GPU and CPU Double Precision Performance

After checking some data from wikipedia, vendor's specifications and so on, I made this chart above. The chart is about GPU and CPU performance in double precision. While the Intel Core I7 980X (extreme edition) gives us around 110GFLOPS (Source: Tom's Hardware), GPUs such as AMD Radeon 6970 and NVidia C2090 offer more than 660GFLOPS. Obviously these benchmarks represent peak performance under specific conditions for each platform.

OpenCL Books

OpenCL Programming Guide

• By: Aaftab Munshi; Benedict Gaster; Timothy G. Mattson; James Fung; Dan Ginsburg

• Publisher: Addison-Wesley Professional

Chapter 1. An Introduction to OpenCL
Chapter 2. HelloWorld: An OpenCL Example

Chapter 3. Platforms, Contexts, and Devices

Chapter 4. Programming with OpenCL C

Chapter 5. OpenCL C Built-in Functions

Chapter 6. Programs and Kernels

Chapter 7. Buffers and Sub-Buffers

Chapter 8. Images and Samplers

Chapter 9. Events

Chapter 10. Interoperability with OpenGL

Chapter 11. Interoperability with Direct3D

Chapter 12. C++ Wrapper API

Chapter 13. OpenCL Embedded Profile

Chapter 14. Image Histogram

Chapter 15. Sobel Edge Detection Filter

Chapter 16. Parallelizing Dikjstra’s Single Source Shortest Path Graph Algorithm

Chapter 17. Cloth Simulation in the Bullet Physics SDK

Chapter 18. Simulating the Ocean with Fast Fourier Transform

Chapter 19. Optical Flow

Chapter 20. Using OpenCL with PyOpenCL

Chapter 21. Matrix Multiplication with OpenCL

Chapter 22. Sparse Matrix-Vector Multiplication

Appendix A. Summary of OpenCL 1.1

The OpenCL Programming Book

Publisher: Fixstars Corporation Author: Fixstars Corporation (Ryoji Tsuchiyama, Takashi Nakamura, Takuro Iizuka, Akihiro Asahara, Satoshi Miki)

Introduction to Parallelization

Why Parallell

Parallel Computing (Hardware)

Parallel Computing (Software)

Conclusion

OpenCL

What is OpenCL?

Historical Background

An Overview of OpenCL

Why OpenCL?

Applicable Platforms

OpenCL Setup

Available OpenCL Environments

Developing Environment Setup

First OpenCL Program

Basic OpenCL

Basic Program Flow

Online/Offline Compilation

Calling the Kernel

Advanced OpenCL

OpenCL C

OpenCL Programming Practice

Case Study

FFT (Fast Fourier Transform)

Mersenne Twister

Notes

Heterogeneous Computing with OpenCL

Benedict Gaster, Lee Howes, David R. Kaeli, Perhaad Mistry, Dana Schaa

1. Introduction to Parallel Programming
2. Introduction to OpenCL
3. OpenCL Device Architectures
4. Basic OpenCL Examples
5. Understanding OpenCL's Concurrency and Execution Model
6. Dissecting a CPU/GPU OpenCL Implementation
7. OpenCL Case Study: Convolution
8. OpenCL Case Study: Video Processing
9. OpenCL Case Study: Histogram
10. OpenCL Case Study: Mixed Particle Simulation
11. OpenCL Extensions
12. OpenCL Profiling and Debugging
13. WebCL

AMD Radeon™ E6760 Embedded GPU OpenCL Compliant

AMD has released the first embedded GPU offering suport for OpenCL.

More information:

http://www.amd.com/us/press-releases/Pages/embedded-gpu-2011may02.aspx

http://www.amd.com/us/Documents/49807_E6760_GPU_brief.pdf

Why does NVidia not do the same for Tegra?

Anyway, let's create OpenCL HP applications for embedded systems.

Anjuta Project Wizards for AMD, NVidia and Intel OpenCL SDK

Aiming at increasing the OpenCL developing, I created some wizards to start up an OpenCL application project using the SDK from NVidia, AMD or Intel. I've used Anjuta DevStudio on Linux.

This is just a first approach, so don't be disappointed if you need to do some changes. The wizards give you a simple functional code and you can work on this one.

System requirements (they depend on your goal):

• NVidia: GPU Computing SDK code samples, CUDA Toolkit (http://developer.nvidia.com)
  
• AMD APP SDK (http://developer.amd.com)
• INTEL OpenCL SDK (http://software.intel.com/en-us/articles/opencl-sdk)
  
• Anjuta DevStudio 2.30+

OpenCL Group in Linkedin

I find important sharing information among people using OpenCL. Recently, I joined to OpenCL group in Linkedin. Some discussions and points of view that you'll find there:

1. the lack of domain specific libraries compared to CUDA, for instance. However, AMD has released stuff like http://developer.amd.com/libraries/appmathlibs/Pages/default.aspx. Anyway, we should have always in mind that OpenCL is younger than CUDA.
2. questions about why use OpenCL in environments without GPUs were answered by people arguing about the unknown user space (where your app will run) and stuff like http://www.khronos.org/developers/library/2010_siggraph_bof_opencl/OpenCL-BOF-Intel-SIGGRAPH-Jul10.pdf that points the direction of applications implemented on Intel's SDK.

The group is open and some helpful and qualified helpful members are creating interesting discussions.

Parallel Primitives Library

Lets try to improve OpenCL libraries for everything...

Libs like this:

clpp is an OpenCL Data Parallel Primitives Library. It is a library of data-parallel algorithm primitives such as parallel-prefix-sum ("scan"), parallel sort and parallel reduction. Primitives such as these are important building blocks for a wide variety of data-parallel algorithms, including sorting, stream compaction, and building data structures such as trees and summed-area tables.

If you want to join the project, please simply send a message to our mailing list: http://groups.google.com/group/cl-pp

Thanks to Polar Lights?? initiative. ;)

Profiling your application

I was away for a time, but I'm coming back. In my recent projects, I had taken a few results from profiling tools. For a better understanding, I grabbed some resources from Marcus Bannerman's website. Putting the link...

Profiling in Linux

I didn't ask no authorization to publish it in this blog, but I'm sure that he will not be upset about that.

Take a look at other articles from his website. There are a lot of stuff about OpenCL http://www.marcusbannerman.co.uk