.2.2 Limitations

Green Computing could be quite costly.

Some computers that are green may be considerably

underpowered.

Rapid technology change.

3.3 Materials Recycling
3.3.1 E-Waste Definition

E-Waste for short - or Waste Electrical and Electronic

Equipment - is the term used to describe old, end-of-life
or discarded appliances using electricity.

It includes computers, consumer electronics, fridges etc

which have been disposed of by their original users.

"e-waste" is used as a generic term embracing all types

of waste containing electrically powered components.

e-Waste contains both valuable materials as well as

hazardous materials which require special handling and
recycling methods.

Examples: Computers, LCD / CRT screens, cooling

appliances, mobile phones, etc., contain precious metals,
flame retarded plastics, CFC foams and many other
substances.
3.3.2 Categories of e-waste
Large Household Appliances

Washing machines, Dryers, Refrigerators,

Airconditioners, etc.
Small Household Appliances

Vacuum cleaners, Coffee Machines, Irons, Toasters,

etc.
Office, Information & Communication Equipment

PCs, Laptops, Mobiles, Telephones, Fax Machines,

Copiers, Printers etc.
 Entertainment & Consumer Electronics

Televisions, VCR/DVD/CD players, Hi-Fi sets,

Radios, etc
Lighting Equipment

Fluorescent tubes, sodium lamps etc. (Except: Bulbs,

Halogen Bulbs)
Electric and Electronic Tools

Drills, Electric saws, Sewing Machines, Lawn

Mowers etc.
Toys, Leisure, Sports and Recreational Equipment Electric
train sets, coin slot machines, treadmills etc.
Medical Instruments and Equipment Surveillance and
Control Equipment Automatic Issuing Machine

Figure 3.2 E-waste types

3.4 Recycling Technologies

e-Waste management practices comprise of various

means of final disposal of end-of-life equipment which have
different impacts on human health and the environment. It can
be distinguished between state-of-the-art recycling technologies,
which comply with high environmental and occupational health
standards and hazardous technologies that bear a great risk for
both health and the environment and are often applied in
countries, where no strict standards exist.

Figure 3.3 e-waste disposal

3.4.1 Hazardous Technologies

Incineration:

Incineration is the process of destroying waste

through burning. Because of the variety of substances found
in e-waste, incineration is associated with a major risk of
generating and dispersing contaminants and toxic
substances.
 Open Burning:

Since open fires burn at relatively low temperatures,

they release many more pollutants than in a controlled
incineration process at an MSWI-plant. Inhalation of open
fire emissions can trigger asthma attacks, respiratory
infections, and cause other problems such as coughing,
wheezing, chest pain, and eye irritation. Often open fires
burn with a lack of oxygen, forming carbon monoxide,
which poisons the blood when inhaled.
Land filling:

Land filling is one of the most widely used methods

of waste disposal. However, it is common knowledge that all
landfills leak. The leach ate often contains heavy metals and
other toxic substances which can contaminate ground and

3.4.2 State-of-the-art Recycling Technologies

The state-of-the-art recycling of e-waste comprises three steps

Figure 3.4 e-waste management services

Detoxication

The first step in the recycling process is the removal

of critical components from the e-waste in order to avoid
dilution of and / or contamination with toxic substances
during the downstream processes. Critical components
include, e.g., lead glass from CRT screens, CFC gases from
refrigerators, light bulbs and batteries.

Shredding

Mechanical processing is the next step in e-waste

treatment, normally an industrial large scale operation to
obtain concentrates of recyclable materials in a dedicated
fraction and also to further separate hazardous materials.

Refining

The third step of e-waste recycling is refining.

Refining of resources in e-waste is possible and the technical
solutions exist to get back raw with minimal environmental
impact. Most of the fractions need to be refined or
conditioned in order to be sold as secondary raw materials or
to be disposed of in a final disposal site, respectively.
During the refining process, to three flows of materials is
paid attention: Metals, plastics and glass.
3.4.3 E-waste Management – Six Steps

Figure 3.5 six steps to e-waste management

3.4.4 Benefits of recycling

Recycling raw materials from end-of-life electronics is

the most effective solution to the growing e-waste problem.
Most electronic devices contain a variety of materials, including
metals that can be recovered for future uses. By dismantling and
providing reuse possibilities, intact natural resources are
conserved and air and water pollution caused by hazardous
disposal is avoided. Additionally, recycling reduces the amount
of greenhouse gas emissions caused by the manufacturing of
new products.
3.6 GREEN DATA CENTRE

The demand for data centre capacity world wide has been
on the rise. This has also lead to a steady increase in
carbon emissions.

