Good day Today,
After looking into https://bitbucket.org/francesco-asnicar/pc-boinc/src/master/ and compiling the code myself in ubuntu, I noticed that de DoubleVector always uses the SSE code and not the AVX code.
file: DoubleVector.hpp
staring at Line: 60
#ifdef __AVX__
#ifdef __FMA__
#ifdef __AVX2__
#pragma message "Using FMA+AVX2 instructions"
#else
#pragma message "Using FMA+AVX instructions"
#endif
#else
#pragma message "Using AVX instructions"
#endif
typedef AvxDoubleVector DoubleVectorLong;
typedef SseDoubleVector DoubleVector;
#elif defined (__SSE2__)
Changed line 72 to:
typedef AvxDoubleVector DoubleVector;
It seems that haswell in 2017 was not very performant with AVX at the time and therefor it was disabled.
Running the compiled verion as follows:
bin/pc input/tile2.txt output/output2.txt 0.05 1 100000 0
The avx realtime is 22 second, the SSE version realtime is 32 seconds under ubuntu in hyper-v.
This only helps if the 5th parameter (100000) is large enough. If it is small like the test_run.sh scripts it almost makes no difference.
Could someone verify that the compiled gene_pcim_v1.11_win64__avx.exe is running in SSE instead of AVX mode?
Reply
Quote
valtercProject administrator
Project tester
Send message
Joined: 30 Oct 13
Posts: 624
Credit: 34,677,535
RAC: 0
The Windows source code is exactly the same of the Linux one, it was compiled using gcc/MinGW so I guess the behavior is the same.
Also keep in mind that the second-last command line parameter of the application is the number of columns in the input dataset. Right now, with the current experiment (FANTOM, hgnc_data_mat.csv), this number is 1829 (and this is the largest dataset that we ever used).
I suggest, if you want to benchmark the application in a real case, to download the small validation suite I provided: https://gene.disi.unitn.it/test/files/gene_validation.zip
Reply
Quote
I'm looking into it and see that it is running several sets. Each set runs for about 13 seconds. After that a new set begins.
I'm going to test the avx/SSE doublevector code in the pc algorithem part.
But it is allready heavely optimized.
I also understand why avx does not always help, the algoritem parts it needs to go though are small so it cannot strech it legs.
Reply
Quote