CFP last date
20 June 2024
Reseach Article

Distributed OpenCL Distributing OpenCL Platform on Network Scale

Published on July 2012 by Barhh Eskikaya, D Turgay Alt?lar
Advanced Computing and Communication Technologies for HPC Applications
Foundation of Computer Science USA
ACCTHPCA - Number 2
July 2012
Authors: Barhh Eskikaya, D Turgay Alt?lar
e119926b-14a1-4e0e-b8bd-4d65d1212695

Barhh Eskikaya, D Turgay Alt?lar . Distributed OpenCL Distributing OpenCL Platform on Network Scale. Advanced Computing and Communication Technologies for HPC Applications. ACCTHPCA, 2 (July 2012), 26-30.

@article{
author = { Barhh Eskikaya, D Turgay Alt?lar },
title = { Distributed OpenCL Distributing OpenCL Platform on Network Scale },
journal = { Advanced Computing and Communication Technologies for HPC Applications },
issue_date = { July 2012 },
volume = { ACCTHPCA },
number = { 2 },
month = { July },
year = { 2012 },
issn = 0975-8887,
pages = { 26-30 },
numpages = 5,
url = { /specialissues/accthpca/number2/7559-1013/ },
publisher = {Foundation of Computer Science (FCS), NY, USA},
address = {New York, USA}
}
%0 Special Issue Article
%1 Advanced Computing and Communication Technologies for HPC Applications
%A Barhh Eskikaya
%A D Turgay Alt?lar
%T Distributed OpenCL Distributing OpenCL Platform on Network Scale
%J Advanced Computing and Communication Technologies for HPC Applications
%@ 0975-8887
%V ACCTHPCA
%N 2
%P 26-30
%D 2012
%I International Journal of Computer Applications
Abstract

This paper presents a framework that extends OpenCL by distributing computing process to many computing resources connected via network and enables the computing resources to run in parallel. Using JSON RPC (Remote Procedure Call technique relying on JavaScript Object Notation) in communication layer, Distributed OpenCL framework provides platform and operating system independency. Using this framework, a host program executed on a computer that has no OpenCL support is able to use other computing resources distributed on network in parallel. Results show that OpenCL platform model can be extended to network scale to provide a vendor, architecture and operating system independent and a parallel computing environment with reasonable communication overhead.

References
  1. "CUDA. " [Online]. Available: http://www. nvidia. com/cuda
  2. "BrookGPU" [Online]. Available: http://graphics. stanford. edu/projects/brookgpu/
  3. "Directcompute. " [Online]. Available: http://msdn. microsoft. com/directx
  4. "OpenCL - The open standard for parallel programming of heterogeneous systems" [Online]. Available: http://www. khronos. org/opencl/
  5. "The OpenCL Specification" [Online]. Available: http://www. khronos. org/registry/cl/specs/opencl-1. 1. pdf
  6. "JsonRpc-Cpp - OpenSource JSON-RPC implementation for C++" [Online]. Available: http://jsonrpc-cpp. sourceforge. net/
  7. "NVIDIA GPU Models" [Online]. Available: http://www. geforce. com/hardware/notebook-gpus
  8. "ATI GPU Models" [Online]. Available: http://www. amd. com/us/products/notebook/graphics/Pages/notebook-graphics. aspx
  9. A. Barak, T. Ben-Nun, E. Levy and A. Shiloh "A Package for OpenCL Based Heterogeneous Computing on Clusters with Many GPU Devices" in IEEE International Conference on Cluster Computing.
  10. "The Virtual OpenCL (VCL) Cluster Platform" [Online]. Available: http://www. mosix. org/vcl/VCL_wp. pdf
  11. "rCUDA. " [Online]. Available: http://www. rcuda. net/.
  12. Ryo Aoki, Shuichi Oikava, Ryoji Tsuchiyama, Takashi Nakamura "Improving Hybrid OpenCL Performance by High Speed Networks" 2010 First IEEE international Conference on Networking and Computing
  13. "CLuMPI. " [Online]. Available: http://clumpi. sourceforge. net
  14. V. Kindratenko et al. "GPU Clusters for High-Performance Computing"
Index Terms

Computer Science
Information Sciences

Keywords

Opencl Gpu Gpgpu Hpc Gpu Clusters