Difference between revisions of "Cray"

Revision as of 00:25, 25 October 2012

Cray XK6 machines

1. Configuring TAU:

module load cudatoolkit
./configure -arch=craycnl -cuda="$CRAY_CUDATOOLKIT_DIR" -cudalibrary="$CRAY_CUDATOOLKIT_POST_LINK_OPTS" -bfd=download

Setup your environment:

export PATH=<path to tau2>/craycnl/bin:$PATH
export LD_LIBRARY_PATH=<path to tau2>/craycnl/lib:$LD_LIBRARY_PATH

2. CUDA

Build as normally would, and modify your run command to be:

aprun -N 1 tau_exec -T serial,cupti -cupti ./matmult

3. OpenACC

Both PGI and Cray uses the CUDA driver API to interact with the GPU, so setup TAU to collect those calls:

export TAU_CUPTI_API=driver

Compile as normally would and run with tau_exec as well:

aprun -N 1 tau_exec -T serial,cupti -cupti ./matmult

@@ Line 1: / Line 1: @@
+<!--
 This document describes how to configure TAU and PDT on Cray XT3/RedStorm or Cray CNL. Choose the subsection that applies to you. Both cases assumes that the backend nodes are running Catamount OS.
@@ Line 88: / Line 90: @@
 ./configure -fortran=pathscale -c++=CC -cc=pathcc -arch=craycnl -pdt=/spin/proj/perc/TOOLS/pdt/pdtoolkit-3.12 -pdt_c++=g++ -mpi
 </pre>
+-->
+== Cray XK6/XK7 ==
+Cray XK6 machines
+=== GPU performance tracking ===
+. Configuring TAU:
+ module load cudatoolkit
+ ./configure -arch=craycnl -cuda="$CRAY_CUDATOOLKIT_DIR" -cudalibrary="$CRAY_CUDATOOLKIT_POST_LINK_OPTS" -bfd=download
+Setup your environment:
+ export PATH=<path to tau2>/craycnl/bin:$PATH
+ export LD_LIBRARY_PATH=<path to tau2>/craycnl/lib:$LD_LIBRARY_PATH
+. CUDA
+Build as normally would, and modify your run command to be:
+ aprun -N 1 tau_exec -T serial,cupti -cupti ./matmult
+. OpenACC
+Both PGI and Cray uses the CUDA driver API to interact with the GPU, so setup TAU to collect those calls:
+ export TAU_CUPTI_API=driver
+Compile as normally would and run with tau_exec as well:
+ aprun -N 1 tau_exec -T serial,cupti -cupti ./matmult