OpenCL on CPU and GPU

Introduction

OpenCL is a standard which defines a framework, an API and a programming language for parallel computation on heterogeneous systems like desktop computers, high-performance computing servers as well as hand-held devices.

The standard is maintained by the Khronos Group and supported by a large consortium of industry leaders including Apple, Intel, AMD, NVIDIA and ARM. Influenced by NVIDIA’s CUDA from the GPU side and by OpenMP which originates from the classical CPU side, the open OpenCL standard is characterized by a formulation which is abstract enough to support both CPU and GPU computing resources.

This is an ambitious goal, since providing an abstract interface together with a peak performance is a challenging task. OpenCL employs a strict isolation of the computation work into fundamental units, the kernels. These kernels can be developed in the OpenCL C programming language, a subset of the C99 language, with some additional OpenCL specific keywords.

In general, these kernels are hardware independent and compiled by the OpenCL runtime when they are loaded. To be able to fully exploit the parallel execution of the kernel code, several kernel instances, the work items, are started to process a set of input values. The actual number of concurrently running work items is determined by the OpenCL system. How a concrete algorithm can be partitioned into work items has to be decided by the programmer.

Assigned exercise machines

UID               Host name                              IP address
068		rz-tesla0.rz.uni-karlsruhe.de	172.21.94.100
074		rz-tesla0.rz.uni-karlsruhe.de	172.21.94.100
075		rz-tesla0.rz.uni-karlsruhe.de	172.21.94.100
077		rz-tesla0.rz.uni-karlsruhe.de	172.21.94.100
082		rz-tesla0.rz.uni-karlsruhe.de	172.21.94.100
084		rz-tesla0.rz.uni-karlsruhe.de	172.21.94.100
085		rz-tesla0.rz.uni-karlsruhe.de	172.21.94.100
088		rz-tesla0.rz.uni-karlsruhe.de	172.21.94.100
089		rz-tesla0.rz.uni-karlsruhe.de	172.21.94.100
090		rz-tesla0.rz.uni-karlsruhe.de	172.21.94.100
102		rz-tesla0.rz.uni-karlsruhe.de	172.21.94.100
103		rz-tesla0.rz.uni-karlsruhe.de	172.21.94.100
104		rz-tesla0.rz.uni-karlsruhe.de	172.21.94.100
110		rz-tesla2.rz.uni-karlsruhe.de	172.21.94.102
113		rz-tesla2.rz.uni-karlsruhe.de	172.21.94.102
117		rz-tesla2.rz.uni-karlsruhe.de	172.21.94.102
121		rz-tesla2.rz.uni-karlsruhe.de	172.21.94.102
123		rz-tesla2.rz.uni-karlsruhe.de	172.21.94.102
125		rz-tesla2.rz.uni-karlsruhe.de	172.21.94.102
128		rz-tesla2.rz.uni-karlsruhe.de	172.21.94.102
012		rz-tesla2.rz.uni-karlsruhe.de	172.21.94.102
132		rz-tesla2.rz.uni-karlsruhe.de	172.21.94.102
134		rz-tesla2.rz.uni-karlsruhe.de	172.21.94.102
137		rz-tesla2.rz.uni-karlsruhe.de	172.21.94.102
139		rz-tesla2.rz.uni-karlsruhe.de	172.21.94.102

Reference Material

• Khronos Group OpenCL

  http://www.khronos.org/opencl/

• OpenCL 1.2 Quick Reference Card

  http://www.khronos.org/files/opencl-1-2-quick-reference-card.pdf

• OpenCL 1.2 Full Documentation

  http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/

• Intel SDK for OpenCL

  http://software.intel.com/en-us/articles/vcsource-tools-opencl-sdk/

• AMD OpenCL Zone

  http://developer.amd.com/zones/OpenCLZone/

• NVIDIA OpenCL

  http://www.nvidia.com/object/cuda_opencl_1.html