"There were experiments with moving both pieces of code to the GPU, and GPU did well in terms of speed, but they say getting data in and out of GPU made the approach unfeasible."
That would be the problem with GPUs :) Co-processors always have this problem. GPUs were designed for 3d video games- the data flow into the GPU is low-volume (textures & triangles) and the data flow out is (I assume) pretty minimal. 2011/3/13 [email protected] <[email protected]>: > Vectorizable code would be the major target, yes. > > Do you know if the guys from Yandex were using CUDA or OpenCL approaches? or > the old kind of opengl hacks. > > > On Sun, Mar 13, 2011 at 8:31 AM, Earwin Burrfoot <[email protected]> wrote: >> >> On Sun, Mar 13, 2011 at 00:15, Ken O'Brien <[email protected]> wrote: >> > To clarify, I've not yet written any code. I aim to bring a large >> > speedup to >> > any functionality that is computationally expensive. I'm wondering which >> > components are candidates for this. >> > >> > I'll be looking through the code but if anyone is aware of >> > parallelizable >> > code, I'll start with that. >> More like 'vectorizable' code, huh? >> >> Guys from Yandex use modified group varint encoding plus handcrafted >> SSE magic to decode/intersect posting lists and claim tremendous >> speedups over original group varint. >> They also use SSE to run the decision trees used in ranking. >> >> There were experiments with moving both pieces of code to the GPU, and >> GPU did well in terms of speed, but they say getting data in and out >> of GPU made the approach unfeasible. >> >> > I'll basically replicate existing functionality to run on the gpu. >> > >> > On 12/03/11 21:08, Simon Willnauer wrote: >> >> >> >> On Sat, Mar 12, 2011 at 9:21 PM, Ken O'Brien<[email protected]> wrote: >> >>> >> >>> Hi, >> >>> >> >>> Is anyone looking at GPU acceleration for Solr? If not, I'd like to >> >>> contribute code which adds this functionality. >> >>> >> >>> As I'm not familiar with the codebase, does anyone know which areas of >> >>> functionality could benefit from high degrees of parallelism. >> >> >> >> Very interesting can you elaborate a little more what kind of >> >> functionality you exposed / try to expose to the GPU? >> >> >> >> simon >> >>> >> >>> Regards, >> >>> >> >>> Ken >> >>> >> >>> >> >>> >> >>> --------------------------------------------------------------------- >> >>> To unsubscribe, e-mail: [email protected] >> >>> For additional commands, e-mail: [email protected] >> >>> >> >>> >> > >> > --------------------------------------------------------------------- >> > To unsubscribe, e-mail: [email protected] >> > For additional commands, e-mail: [email protected] >> > >> > >> >> >> >> -- >> Kirill Zakharenko/Кирилл Захаренко >> E-Mail/Jabber: [email protected] >> Phone: +7 (495) 683-567-4 >> ICQ: 104465785 >> >> --------------------------------------------------------------------- >> To unsubscribe, e-mail: [email protected] >> For additional commands, e-mail: [email protected] >> > > -- Lance Norskog [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
