AMD Beats Intel - Again On Power Saving
Wednesday, January 16, 2008 by PPV
Remember my post titled AMD beats Intel on power efficiency? In that post, research firm Neal Nelson reports that AMD Beats Intel in 36 of 57 Power Efficiency Tests.
Now the same research firm conducted another test, but this time on servers using quad core processors from AMD and Intel. Neal Nelson’s Power Efficiency Benchmark used 4GB, 8GB and 16GB memory for their test.
It is interesting to note that FB-DIMM which is used by Quad-Core Intel Xeon consumes more power compared to DDR-II modules which is used on Quad-Core AMD Opteron servers. So this I think is one of the factors why the Intel Server consumed more power.
These new test results were collected with Neal Nelson's Power Efficiency Benchmark which is a client server test where up to 500 world wide web users from 32 separate computers submit individual transactions to a server running the Apache2 web server software, the MySQL relational database and Novell's SUSE Linux Enterprise Server operating system. The benchmark has a complex multi-user load with a large memory footprint, a high volume of context switches, significant network traffic and substantial amounts of physical disk I/O.
"I am certain that these test results are correct," Nelson said. He added, "I am so certain that I have begun to offer my customers a money back guarantee if their real world experiences do not agree with my benchmark data."
But ZDNet sees it the other way. Read their blog titled Beware of rigged CPU efficiency study.
Very similar in my country where two Giant TV Network wars on TV ratings. lol. It's very hard to tell who's telling the truth.
White paper with detail test results: here
Now the same research firm conducted another test, but this time on servers using quad core processors from AMD and Intel. Neal Nelson’s Power Efficiency Benchmark used 4GB, 8GB and 16GB memory for their test.
It is interesting to note that FB-DIMM which is used by Quad-Core Intel Xeon consumes more power compared to DDR-II modules which is used on Quad-Core AMD Opteron servers. So this I think is one of the factors why the Intel Server consumed more power.
These new test results were collected with Neal Nelson's Power Efficiency Benchmark which is a client server test where up to 500 world wide web users from 32 separate computers submit individual transactions to a server running the Apache2 web server software, the MySQL relational database and Novell's SUSE Linux Enterprise Server operating system. The benchmark has a complex multi-user load with a large memory footprint, a high volume of context switches, significant network traffic and substantial amounts of physical disk I/O.
"I am certain that these test results are correct," Nelson said. He added, "I am so certain that I have begun to offer my customers a money back guarantee if their real world experiences do not agree with my benchmark data."
But ZDNet sees it the other way. Read their blog titled Beware of rigged CPU efficiency study.
Very similar in my country where two Giant TV Network wars on TV ratings. lol. It's very hard to tell who's telling the truth.
White paper with detail test results: here
