After Intel CEO Paul Otellini bragged that Intel's Xeon chip was the most energy-efficient processor, Neal Nelson compared it to AMD's offering and found that Intel wasn't being exactly truthful.
Recent tests by Neal Nelson & Associates, an independent computer performance consulting firm, have reported that in 36 of the 57 cases tested, an AMD Opteron-based server delivered better power efficiency than a comparably configured Intel Xeon-based server.
The tests were performed on servers configured with 2, 4, 6 and 8 gigabytes of main memory at various transaction processing load levels. The results show that for certain configurations and at certain load levels the Intel Xeon based server was 2.4 to 11.7 percent more power efficient while in other cases, the AMD Opteron based server was 9.2 to 23.1 percent more power efficient. In addition, when the systems were idle and waiting for transactions to process, the AMD server was 30.4 to 53.1 percent more power efficient.
Power consumption while the servers are idle is particularly significant since many servers spend most of their time waiting for work. A November 16, 2006-press release from IBM quotes a report by the Robert Frances Group, which states that on average servers in datacentres are idle 80 to 85 percent of the time.
The test results also showed that:-
- Larger memory configurations deliver both higher throughput and better power efficiency
- Intel's power efficiency advantages decrease as memory size increases,
- AMD's power efficiency advantages increase as memory size increases,
- For CPU-intensive workloads, the Xeon delivers 8.0 to 14.0 percent higher peak throughput,
- For primarily I/O intensive workloads the Opteron delivers 11.3 to 19.4 percent higher peak throughput.
Neal Nelson conducted these tests in response to a statement made by Intel CEO Paul Otellini in a July 18, 2007 analyst conference call. During that call Mr Otellini referred to Intel's "lead in power efficiency." Neal Nelson decided to use his company's benchmark toolset to determine if Intel actually had a lead in power efficiency.
In a somewhat dry comment, Neal Nelson said: "It appears that Mr Otellini's statement is inconsistent with the test results."
The tests were not financed or sponsored by any company or group.
An Intel spokesperson said: "Intel has no specific comment to make. We stand by our power efficiency claims. There is already an INDUSTRY STANDARD power efficiency benchmark called SPECPower and the results for that benchmark are at: www.spec.org/power_ssj2008/results/res2007q4/".