BUILDING A PRIVATE HPC CLOUD FOR COMPUTE AND DATA-INTENSIVE APPLICATIONS
OpenCL Standard for Parallel Programming
1. OpenCL- The Open Standard for
Parallel Programming of
Heterogeneous systems
1
DEPT. OF IT - CNW 08/17/14
Judyfer.W.S, 511309
M.E, Communication and Networking
St.Xaviers Catholic College of Engineering, Nagercoil
2. OVERVIEW
2
DEPT. OF IT - CNW 08/17/14
Science
Computing
Serial Computing
Parallel Computing
Why Parallel Computing?
Parallelism- Future of Computing
Why Heterogeneous Computing?
What is OpenCL?
OpenCL Objectives
Anatomy of OpenCL
OpenCL Architecture
Simple Program
4. Computing
The ACM Computing Curricula 2005 defined "computing" as follows:
"In a general way, we can define computing to mean any
goal-oriented activity requiring, benefiting from, or creating
computers. Thus, computing includes designing and building
hardware and software systems for a wide range of purposes;
processing, structuring, and managing various kinds of
information; doing scientific studies using computers; making
computer systems behave intelligently; creating and using
communications and entertainment media; finding and gathering
information relevant to any particular purpose, and so on. The list
is virtually endless, and the possibilities are vast."
08/17/14DEPT. OF IT - CNW
4
8. Why Parallel Computing?
Real World Problems
Save Time and Money
Solve larger and more complex problems
Provide Concurrency
Take advantage of Non-local Resources
Make better use of underlying parallel hardware
08/17/14DEPT. OF IT - CNW
8
9. Parallelism- Future of Computing
The race is already on for Exascale Computing!
Exaflop = 1018
calculations per second
08/17/14DEPT. OF IT - CNW
9
10. Why heterogeneous computing?
Increased computational power
does come from parallelism with multiple CPUs and
programmable GPUs
08/17/14DEPT. OF IT - CNW
10
11. OpenCL
Open
Royalty-free
OpenCL 1.0- Dec8,2008- The Khronos Group
Low level language
Supports both data parallel and task parallel programming
modes.
08/17/14DEPT. OF IT - CNW
11
12. OpenCL objectives
1. To grow the market of Parallel Computing.
2. Cross-vendor software portability.
3. Support for wide diversity of application.
4. Rapid deployment in the market.
08/17/14DEPT. OF IT - CNW
12
13. Anatomy of OpenCL
The OpenCL development framework is made up of
three parts:
1. Language specification
2. Platform layer API
3. Runtime API
08/17/14DEPT. OF IT - CNW
13