This is so because servers will not only handle greater

volume but will also require greater processing.

Data centres house a suit of large computers and

associated networks of the organization, forming the heart
of most businesses.

Data servers can be seen as powerful computers that have

the capacity to store as well as process vast amount of
multiformatted data.

As Cloud computing makes rapid strides, data, in its

myriad multimedia format will have to be stored and
instantly made available upon request.

The business users need to store data endlessly and also

comply with the legislations. The demand of storing and
processing of data is unabating.

The businesses that particularly deal with contents have to

improve their data centres through innovative strategies in
data management.

The data management solutions need to be agile so as to

cater to rapidly changing data needs.

Dynamic and agile data management implies ability to

modify, update, backup, and mirror data even as the
organizational needs of the data keep changing.
 Innovation, together with disciplined operational
management of the data centre is required. Costs and car-
bon emissions are also closely tied together in case of data
centres.

3.6.1 Influencing factors of green data centres

Data centre design, layout, and location:

Physical building in which the data centre resides.

This can be one building, or multiple buildings that
house the machines but are themselves spread across
geographical regions.

Architecture and design of the building, its

geographical region and the material used in
construction of the building
considerations h ere.
 The size and design of rooms in which servers are
housed and also the location of the server rooms
within the data centre can play a role in carbon
reduction.

Cooling, air conditioning, power source and

power consumption:

This includes the cooling strategies of the servers;

and the air conditioning relating to the actual
building.
This also includes use of green energy sources.

The impact of the physical location of the rooms to

be cooled, that are housing the servers.
Power management:

This includes lights and operational aspect, number

of people working, opening and closing of doors.

This would include procurement and installation of

green products and use of green services.
Servers:

Their numbers, positioning and corresponding

energy-efficient computing plays pivotal role.

Physical location of the racks, t heir positioning (hot

isle/cold isle), architecture and the physical rooms in
which they are placed are also important.
Design of each server—water cooled, air cooled,
and other efficiencies are also to be considered.
 Data strategy:

The main concerns here are including security and

backup.

Virtualization within each server, and combined

virtualization.
Organization of a cluster of servers—private cloud,
Space storage and usage strategy.

Virtualization aims to pool resources together to

deliver data centre services by pooling resources that
may be otherwise underutilized.

Adopting virtualization strategies will foster the

development of many virtualization architectures that
will enhance the data centre energy efficiency.

Networks and communications equipment:

This made up of land-based as well as wireless

communications such as switchgears, routers, and
modems.

The numbers and capacities of these equipment in

the data centre contribute to its carbon footprint.
3.6.2 Data Centre ICT Equipment—Server Strategies

They are housed within the green data centre and require
specific strategies for positioning, cooling, and usage. Servers
are powerful computers that form a significant part of the IT
assets of an organization. Increasingly these powerful servers
provide the organization with the ability to access, provide,
analyze, and store data, information, knowledge, and
intelligence in myriad different ways.

There is ever increasing demand for more powerful

servers with increased storage and processing facilities. With
more powerful processors and proliferating number of servers
the power consumption continues to climb rapidly. Servers
belong to the data centre manager who is responsible for
providing a service to the rest of the organization rather than
using it directly themselves.

The following are a list of green server strategy considerations:

Online, real-time list of server inventory that enables

location and uses of the servers.

Power consumption bill in real time—mapped to carbon

generation, that provides operational feedback to the
entire organization.

Bit to carbon ration as part of comprehensive data strategy

that provides metrics on not only the used ―bits‖ but
also the carbon generated by the provisioned bits.

Mirroring backup strategies that are balanced by the

―acceptable risks‖ of the data centre director.

