Skip to content
master
Go to file
Code

Latest commit

 

Git stats

Files

Permalink
Failed to load latest commit information.
Type
Name
Latest commit message
Commit time
 
 
HPL
 
 
 
 
 
 
 
 
 
 
 
 

README.md

clusterSoCBench

clusterSoCBench is a collection of benchmarks ported to ARM that can be used to evaluate the CPU+GPGPU performance of an ARM-based cluster.

For more details please visit our publication available here as paper.pdf:

R. Azimi, T. Fox and S. Reda, "Understanding the Role of GPGPU-accelerated SoC-based ARM Clusters", to appear in IEEE Cluster 2017.

Build instructions

Copy the source code to all nodes and compile each benchmark using make. For HPL, we used OPENBLAS. For message passing library:, we used OpenMPI.

More information about the individual benchmarks can be found here:

TeaLeaf: [https://github.com/UK-MAC/TeaLeaf]

CloverLeaf: [https://github.com/UK-MAC/CloverLeaf]

Jacobi: [https://github.com/parallel-forall/code-samples/tree/master/posts/cuda-aware-mpi-example/src]

Caffe: [https://github.com/BVLC/caffe]

Contacts

If you use any of our ideas, please cite our paper. Our contact information is:
reza_azimi at brown dot edu, tyler_fox at brown dot edu or sherif_reda at brown dot edu

About

A set of benchmarks to stress GPGPU accelerated cluster of ARM SoCs (tested with Nvidia TX1)

Topics

Resources

Releases

No releases published

Packages

No packages published
You can’t perform that action at this time.