Intel, aiming to re-establish technological dominance and stem the momentum of rival Advanced Micro Devices in x86 servers, has a new Xeon chip for multi-processor machines.
Intels Xeon MP 7100 series—code-named Tulsa—arrived on Aug. 29 and will appear in high-end x86 servers from more than 40 system makers.
Thus far, the dual-core 7100 family has received accolades from manufacturers—in particular for its massive 16MB Level 3 cache—but some industry observers questioned whether it will be enough to change users opinions on AMDs Opteron processors.
“If youve already looked at AMD, this isnt going to change your mind,” said Nathan Brookwood, an analyst with Insight 64, in Saratoga, Calif. “The real solution for Intel comes about this time next year with Tigerton,” a quad-core chip that will be based on Intels new Core microarchitecture.
While other new Xeons, such as the Xeon DP 5100 “Woodcrest” series, are based on the energy-efficient Core architecture, “Tulsa … continues Intels reliance on its aging NetBurst architecture,” said Charles King, an analyst with Pund-IT, in Hayward, Calif., in a report. That means, despite impressive performance, “Tulsa is also a power hog in comparison to equivalent Opteron processors,” King said.
The 7100 series—topped by a 3.4GHz model 7140M—uses up to 150 watts, although low-power 7100 models use 95 watts, equaling the Opteron.
Tom Kilroy, vice president and general manager of Intels Digital Enterprise Group, in Santa Clara, Calif., said the focus for users of such high-end systems tends to be more about performance than power.
Tulsa, Kilroy said, offers a twofold performance increase versus its earlier Xeon MP chips.
“I dont see energy efficiency really as important in this segment,” Kilroy said.
That is a key difference between Intel and AMD, according to AMD officials.
“We dont require customers to choose between high performance and great power savings,” said John Fruehe, worldwide market development manager for the AMD server and workstation business, in Austin, Texas, in a statement. “We believe they should be able to get both in the same platform.”
Theres enough promise in the Tulsa chip to entice users to take a look.
Jevin Jensen, director of IS at Mohawk Industries, in Dalton, Ga., said he wants to compare a Tulsa-based server to one running AMDs newest “Rev F” Opterons. “I have contacted both vendors about getting four-way demo units of each,” Jensen said. “On paper, it appears AMD may still have a slight advantage … but I will let our real-world tests decide the outcome.”
Dell, Hewlett-Packard, IBM, Fujitsu and Unisys are all among the larger OEMs to unveil servers featuring Tulsa chips.
Check out eWEEK.coms for the latest news, views and analysis on servers, switches and networking protocols for the enterprise and small businesses.