Data capacity forecasting: Server capacities need to be

estimated on a continuous basis as the business changes.
The correlation between business change and growth,
and corresponding data centre capacity, is ascertained
based on statistical analysis, trend spotting, and
estimating the impact of technological innovativeness.
 Carbon-cost visibility: Lack of visibility of server costs
and particularly its mapping to individual or
departmental use of space.

Efficient decommissioning: Once the purpose of a

server is consummated, there is a need for a formal yet
quick way of decommissioning the server.

Incorporation right redundancy

Enhanced server distribution: Need to distribute,

through proper assignment, the use of the data space
across and between various departments/users.

Incorporate server switching: Data servers should be

capable of being switched from one type of usage to
another.

Incorporate Cloud computing and server virtualization.

3.6.3 Data Strategy and the Carbon Emitting Bit

Data strategy encompasses the use, storage, mirroring,

security, backups, clean ups, and architectures for data. It covers
both external and internal approaches to data management. Data
efficiency in relational database management systems includes
use of techniques such as data normalization and incremental
storage. Such practices enable creation of nonredundant and
flexible data structures which tend to save data storage space
when multiplied on a large scale. Using the correct data type
would also affect the amount of data space that is being used in
every ―bit‖ of data. Every ―bit‖ adds to the carbon generation
from the data centre. Following are the impact of one extra bit in
a data centre on the green performance of the organization:
 Additional free s pace provisioning.

Speed and density.

Backup

Mirroring.

Quality and reliability.

Security.

Provisioning. Each bit requires provisioning for spare

capacity, with corresponding need for spare room space,
people and infrastructure.

[1 bit + m b it (additional) leads to → 1.m bit × n w atts

(direct energy need) → leads to nxp watts (support
energy-infrastructure) influences → People (attitude)]
In addition to the data server strategies discussed thus
far, there is also a need to compliment those strategies with
astute IT governances that ensure incremental improvements to
the data centre performance. IT governance with additional
focus on data centres help to manage the overall number of
servers, their lifecycle and the underlying server virtualization
strategies.
3.6.4 Data Servers Optimization

Optimization of servers deals primarily with the

numbers, usage, and collaborations amongst the servers. This
data server optimization can be improved through better
organization of the databases including t heir design,
provisioning for redundancy, and improved capacity forecasting.
Optimization also includes consolidation of various physical
servers that would reduce their total numbers. Some of the
techniques that could be considered by an organization for
server optimization are:

Undertake intense and iterative capacity planning for the

data centre.

Undertake in-depth optimization through identification

of unused capacity of servers and storage disks within
them.

Implement full storage virtualization that will enable

hosting of multiple data warehouses on the same server.

Efficient server operations.

Efficient management of air-conditioning and cooling

equipment that require, at times, even more power to
cool the servers than required to operate them.

Decommissioning servers once their service level

agreement has expired.

Applying virtualization during architecture and design of

the servers, corresponding operating systems, and even
applications.

Making use of infrastructural and hardware economies

of scale.

Increasing B2B relation for a more common and

efficient solution service
3.6.5 Data Servers Virtualization

Data server virtualization, as a key strategy, includes

creation of many virtual servers from one physical server.
Virtualization has been popular as an efficient hardware
resource utilization; however, it also has significant impact on
reducing carbon emissions. A rough virtualization, data centres
can consolidate their physical server infrastructure as multiple
virtual serversarehosted on lesser number of servers. This data
centre specific program aims to improve energy monitoring,
advanced 3-D power management and thermal modeling
capabilities, better design techniques, cutting-edge virtualization
technologies, enhanced power management systems, and new
energy- efficient liquid cooling infrastructures. These initiatives
can not only improve building use, data server use, but also
reduce carbon emissions by almost 7,500 tons a year.
Virtualization has to be supported by the operating system that
would separate the underlying hardware from corresponding
application software. Various types of virtualization are:

presentation virtualization- wherein users get a feel for

owning the presentation of an application, whereas it is
actually shared

application virtualization- enables multiple users to use

the same application

desktop virtualization- applies the virtualization

techniques of the servers at a local, desktop level
storage virtualization-applied to databases

network virtualization- relates to the communications

and networking equipments of the data centre.
3.6.6 Physical Data Server Organization and Cooling
The physical arrangements of data servers, their
organization, and the manner in which the floor space
and racks are physically organized also impacts the
overall carbon emission from that data centre

Fig 3. 7: Two ways of physical arrangement of servers

Data centres use a number of techniques to cool their servers.
Water cooling has been popular to handle the
heat dissipation issues
Air cooling of servers using the concepts of hot-aisle
and cold-aisle

They not only reduce the carbon footprint of the

organization but, at the same time, improve its economic
performance by reducing power consumption. Therefore,
physical organization of the data servers, their operational
effectiveness and cooling strategies all play a crucial role inthe
overall reduction in carbon footprint.
 Physical arrangements of servers require the following
considerations:
Server optimization
Disk identification.
Implement a multitiered storage solution.

Specify low-power consumption, low voltage servers

together with high-efficiency Power Supply Units
Equipment Reuse.
Re-engineer Layout.
3.6.7 Cloud Computing and Data Centres

Cloud computing provides lot of opportunities for

organizations to consolidate their hardware and data space
requirements. Cloud computing cuts the costs of services and
also reduce carbon emissions. Saas can be used to to access key
enterprise applications such as customer relationship
management (CRM) and supply chain management (SCM)
through the Internet. The opportunities to reduce carbon
emissions increase with consolidation of both hardware and
software applications. Furthermore, the payment models for
SaaS-based applications is usually based on its usage. The
typical data centre planning that makes provision for
eventualities can be sidestepped for an overall planning by the
cloud service provider.

3.6.8 Networking and Communications Infrastructure

The data centres hold the communication equipment and

related assets. These communications infrastructure support the
internal and external networks of an organization and play a
significant role in its carbon footprint. Some of the
communication devices that contribute to carbon emissions
includes switches, routers, the LAN, WAN, and associated
mobile transmission devices. Monitoring of networks, their
interoperability, their uptimes and full-times, are also factors
contributing to the carbon footprint. Reduction of
communication traffic reduces server load, thus minimizing the
memory and processing time of the server. Following are the
categories of which demands attention in the context of carbon
emissions:

Local Area Networks (LAN): Local networks of the

organization that are made up of the physical connections
amongst the machines and data centre. Usually they lack
planning and architecture for LANs is a major factor in
consuming substantial power and thereby adding to the
cooling requirements.

Wide Area Networks (WAN): This enables communication

amongst its desktop and laptop machines with and beyond
its data centre. WAN comprises use of communication lines
that make up the virtual private network (VPN) of the
organization.

Mobile Networks: The mobile communications

infrastructure stack is made up of TCP/IP at the base,
followed by Well-integrated and optimized networks. They
incorporate combination of centralized and decentralized
approaches and plug-in sensors which can increasingly play
a major role in reducing carbon effects.
Other techniques: Personal Area Networking (PAN),
Metropolitan Area Networks (MAN), the IEEE 802.1x
group of standards and Infrared, Bluetooth, RFID, WiMax,
and Wireless VoIP also produce carbon.

Wireless L AN/WAN: While wireless communication may

give the impression of reduced hardware and infrastructure
it may still be inefficient and result in substantial carbon if
not properly architected during installation and not
monitored during operation.

WiMax: Mobile standard for point-to-point communication

that is based on radio frequency standardized technology
(IEEE 802.16) that tends to consume power, especially
when it is on but not in use. WiMax, made up of transceivers
to base antennas, need standards to ensure these networks
are switched on-and-off depending on their usage pattern.
3.5 BEST WAYS FOR GREEN PC

The hardware aspect of Green IT deals with the

architecture and design of IT hardware, the manner in which it is
procured and operated. The operational energy consumption is a
major issue for computer manufacturers. The impact a good,
energy optimum design can have on the overall energy
consumed by a piece of hardware over its entire life has to be
studied with utmost care. A purpose-built computer chip, or an
efficient laptop battery design has potentially greater impact in
reducing carbon emissions over its lifetime than its operation
would have.

