By Matthew Dublin

From all accounts, 2010 looks to be the year of the multicore processor, but does this finally mean the emergence of HPC at the desk side or just really expensive space heaters that you can Tweet with? Despite a delayed rollout in 2009, Intel is planning on releasing a 6 core processor code named "Gulftown" sometime in Q2 of this year. The chip is capable of running 12 threads in parallel and will supposedly increase processing performance by some 50% over quad-core processors while drawing roughly the same of power. Intel is also working on a 6-core version of the Nehalem processor, which was originally released with 8-cores, in order to reduce heat issues, and will also be releasing an HPC version of the Nehalem which is slated to be called the Xeon 7500. The chipmaker's tera-scale computing research program is also touting their monster multicore experimental "single-chip cloud computer," a 48-core chip which they describe as architecturally resembling a cloud of integrated computers into silicon. Whatever that means...

And not to be outdone, AMD is also releasing a 21-core processor called Magny-Cours that is clocked at 2.2Ghz, chock full of memory channels, and will also run cooler when idle than AMD's 6-core Opteron.

The AMD Server team has also kicked off a contest for the best response to the question "What Would You Do With 48 Cores?" The winner will be awarded four new AMD Opteron processors, a TYAN S8812 motherboard that features 4 processor sockets with the capacity for you to install up to 8 DIMMs per sockets, and one copy of Windows Server 2008-Approximate retail value of all prizes is $8,189 USD. Contestants can write an essay, blog, or create a YouTube video expounding upon how they would utilize a 48-core machine to help society.

As Douglas Eadline points out, there are other concerns for the HPC users in addition to just multicore count, including parallel I/O, memory contention, and GP-GPUs, not to mention heat, power, and noise. Questions remain as to whether or not 48 cores or more even make sense for a single node, and how GP-GPUs, which definitely have a place in a lot of HPC application areas, can be leveraged in union alongside multicore processing to tackle large data sets. So as per usual, there's a lot of marketing noise about tremendously powerful hardware coming down the pipeline which is drowning out the voices of software developers scratching their heads, wondering how they can win the catch-up game.