UNIVERSITY OF CALIFORNIA, LOS ANGELES
This project focuses on understanding the principles and methods for the design of green networks at the edge of the Internet. The total power consumption of edge networks is estimated to be quite significant, so even moderate improvements in energy-usage in an individual device can result in non-trivial savings overall. Obtaining these moderate improvements in the energy-efficiency of edge networks is challenging for two reasons: the diversity of edge networks and the dynamics in their workload. The technical approach in the project is based on first modeling, measuring at run-time, and predicting the power consumption in different components of networked systems in their usage context, and devising run-time adaptation and coordination mechanisms to manage energy via shutdown and performance scaling while maintain user and application performance. The project has the potential of significant societal benefit, targeted as it is on sustainable technologies. Moreover, the techniques it develops for energy efficiency can be broadly applied to other areas of computing: large server systems, mobile devices, and consumer appliances. Beyond its impact on technology, the project will also contribute to workforce development by training EE and CS students in sustainability.
Choose a quarter and click "Go."
| AWARD OVERVIEW |
| Award Number |
0905580 |
Funding Agency |
National Science Foundation |
| Total Award Amount |
$600,000 |
Project Location - City |
Los Angeles |
| Award Date |
08/26/2009 |
Project Location - State |
CA |
| Project Status |
More than 50% Completed |
Project Location - Zip |
90095-1406
|
| Jobs Reported |
0.00 |
Congressional District |
33 |
| Project Location - Country |
US |
|
|
Recipient Information
(Grants)
| Recipient Information (Grants) |
|
Recipient Name
|
UNIVERSITY OF CALIFORNIA, LOS ANGELES |
| Recipient DUNS Number |
092530369
|
| Recipient Address |
11000 KINROSS AVE STE 102 |
| Recipient City |
LOS ANGELES |
| Recipient State |
California |
| Recipient Zip |
90095-2000 |
| Recipient Congressional District |
33 |
| Recipient Country |
USA |
Required to Report Top 5
Highly Compensated Officials |
No |
Projects and Jobs Information
| Projects and Jobs Information |
| Project Title |
NETSE: MEDIUM: COLLABORATIVE RESEARCH: GREEN EDGE NETWORKS |
| Project Status |
More than 50% Completed |
| Final Project Report Submitted |
No |
| Project Activities Description |
Colleges, Universities, and Professional Schools |
| Quarterly Activities/Project Description |
During the quarter following activities were undertaken. First, using our previous developed instrumented Intel Atom processor platform with subsystem-level decomposition of energy, we studied the effect of compression algorithm choice on overall energy efficiency of data transmission. Our key finding was that the optimal algorithm is a strong function of the data rate and specific power parameters, so that adapting the algorithm selection dynamically based on variations in workload and platform characteristics is highly desirable. Second, we studied spatiotemporal variability in energy consumption in wireless (WiFi, cellular) communications on netbook and mobile phone class end nodes. Our findings indicate significant variability, which future work will exploit. Our third activity during the quarter focused on dynamically learning models that seek to predict on-line sub-system level energy consumption using data from system performance counters (as provided by most CPUs) and a whole system energy sensor. If successful, this approach will bring the benefits of deep inspection of power consumption to off the shelf systems that do not have built in fine-grained sensing instrumentation. Initial results are promising, being able to disambiguate processor and memory power consumption over short time windows to ~ 10 to 20% range, which is sufficient for many forms of adaptation. |
| Jobs Created |
0.00 |
| Description of Jobs Created |
Not applicable. |
Purchaser Information
(Grants)
| Purchaser Information |
| Contracting Office ID |
Not Reported |
| Contracting Office Name |
Not Available |
| Contracting Office Region |
Not Available |
| TAS Major Program |
49-0101 |
| Award Information |
| Award Date |
08/26/2009 |
| Award Number |
0905580 |
| Order Number |
|
| Award Type |
Grants |
| Funding Agency ID |
49 |
| Funding Agency Name |
National Science Foundation |
| Funding Office Name |
Not Available |
| Awarding Agency ID |
49 |
| Awarding Agency Name |
National Science Foundation |
| Amount of Award |
$600,000 |
| Funds Invoiced/Received |
$594,612 |
| Expenditure Amount |
$594,612 |
| Infrastructure Expenditure Amount |
$0 |
| Infrastructure Purpose and Rationale |
Not Reported |
| Infrastructure Point of Contact Name |
Not Reported |
| Infrastructure Point of Contact Email |
Not Reported |
| Infrastructure Point of Contact Phone |
Not Reported |
| Infrastructure Point of Contact Address |
Not Reported |
| Infrastructure Point of Contact City |
Not Reported |
| Infrastructure Point of Contact State |
Not Reported |
| Infrastructure Point of Contact Zip |
Not Reported |
Product or Service Information
(Grants)
| Product or Service Information |
| Primary Activity Code |
611310 |
| Activity Description |
Colleges, Universities, and Professional Schools |
| Sub-Awards Information |
| Sub-awards to Organizations |
0 |
| Sub-award Amounts to Organizations |
$0 |
| Sub-Awards to Individuals |
0 |
| Sub-Award Amounts to Individuals |
$0 |
| Number of Sub-awards less than $25,000/award |
0 |
| Amount of Sub-awards less than $25,000/award |
$0 |
| Number of payments to vendors greater than $25,000 |
0 |
| Total Amount of payments to vendors greater than $25,000/award |
$0 |
| Number of payments to vendors less than $25,000/award |
114 |
| Total Amount of payments to vendors less than $25,000/award |
$62,185 |
| Location Information |
| Latitude, Longitude |
34º 4' 6",
-118º 26' 41" |
| Congressional District |
33 |
| Address 1 |
|
| Address 2 |
|
| City |
Los Angeles |
| County |
Los Angeles |
| State |
CA |
| Zip |
90095-1406 |
|
 |