Following is a more detailed description of these IT

hardware assets of an organization:

Data servers: deals with the physical machines and the

specific buildings in which they are housed. These
servers have both wired and wireless networks and other
corresponding communications equipment associated
with them that are directly emitting carbon.
End-user computers: laptops, desktops, their capacities,
operational efficiencies, and their disposal need to be
discussed from t heir P-O-D viewpoint. While the
efficient design and manufacturing of these end-user
devices remains the perceiver of the hardware
manufacturers, the efficient operation and disposal is
with the user organization.

Mobile devices: the mobile devices and associated

hardware, their batteries including the recharging
mechanism and disposal of the batteries and the policies

when the devices become outdated. The

mobile devices P-O-D is heavily affected by the user’s
attitude. Thus, a sociocultural issue is an important
contributor to the carbon behavior of these devices.
 Peripherals: printers, photocopiers, shredders, and so on.
These electronic gadgets are of immense interest in
Green IT due to their large numbers and overuse. These
devices has more operational waste and the carbon
associated with the eventual disposal of these fast
moving items.

The carbon emissions from each of these Green IT

hardware group mentioned above is affected by its procurement,
operations, and disposal (Green P-O-D) phases in its lifecycle.
Procurement focuses on well-designed, low-carbon emitting
data servers or monitors, buying it from a green supplier and
using the most efficient means of packaging and transporting the
equipment. Operation is the ongoing use of hardware in an
efficient and effective manner. Attitude of the end-user, affected
usually by visible metrics, plays a significant part here. Disposal
of IT equipment requires utmost care. The Green P-O-D phases
are practiced based on the policies of the organization.
3.5.1 Best practices for Green PC
The following are some of the tips to accomplish green PC:

Buy energy efficient hardware:

New offerings from major hardware vendors include

notebooks, workstations, and servers that meet the EPA's
Energy Star guidelines for lower power consumption.

Multicore processors increase processing output without

substantially increasing energy usage.

Also look for high efficiency (80%) power supplies,

variable speed temperature controlled fans, small form
factor hard drives, and low voltage processors
Use power management technology and best practices:

Modern operating systems running on Advanced

Configuration and Power Interface (ACPI)-enabled
systems incorporate power-saving features that allows to
configure monitors and hard disks to power down after a
specified period of inactivity.

Systems can be set to hibernate when not in use, thus

powering down the CPU and RAM as well.

Hardware vendors have their own power management

software, which they load on their systems or offer as
options.

There are also many third-party power management

products that can provide further flexibility and control
over computers' energy consumption.

Use virtualization technology to consolidate servers:

The number of physical servers can be reduced, and thus

the energy consumption, by using virtualization
technology to run multiple virtual machines on a single
physical server.

Because many servers are severely underutilized the

savings can be dramatic.

Consolidate storage with SAN/NAS solutions:

Just as server consolidation saves energy, so does

consolidation of storage using storage area networks and
network attached storage solutions.
 The Storage Networking Industry Association (SNIA)
proposes such practices as powering down selected
drives, using slower drives where possible, and not
overbuilding power/cooling equipment based on peak
power requirements shown in label ratings.

Optimize data centre design:

Data centres are huge consumers of energy, and cooling

all the equipment is a big issue.

Data centre design that incorporates hot aisle and cold

aisle layout, coupled cooling, and liquid cooling can
tremendously reduce the energy needed to run the data
centre.

Another way to "green" the data centre is to use low-

powered blade servers and more energy-efficient
uninterruptible power supplies, which can use 70 percent
less power than a legacy UPS.

Optimum data centre design for saving energy should

also take into account the big picture, by considering the
use of alternative energy technologies and catalytic
converters on backup generators, and from the ground
up, by minimizing the footprints of the buildings
themselves.

Energy-monitoring systems provide the information you

need to measure efficiency.

Use thin clients to reduce GPU power usage

Another way to reduce the amount of energy consumed

by computers is to deploy thin clients.
 Because most of the processing is done on the server, the
thin clients use very little energy.

A typical thin client uses less power while up and

running applications than an Energy Star compliant PC
uses in sleep mode.

Thin clients are also ecologically friendly because they

generate less e-waste.

