1. IT Applications for the Abatement of
Climate Change
Virtualization & Cloud Computing
-Aparna Bulusu,
Faculty, Comp Science Dept.
St. Ann’s College for Women
Hyderabad
2. Agenda
Impact of IT on Environment
Data Centers – Reasons for expansion
Energy Usage Of & Emissions from Data Centers
Green Initiative 1 - Virtualization
How Virtualization helps
Green Initiative 2 - Cloud Computing
Cloud computing for achieving a green data center
Conclusion
3. Impact of IT on Environment
If the Internet was a country it would be the 5th biggest
energy consumer in the world
IT industry’s global carbon footprint – 2% (Equivalent to the
airline industry)
The upside – Greening WITH IT
The Downside – ICT related emissions are on the rise
Greening OF IT
4. Data Centers
Data Center: A Server Farm that houses computer systems
and associated communication & network infrastructure
and storage systems.
Types range from a Tier -1 data center – Simple server
room with around 96.5% availability
Tier 4 – Hosts mission critical applications with fully
redundant sub-systems and compartmentalized security
systems. Availability at 99.995%
Rigorously controlled temperature, humidity levels
coupled with uninterrupted power supply.
5. Microsoft Chicago Data Center in Northlake, Illinois 40-foot shipping containers filled with up to 2,000
servers. The first phase can hold up to 56
containers, and a second phase (currently shell
space) offers identical capacity, giving the Chicago
facility a total capacity of 112 containers holding
224,000 servers.
6. Some disturbing facts..
Power usage – 100 times that of an conventional office
building
Over 10% of TCO is spent on power needs
Data Center power consumption has doubled from 20002006,will double once again by 2012
15% of all servers run 24/7 but don’t do anything
Switching them off can realize global savings of $24.7 bn
Data Centers account for 14% of ICT footprint
Data center emissions will triple by 2020 (Gartner
Research)
9. AIM: Reduce PUE from 2 to 1.7 ~ 1.4
(Power Usage Effectiveness : Total Power Consumed/ Power
used by ICT equipment)
Decrease in IT power load Automatic reduction in
Cooling
e.g. If computer load power requirements are reduced
by 1.0 kilowatt (kW) - approximately 0.6 kW of airconditioning power is saved.
PRIORITY: Reduce the baseline consumption of data centers
10. Power consumption trends will influence and
determine future rate of emissions
Lesser / cleaner power consumed => Lower carbon footprint
Can be Achieved by changes in Policy, Infrastructure
and/or Technology
12. Green Initiative 1 – Virtualization
Virtualization : Creation of a virtual machine through use of an
additional layer of software called as ‘hypervisors’
Kinds of Virtualization: Based on Level of Virtualization
Full virtualization: Almost complete simulation of the actual
hardware.
Partial virtualization: Some but not all of them target environment
is simulated
Para virtualization: A hardware environment is not simulated;
however, the guest programs are executed in their own isolated
domains, as if they are running on a separate system.
The main operational difference between full virtualization and
para virtualizarion is that in the former, the hypervisor mediates
between the guests and the hardware, while in the latter the guest
OS is modified to cooperate with the hypervisor and the access is
shared
13. Virtualization Contd…
Server virtualization: Multiple virtual servers run on a single
physical server to fully utilize the hardware resources of the
physical server.
Run multiple applications and operating systems independently on
a single server.
Move workloads easily from one virtual workspace to another.
Configure and deploy new servers in a few minutes instead of
hours.
Reduce equipment and maximize server resources.
14. Desktop Virtualization
One server servicing many dumb terminals.
One Centralized system managing user
desktops, profiles , downloads and license.
Manage one server, one license , one
upgrade/downgrade, update one system
Maintenance and debugging: Only 1 computer
needs to be fixed and only one specialist
needs to be there who has a knowledge of all
these applications.
15.
16. How it works..
Install Virtualization software
Create Virtual Machine Instances
Configure Master Virtual machine settings on
host computer (including OS and applications)
Rapidly deploy virtual machines by copying the
master image to multiple Virtualization hosts
Achieve Zero down-time and 100% server
availability
17. Virtualization .. How it scales up
Application during normal usage (users and systems)
Application requiring additional resources due to an increase in
users (load)
Additional virtual machines added on the fly to support increased
usage
18. How Virtualization helps..
Significant power savings can be gained by consolidating underutilized
servers
For e.g. Consider consolidating four underutilized servers into one
A typical server power consumption at different loads
CPU clock speed
idle
load1 load2 load3
2000MHz
283W 290W 297W 305W
2333MHz
284W 296W 309W 317W
2666MHz
285W 305W 324W 335W
3000MHz
286W 313W 340W 354W
load4
311W
326W
347W
368W
Difference in power consumption between idle and max load is negligible.
Direct and indirect cost savings
Environmental benefits
Reduced physical data center footprint
19. Average Cost Reductions that can be availed through
Virtualization
80% less maintenance per year
25% capital cost savings
75% less to operate
25% productivity Increase
BOTTOM LINE – Dramatically Decreases The TCO by 50%
20. Summing it up - Virtualization Helps in ….
Cost Savings – Significant initial capital & ongoing operational
TCO savings including extended life-cycle
Enhanced Security & Compliance – By limiting data exposure
& letting IT manage endpoint compliance
Reduced Licensing Costs – Use of virtual machines means
significant reductions in licensing costs.
Management Improvement –
consolidation, maintenance, upgrades, disaster planning =>
fewer staff
IT Enabler – Focus is Shifted – Graduate from managing
computer systems to improving corporate services
Reduced Energy Consumption – Thin-clients typically use
anywhere from 75 – 90% less power.
