Changes between Version 22 and Version 23 of GEMM_Performance_Cypress


Ignore:
Timestamp:
Aug 19, 2010 5:08:57 AM (14 years ago)
Author:
nakasato
Comment:

--

Legend:

Unmodified
Added
Removed
Modified

  • GEMM_Performance_Cypress

    v22 v23  
    1111Comment(20100812): With the latest Catalyst 10.7, slow transfer speed for GPU to CPU memory is gone. We have roughly ~ 6 GB/sec for both directions. However, it just represents practical maximum transfer speed with a combination of Cypress and X58 chipset. The actual transfer speed from/to the main memory is rather slow ~ 2.5 GB/sec at most with the pinned memory allocation. This is the problem! 
    1212 
    13 Update(20100820): We have implemented a proper treatment of alpha and beta. In the latest plot, we compare our results with ACML-GPU 1.1 and MAGMA BLAS 0.3 running on Fermi. The lines without "I/O" only measure kernel execution time.  
     13Update(20100820): We have implemented a proper treatment of alpha and beta. In the latest plot, we compare our results with ACML-GPU 1.1 and MAGMA BLAS 0.3 running on Fermi (we plot the numbers in "results_dgemm.txt" that is included in http://icl.cs.utk.edu/projectsfiles/magma/downloads/magmablas_gemm_fermi.tar.gz). Note the lines without "I/O" show the performance that does not take into account data transfer time between CPU and GPU. 
    1414 
    1515= Results =