1. Kristopher Blair Jason Duo Li Daniel Moberly Michael Sedillo Profiling performance of a Web Proxy on a multi-core platform

2.  Focusing on speed of web proxies for use in mobile networks – extremely common in modern age  File size is important in mobile networks – small devices and slower connections, therefore in need of smaller delays  File size vs. computation time tradeoffs are very important  Mobile proxies trans-code large images to small images for use on mobile devices – need to optimize the time it takes

3.  Web proxy in iMPACT’s mobile computing lab – written to optimize JPEG image delivery  Image trans-coding: web proxy trans-codes JPEG images for use on mobile devices  Want to analyze bottlenecks to see how the speed of the proxy can be improved  Make sure proxy functions within acceptable levels of power and thermal output

4.  Verify previous findings of the iMPACT research team  Evaluate performance versus power consumption and thermal outputs  Verify that the current web proxy design functions within acceptable levels of power consumption and thermal output.

5. Our network layout

6.  Timestamps for each event on each machine allows for timing analysis  Linux Lm_sensors for thermal data  “Watts Up?” meter for power consumption data

7.  Earlier work shows that network overhead is significant, especially in smaller files – TCP overhead is quite significant  Larger files still show overhead in connection setup Results of previous analysis On small files (300B – 70KB)

8.  Expect thermal output and power consumption to fall along standard curves Expected thermal output (°F) vs. Workload (Request /s) Expected power consumption (Watts) vs. Workload (Request/s)

9. Presentatio n, Phase I, II & III report lm_sensor Configuration Watts upProxy Coding Modificatio n Data Collection Data Analysis DanDuoKrisDan DuoMichaelDuo Michael Kris