There's no hard drive, less memory, and fewer

components to be dealt with at the end of their lifecycles.
Use more efficient displays

Replace the old CRT monitors with LCD displays. This

can save up to 70 percent in energy costs.
Not all LCD monitors are created equal when it comes to
power consumption.

High efficiency LCDs are available from several

vendors.
Recycle systems and supplies

To reduce the load on already overtaxed landfills and to

avoid sending hazardous materials to those landfills, old
systems and supplies can be reused, repurposed, and/or
recycled.

This hand-me-down method allows two workers to get

better systems than they had, while requiring the
purchase of only one new machine.

Old electronics devices can also be reused by those

outside the company.
 Much electronic waste can be recycled, the parts used to
make new items. Things like old printer cartridges, old
cell phones, and paper can all be recycled.
Reduce paper consumption

Another way to save money while reducing your

company's impact on the environment is to reduce your
consumption of paper.

This can be done by switching from a paper-based to an

electronic workflow: creating, editing, viewing, and
delivering documents in digital rather than printed form.

Send documents as e-mail attachments rather than

faxing.

When printing is unavoidable, you can still reduce waste

and save money by setting your printers to use duplex
(double-sided) printing.
Encourage telecommuting

The ultimate way to have a greener office to have less

office.

By encouraging as many workers as possible to

telecommute, you can reduce the amount of office space
that needs to be heated and cooled, the number of
computers required on site, and the number of miles
driven by employees to get to and from work.

Telecommuting reduces costs for both employers and

employees and can also reduce the spread of contagious
diseases.
3.7 GREEN GRID FRAMEWORK

The Green Grid is a nonprofit consortium whose mission

is to become the global authority on resource efficiency in
information technology and data centres. According to The
Green Grid website, the organization provides a forum for IT
directors, facilities managers and C-level executives to come
together and discuss different options for improving resource
efficiency. Findings and recommendations from these forums
are published on a regular basis. Metrics created and endorsed
by The Green Grid include:

Electronic Disposal Efficiency (EDE) - the percentage of

decommissioned information technology electronics and

electrical equipment that is disposed of through known

responsible entities.

Power Usage Effectiveness (PUE) - the ratio of total

facilities energy to IT equipment energy.

Data Centre Infrastructure Efficiency (DCIE) - the ratio

of IT equipment power to total facility power.
 Carbon Usage Effectiveness (CUE) - the product of the

amount of carbon dioxide emitted per kilowatt hour

(CEF) and the data centre's annual PUE.

Water Usage Effectiveness (WUE) - the ratio of the

annual site water usage in liters to the IT equipment
energy usage in kilowatt hours (Kwh).

Data Centre Productivity (DCP) - the quantity of useful

information processing completed relative to the amount
of some resource consumed in producing the work.

Deliverables of Green Grid:

Data Collection

Data Centre Standards and Metrics Inventory – this

study will document existing standards and metrics for
energy efficiency, identify coverage gaps and make
recommendations for future development.

The Green Grid Metrics: Describing Data Centre

Power Efficiency – this study will be an update to The
Green Grid's existing study on data centre efficiency
metrics and will look at workload classification through
a data centre segmentation model.
Operationalizing Energy-Efficiency Data Collection –
this study will identify the requirements for collecting
and aggregating data centre power consumption data.

Data Assessment
Data Centre Efficiency Baseline Market Study – this
study on the current state of the industry will allow The
Green Grid to identify key factors driving companies to
take action on data centre power consumption and the
challenges in doing so. Collecting and analyzing this
data will help to provide companies with a baseline to
compare their own initiatives, goals and performance.
Operational Best Practices – these studies will focus
on right-sizing the data centre and will outline best
practices in the adoption
consolidation technologies.

Database for Data Centre Performance – The Green

Grid will begin development work on a database focused
on data centre characteristics and performance schema.

Technology Proposals
Initial Technology Roadmap – this roadmap provides
an initial assessment of existing and emerging
technologies affecting data centre efficiency and
performance, taking into consideration both return on
investment and risk to the end user.
Power Distribution Options for the Data CentreStudy

– this study will look at the qualitative advantages

and

disadvantages of data centre power distribution

configurations.

Cooling Options Study – this study will focus on the

qualitative advantages and disadvantages of data centre
cooling architectures.