{"id":4671,"date":"2012-11-13T11:33:35","date_gmt":"2012-11-13T10:33:35","guid":{"rendered":"http:\/\/www.walkingrandomly.com\/?p=4671"},"modified":"2013-01-31T14:07:55","modified_gmt":"2013-01-31T13:07:55","slug":"intels-xeon-phi-gpu-level-power-without-the-hassle","status":"publish","type":"post","link":"https:\/\/walkingrandomly.com\/?p=4671","title":{"rendered":"Intel’s Xeon Phi – GPU level power without the hassle?"},"content":{"rendered":"

Intel have finally released the Xeon Phi<\/a> – an accelerator card based on 60 or so customised Intel cores to give around a Teraflop<\/a> of double precision performance.\u00a0 That’s comparable to the latest cards from NVIDIA (1.3 Teraflops according to http:\/\/www.theregister.co.uk\/2012\/11\/12\/nvidia_tesla_k20_k20x_gpu_coprocessors\/<\/a>) but with one key difference—you don’t need to learn any new languages or technologies to take advantage of it (although you can do so if you wish)!<\/p>\n

The Xeon Phi uses good, old fashioned High Performance Computing technologies that we’ve been using for years such as OpenMP<\/a> and MPI<\/a>.\u00a0 There’s no need to completely recode your algorithms in CUDA<\/a> or OpenCL<\/a> to get a performance boost…just a sprinkling of OpenMP pragmas might be enough in many cases.\u00a0 Obviously it will take quite a bit of work to squeeze every last drop of performance out of the thing but this might just be the realisation of ‘personal supercomputer’ we’ve all been waiting for.<\/p>\n

Here are some links I’ve found so far — would love to see what everyone else has come up with.\u00a0 I’ll update as I find more<\/p>\n