RSS Feed NEAL NELSON BENCHMARK LABORATORY

Server Power Efficiency - Test Methodology

Benchmark
Laboratory Home 		Summary
Test Results 		White
Papers 		About
This Site 		Press
Archive 		Our
Email List 		Contact
Information 		Neal Nelson
& Assoc. Home
A server's power efficiency is tested by building a relational database on the system and then measuring the power consumed as the server processes transactions. MySQL is used as the RDBMS and the test transactions simulate credit card purchases at a service station. The transactions are submitted as 'html' requests from 32 client machines to the Apache2 web server sotware package. There is one transaction per web request page.

The server's power usage is measured at the wall as the server is subjected to six different levels of transaction arrival rates: Idle, 2,400, 4,800, 7,200, 9,600 and 12,000 transactions per minute.

All servers are subjected to the identical transaction arrival rates. This facilitates quick and accurate comparisons between machines and allows statements like "When transctions were being processed at an arrival rate of 2,400 TPM, system 'A' consumed 133.6 watts while system 'B' consumed 145.8 watts."

All servers are loaded with the same operating system, SUSE Linux Enterprise Server version 10, and all machines are confgured with exactly the same 'tunables'.

Click here to read white papers that describe this procedure in more detail.

Copyright © 2006-2008
Neal Nelson & Associates 		Trademarks which may be mentioned on this site are the property of their owners.