Over the last few years, interest in “green computing” has motivated research into energy-saving techniques for enterprise systems, from network proxies and virtual machine migration to the return of thin clients. When selecting computers, there are many other considerations than just energy, such as computational resources, and price.

1. A
Seminar on
Green Computing
Software or Hardware: The Future of Green Enterprise
Computing
PRESENTED BY:NIRAJ KUMAR
Final Year (computer Science And Engineering)

2. Abstract
Green computing motivated
Considerations than just energy.
Save more energy than stateof-the-art
power management software or thin
clients

3. INTRODUCTION
Green computing has emerged as a new area
of research
The first International Energy Conservation
Month was October, 1979
In the forty years after the 1970s energy
crises, usage reduction has proven less
effective than improving effieciency

4. Energy Conservation Today
 Enterprises Today
Enterprise computing
government mandates
 Sleep Proxies
 managing the power state of all of the devices
 low-power embedded computer sleep proxy
 Clients and Servers
Improve efficiency
Virtual machine migration

5. Hardware or Software?
The prior section described a variety of hardware
and software solutions for conserving energy.
Our hypothesis is that computing is no different
than any other energy domain: efficiency is more
effective than reducing energy usage.
This section tries to answer a simple question–
‘Might buying laptops for everyone be the easiest
and most cost-effective solution?’

6. Hardware or Software?
• Energy Star Data

7. Hardware or Software?
Back of the Envelope
Two approaches for energy saving…
Light Green states.
Sleep Server reports.
Enterprise Measurements
Why We Could Be Wrong

8. Enterprise Measurements

9. Why We Could Be Wrong
Can laptops handle desktop workloads?
What about thin clients?
Do laptops save money?
Why not combine techniques?
Does energy consumption of computing
systems matter at the building level?

10. Our first data set comes from an
administrative department
The desktops draw average power of 55 to
100 watts
Thin clients all average between 15 and 17
watts, with a standard deviation of less than 1
watt
The result is an average per-user power cost
of 30 watts, 50% more than laptops
United States VM-Ware Deployment

11. German Computer Science Lab
Isolated thin client setup.
Measured the power draw of four clients and
one of the two servers in the setup .

12. Utilization
• CPU
• Workloads

13. Economic Feasibility
Whole-building Energy

14. Looking Forward
 Green Enterprises
 Enterprises and the Cloud
 Policy
 Future Research
 Operating Systems
 Virtualization
 Networks
 Hardware

15. Conclusion
Hardware efficiency is more important than
software solution both in energy and
economic perspective
Client computing do not provide savings over
laptops

16. References
Windows Remote Desktop.
http://www.microsoft.com/
Energy star. http://www.energystar.gov/,
2011.
Application-Layer Traffic Optimization
(alto).https://datatracker.ietf.org/wg/alto/.
www.google.com
www.wikipedia.com