GAME ON: NVIDIA brings GPUs to 64-bit ARM servers
HPC users targeted with parallel processing code
ARM's march into data centres and the world's most demanding applications has taken a major stride forward with GPU king NVIDIA throwing its weight behind ARM server-makers by releasing a version of its CUDA parallel programming platform that works with the architecture.
HPC types have, in recent years, increasingly turned to GPUs to do the heavy lifting in some data-crunching applications. The reason for doing so is that GPUs possess many cores, an architecture HPC types can take advantage of by handing off small jobs to a dedicated core. With lots of little jobs all going in parallel, more work gets done than would be possible if a central CPU were asked to juggle as many jobs.
Such arrangements translate into lower costs and faster processing, outcomes it is hard to argue against.
Those backing ARM CPUs as an alternative to x86 in the data centre say their efforts bring about the first outcome, leading to ARM chippery's lower acquisition and operational costs being advanced as a good reason to overthrow Intel's dominance of the data centre and HPC rigs.
One barrier to that was, until today, that ARM didn't play as nicely with parallel processing as x86 CPUs.
That barrier is now gone and NVIDIA says three vendors have products ready to roll that bring GPU-assisisted co-processing to market. The three are:
- Cirrascale RM1905D - High-density two-in-one 1U server with two Tesla K20 GPU accelerators; provides high-performance, low total cost of ownership for private cloud, public cloud, HPC, and enterprise applications;
- E4 EK003 - Production-ready, low-power 3U, dual-motherboard server appliance with two Tesla K20 GPU accelerators, designed for seismic, signal and image processing, video analytics, track analysis, web applications and MapReduce processing; and
- Eurotech - Ultra-high density, energy efficient and modular Aurora HPC server configuration, based on proprietary Brick Technology and featuring direct hot liquid cooling.
Separately but at the same event, AppliedMicro also announced that its ARM-based X-Gene “Server on a Chip” is now in a state of “readiness” and that “ … development kits [are] available immediately, and production [will be] silicon available imminently.”
NVIDIA's also offering up a canned quote from IDC's program president for HPC Earl Joseph to the effect that "The availability of accelerated 64-bit ARM servers is one of the most significant developments to hit the HPC market this year.”
Joseph also says he sees “substantial interest within the HPC community in evaluating GPU-accelerated 64-bit ARM systems for next-generation computing projects."
Interest in evaluations is a long way short of commitment to buy or build, so is not a massive threat to AMD or Intel. Throw in the fact that in 2015 will offer HPC buyers "Knights Landing", a 72-core Atom-based product, and it is hardly game over for x86.
But NVIDIA and friends' efforts mean it is game on. ®