22. What is the Cloud??
On demand over the Internet
Shared resources, software, and information are
provided to computers and other devices
Pay-per-use/ Subscription basis
Dynamically scalable and virtualized resources
Switch from company-owned hardware and
software assets to per-use service-based models.
23.
24. Saas ( Software as a Service)
Using a commercial vendor to develop the
software, host its operation and provide Internet-based
access and application support to customers.
Rather than licensing, installing and maintaining
software on clients’ computers or servers, vendors let
users access the software via the Internet on pay-asyou-go basis
Applications (word processor, CRM, etc.) or application
services (schedule, calendar, etc.) execute in the “cloud”
using the interconnectivity of the internet to propagate
data
25. Benefits of Saas
Benefit
Software hosted on third party premises rather than at the
enterprise.
Easy scalability as applications are accessed through web-based
interfaces.
Share costs by opting for multi-tenancy
Systematic support like maintenance/ debugging taken care of by
the providers.
Elimination of version compatibility, Reduced hardware footprint
Common Saas Applications: Web Conferencing, ecommerce, payroll, CRM, ERP etc.
Hugely Successful Saas Ventures: WebEx.com, SalesForce.com
• hardware f
26. Platform as a
Service (PaaS)
Applications are built in the “cloud” on the platform
using a variety of technologies
Simplifies orchestration of cloud services
Development, testing, and production environments
(servers, storage, bandwidth, etc.) are billed monthly
like hosting
Pay-as-you-go model
Environments scale up & down at the click of a button
Concerns include code & data privacy, security and
scalability
27. Benefits of Paas
Pay-as-you-go for development, test, and production
environments
Enables developers to focus on application code
Instant global platform
Elimination of H/W dependencies and capacity
concerns
Inherent scalability
Simplified deployment model
28. Infrastructure as a Service (IaaS)
Based on Virtualization
Computer resources
(processors, memory, storage, bandwidth, etc.) are provided
in an as-needed, pay-as-you-go model
Ability to scale up from a single server to entire data centers
Creates new opportunities such as Cloud bursting: Shifting
usage spike traffic to alternate resources
Infrastructure scales up and down quickly to meet demand
Built on a utility computing architecture to host a SOA
application layer
29. Iaas – Case Study
CapGemini offers cloud computing services in collaboration
with IBM Cloud Services.
Using a combination of physical and cloud servers, Capgemini
was able to show more than 33 percent savings on one
client’s disaster recovery service while decreasing the
recovery time objective (RTO) from 72 to just 24 hours.
By providing self-service features that are built into the Cloud
platform, application development time is reduced by as
much as 50 percent.
30. Amazon EC2
(Elastic Compute Cloud)
Your application will be executed on a virtual
computer (instance).
You have the choice of virtual computer, where
you can select a configuration of CPU, memory &
storage that is optimal for your application.
The whole cloud infrastructure viz.
servers, routers, hardware based loadbalancing, firewalls, storage & other network
equipments are provided by the IaaS provider.
31. Benefits of
cloud computing
Server Utilization – Cloud computing drives energy savings by improving
server utilization
Datacenter Efficiency: Improves efficiency by reducing power consumption
and better utilization of resources
Dynamic Provisioning: Cloud operators can quickly match server capacity
to demand shifts.
Cloud data centers can be located at sites with clean energy and favorable
environments. Policy level changes are also easy to make and implement.
33. Cloud Computing
Success Stories
GE - Global procurement hosting 500k suppliers and 100k
users in six languages on SaaS platform to manage $55B/yr
in spend
Bechtel - Reduced infrastructure cost by 30% in part by
achieving 70% server utilization
Washington DC - Google Apps used by 38k employees
reducing costs to $50/use per year for
email, calendaring, documents, spreadsheets, wikis and
instant messaging
Eli Lilly - Using Amazon Web Services can deploy a new
server in 50 days and a 64-node Linux cluster in 5min vs
100days
NASDAQ - Using Amazon Storage to store 30-80GB/day of
trading activity
34. Proper utilization of ICT
ICT’s largest influence will be in
enabling energy efficiencies in other
sectors
deliver carbon savings five times larger
than the total emissions from the
entire ICT sector in 2020.
7.8 GtCO2e of emissions savings in
2020.
Represents 15% of emissions in 2020
based on a BAU estimation
ICT-enabled energy efficiency translates
into approximately ¤600 billion ($946.5
billion) of cost savings
Source: Smart 2020 report by the Climate Group on behalf
of Global e-Sustainability Initiative & McKinsey
Consultants
35. Improvements
in the Office
• More efficient components: Micro-processors: Advances in chip design such
as ‘multi-core’ processors can save 30-60% of the energy used by the
processor if software is written to take advantage of this capacity.
• Power management: Low power mode which they can enter automatically
after a period of user inactivity
• Laptop computers use as little as a third of the energy of a
desktop.Similarly, flat screen monitors are much more efficient than oldfashioned cathode ray tubes.
• Use Thin clients
• Multi-functional devices
scanning, copying and fax
functions are more efficient than running several separate devices.
36. Conclusion
The ICT industry can greatly reduce its emissions by
relocating computing resources to zero-carbon data centers
powered either by solar, wind, hydroelectric or geothermal
sources..
The virtualized environment increased server utilization from
10 to 60 percent per server, further reducing the need for
additional physical servers
Large, public cloud infrastructures are better positioned to
help reduce the environmental impact of IT through efficiency
and economies of scale.
The global warming crisis requires radical changes in order to
go away, and these changes need to come from the top.
