Nvidia Developing ARM Processor for Servers
January 6th, 2011 By: Rich Miller
Graphics chip specialist Nvidia (NVDA) is developing a CPU based on the ARM architecture that powers popular smartphones and tablets, the company confirmed last night at the Consumer Electronics Show in Las Vegas. Nvidia says its “Project Denver” initiative will result in hybrid servers that use graphics processing units (GPUs) in tandem with CPUs to accelerate computing processes. This concept was used in several of the top-performing supercomputers in the latest Top 500 power rankings.
The announcement at CES prompted plenty of analysis of Nvidia’s move and what it might mean for the market for x86 servers and data center performance. Here’s a sampling of notable discussion and commentary:
- ZDNet – There have rumors on and off for years that Nvidia was developing its own x86 CPU to compete with AMD and Intel. With this move, Nvidia seems to betting that the ARM ecosystem–the tools, compilers and applications already widely used in mobile and embedded devices–will gradually take over the rest of the computing world as well. “The energy around the ARM architecture is now absolutely enormous,” said Nvidia CEO and President Jen-Hsun Huang.
- NVIDIA Blog – Chief Scientist Bill Dally: “ARM is already the standard architecture for mobile devices. Project Denver extends the range of ARM systems upward to PCs, data center servers, and supercomputers. ARM’s modern architecture, open business model, and vibrant eco-system have led to its pervasiveness in cell phones, tablets, and other embedded devices. Denver is the catalyst that will enable these same factors to propel ARM to become pervasive in higher-end systems. Denver frees PCs, workstations and servers from the hegemony and inefficiency of the x86 architecture.”
- The Register – Huang said that Project Denver was one of the most important announcements that Nvidia has made in its history, and he said that a team of engineers with expertise with ARM, Sparc, x64, Power, and MIPS chips has been hard at work for years on figuring out what future smartphones, tablets, PCs, and servers might need in terms of CPUs and GPUs. “I think this is a game changer,” Huang said. Nvidia is not providing much in the way of detail about Project Denver, but Andy Keane, general manager of Tesla supercomputing at Nvidia, told El Reg that Nvidia was slated to deliver its Denver cores concurrent with the “Maxwell” series of GPUs, which are due in 2013.
- Reuters – “If you have the ability to have your own processor architecture and the support of operating-system vendors like Microsoft, then that becomes a formidable dynamic that Intel will have to contend with,” said Hans Mosesmann, an analyst at Raymond James. “You and I five years from now might look back at this … and say this was the turning point in the industry.”
- X-bit labs – While it remains to be seen how successful ARM architecture will be in servers, it should be stressed that Nvidia makes no promises about any actual products or time frames. Perhaps, by the time project Denver evolves into actual devices the market of servers will be completely different from today and highly-parallel accelerators will be more important than x86-based AMD Opteron or Intel Xeon microprocessors.
- Computerworld – The announcement of new CPU cores came on the same day that Microsoft announced it is developing a future version of its Windows OS for the Arm architecture. The new Denver chips will support the upcoming Arm-based Windows OS, said Ken Brown, an Nvidia spokesman. Microsoft’s Windows OS, which is used on most of the world’s PCs, currently works only on x86 chipsets from companies such as Intel and Advanced Micro Devices. An Arm version of Windows could provide users an option to buy Arm-based systems as an alternative to x86 products.
- eWeek – One top-tier OEM keeping an eye on developments in this area is Dell. Forrest Norrod, vice president and general manager of Dell’s Data Center Solutions unit, said a move to alternative instructions sets—like the ARM architecture—in the data center holds some promise, but also faces some challenges. Bringing greater energy efficiency to servers will be attractive to businesses, Norrod said. However, it would mean moving software from x86 to another instruction set is no small feat for companies, he said. “It’s not a trivial expense,” Norrod said.
For more news about servers and processors, check out our Servers Channel.