Difference between revisions of "UTK multicores + GPU"

From HCL
Jump to: navigation, search
Line 6: Line 6:
  
 
== Using Fupermod on hybrid multicore/GPUs node ==
 
== Using Fupermod on hybrid multicore/GPUs node ==
*Compiling : Create two separate directories for configuration with selected CPU cblas (e.g. gsl, acml, mkl)and GPU cblas (e.g. cublas).  
+
* Compiling : Create two separate directories for configuration with selected CPU cblas (e.g. gsl, acml, mkl)and GPU cblas (e.g. cublas).  
**For example: Using acml blas for CPU and cublas for GPU computing
+
** For example: Using acml blas for CPU and cublas for GPU computing
  
 
  cd fupermod/
 
  cd fupermod/
Line 18: Line 18:
 
  cd cuda_config
 
  cd cuda_config
 
  ./configure --with-cblas=cuda
 
  ./configure --with-cblas=cuda
 +
 +
* Build performance model:
 +
 +
/*
 +
* rankfile is for process binding
 +
* appfile tells what processes will execute
 +
*/
 +
 +
$ mpirun -rf rankfile -app appfile_fpm

Revision as of 10:42, 12 July 2012

List of machines

http://icl.cs.utk.edu/iclhelp/custom/index.html?lid=97&slid=180

Display a list of available GPUs

$ nvidia-smi -L

Using Fupermod on hybrid multicore/GPUs node

  • Compiling : Create two separate directories for configuration with selected CPU cblas (e.g. gsl, acml, mkl)and GPU cblas (e.g. cublas).
    • For example: Using acml blas for CPU and cublas for GPU computing
cd fupermod/
mkdir acml_config 
cd acml_config
./configure --with-cblas=acml
mkdir cuda_config 
cd cuda_config
./configure --with-cblas=cuda
  • Build performance model:

/*

* rankfile is for process binding
* appfile tells what processes will execute
*/

$ mpirun -rf rankfile -app appfile_fpm