Technical computing has become an integral part of research and development of new technologies in modern civilization. It is used in industry and academia equally to produce new products, produce weather forecasting, geosciences exploration, financial modeling, car crash simulation, electro-magnetic field propagation from mobile phones etc. However, current computational power limits the type of research needed to push the edge of understanding of these physical processes. Massively parallel processors like Intel® MIC architecture has been developed to increase the computational power to remove these research barriers. However, to exploit MIC co-processor capable of providing more than Teraflops of double precision floating point performance in Technical Computing applications will require careful design of algorithm, data structure and understanding to match the hardware capabilities and idiosyncrasies of the processor. This book is targeted to give the readers an in-depth look at the MIC processor architecture, parallel data structure and algorithm used by various Technical Computing applications that will be suitable for such a coprocessor. We shall also look at the various source code level optimizations that can be done to exploit some of the processor features.
What you’ll learn
- A look at the current incarnation of the Intel MIC architecture.
- How to get optimal performance out of code
- Designing data structures and algorithms to benefit from the computing power at hand
- How to design and port applications to any core architecture to gain the performance that MIC affords.
Who this book is for
This book will be beneficial to practicing scientist, engineers and software developers looking at porting and developing technical computing applications on Intel® MIC architecture specifically current incarnation KNC (Knights Corner). This book may also be beneficial to secondary audiences who are looking to learn parallel programming for massively parallel architecture.
Unless otherwise noted above, most orders ship within 1 to 2 days. We will promptly notify you if there is a stock problem with any items on your order and provide you with an estimated delivery date. If you have a firm need by date, please provide such information in the comment section at checkout.
Page Count (est.): 232
Pub Date: 9/1/2013