Interestingly enough, I bet the modifications that were done to speed up for GPU use also gave significant improvement to the CPU version. Is that the case Eugene?
On Wed, Apr 17, 2013 at 4:40 AM, Mick O'Donnell <[email protected]>wrote: > Very interesting. I was actually using CUDA for a separate project (to > analyse sports injuries), so I'm afraid I don't have any JCUDA benchmarks > to share. But an improvement in the order of 1000 times is fantastic. I > thought it would reduce the time, but I didn't think it would have that > much of an effect. > > If I can assist at all with the re-writes, please let me know, I'd be glad > to help. Even if it's only to test / benchmark results on a different GPU. > I have a Quadro 2000 card here. I won't do anything with JCUDA for now. > > Michael > > > > > On Tuesday, April 16, 2013 4:39:37 PM UTC+2, Eugene Kononov wrote: >> >> >> Just wondering if anyone has looked at use parallel processing for >>> optimizing strategies yet. I've been using the NVIDIA's CUDA library for >>> other heavy lifting jobs with good results. There's a project called JCUDA >>> which supplies a Runtime API with Java bindings for CUDA. If no one has >>> written anything for this yet, I may take it on as a side project. >>> >>> >> >> Yes, we are actively working on this. We are using aparapi: >> https://code.google.com/p/**aparapi/ <https://code.google.com/p/aparapi/> >> >> >> Riunning the JBT optimization on he high end GPU (AMD Radeon 7900 series) >> indicates a speedup in the order of 1000 times faster. The big drawback is >> that it required a complete re-write of the multiple JBT classes, >> essentially flattening the object-oriented model into array processing of >> this type: >> >> final float inA[] = .... // get a float array of data from somewhere >> final float inB[] = .... // get a float array of data from somewhere >> (inA.length==inB.length) >> final float result = new float[inA.length]; >> >> for (int i=0; i<array.length; i++){ >> >> result[i]=intA[i]+inB[i]; >> } >> >> What kind of speed up do you get with CUDA/NVidia? How much modifications to >> the JBT code did it require? >> >> >> >> -- > You received this message because you are subscribed to the Google Groups > "JBookTrader" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To post to this group, send email to [email protected]. > Visit this group at http://groups.google.com/group/jbooktrader?hl=en. > For more options, visit https://groups.google.com/groups/opt_out. > > > -- You received this message because you are subscribed to the Google Groups "JBookTrader" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/jbooktrader?hl=en. For more options, visit https://groups.google.com/groups/opt_out.
