![]() If double-precision is must, then you’ll get mediocre improvement, at best, with current generation of hardware. I dont think these options are for me.ĭoes the PGI accelerator decision seem wise? (this will be my first time with GPU ‘programming’)Ī concern I have from cgorac’s post - do you think I will NOT get a speed improvement by moving processing to GPU if calcs are in double precisions? Intel Parallel studio, $799 contains composer, inspector and amplifier $366ea, for C and C++, again looks like CPU only. Intel Ct, C and C++, appears to be for CPU only - and beta? Open Current sounds like it may be beyond me at the moment. I understand the precoinditioned cinjugate gradient algorithm however not the math. OpenCurrent, looks like made exactly for this application. Does this come with above (ie license for £250?, will email PGI sales) Open MP itself was only partly successful however as got a 50% speedup on CPU only. PGI Accelarator CUDA F, Ive used OpenMP directives and loved the simplicity. If its £(?)250 for a 1 year license I can afford this and sounds like a good option. PGI CUDA F, £(?)899 is too much to risk with uncertainty that it will work. Model sizes are typically 650 65011 nodes with ~1,000,000 iterations for whole runtime.ĬUDA C - including conversion from FORTRAN to C and learning the low level syntax etc of CUDA will be a lot of work - but possibly the most rewarding with respect to speedup - and cheapest. The code size is 10,000 or so lines (solver probably only a few hundred - and where the main gains can be made). I would like to come back to OpenCurrent but for the moment plan 1 will be to implement CUDA as is. Will email sales.Įelson - again thanks, OpenCurrent looks ideal however i need to catchup on the maths of the solver using at the moment. Mkcolg - thanks for the heads up, Ive been looking through the PGI website (the demo videos were very useful) and i think accelerator is likely my best option - cost dependent. ![]() I will take your advice on staying with Fortran, thanks Im going to get a mid range nVidia card for the moment and if successful will upgrade when Fermi is released. Ive been reading around and am thinking:ĬGORAC - yes, double precision is a must. I am new to CUDA and would appreciate any advice. Do I need a more specialist card?ĭo you see any ‘not good plans’ in above? The program runs in double precission (required). For the purposes of CUDA would C++ be usable? Im willing to learn C, but would rather C++. When complete will make the code open source ![]() I want to develop and initially test on my ‘good last year’ gaming machine and if initial results look promising upgrade to a more intensive system. ![]() I am certain have got the options correct but suspect CPU parallelisation (at least letting the compiler decide where to parallelize) is not the answer. I have tried (automated) OpenMP implementation with the intel copmpiler on the Fortran code and have had limited success. I saw Portland Group have released CUDA Fortran which sounds like would skip the onerous step 1 however this is (£?)800 and beyond budget (unless speedup is extreme which I dont know yet). My plan is to convert the FORTRAN to C++, check all works through visual studio express then move to CUDAĪnd test different methods for speedup. Advice on below would be much appreciated. The code is a finite differnce model and I suspect very parallelizable. Im new and planning how to convert a LARGE serial Fortran program to run on GPU. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |