Ascend AI Processor Architecture and Programming: Principles and Applications of CANN - Softcover

Liang, Xiaoyao

Softcover

ISBN 10: 0128234881 ISBN 13: 9780128234884

Publisher: Elsevier, 2020

Synopsis

Ascend AI Processor Architecture and Programming: Principles and Applications of CANN offers in-depth AI applications using Huawei’s Ascend chip, presenting and analyzing the unique performance and attributes of this processor. The title introduces the fundamental theory of AI, the software and hardware architecture of the Ascend AI processor, related tools and programming technology, and typical application cases. It demonstrates internal software and hardware design principles, system tools and programming techniques for the processor, laying out the elements of AI programming technology needed by researchers developing AI applications.

Chapters cover the theoretical fundamentals of AI and deep learning, the state of the industry, including the current state of Neural Network Processors, deep learning frameworks, and a deep learning compilation framework, the hardware architecture of the Ascend AI processor, programming methods and practices for developing the processor, and finally, detailed case studies on data and algorithms for AI.

• Presents the performance and attributes of the Huawei Ascend AI processor
• Describes the software and hardware architecture of the Ascend processor
• Lays out the elements of AI theory, processor architecture, and AI applications
• Provides detailed case studies on data and algorithms for AI
• Offers insights into processor architecture and programming to spark new AI applications

About this title

About the Author

Liang Xiaoyao is a Professor, Supervisor and Academic Leader in the Department of Computer Science and Engineering at Shanghai Jiaotong University, in China. He graduated from Harvard University in the USA, and has held roles at a number of international companies. His research interests include computer architecture, integrated circuit design, general graphics processors, and artificial intelligence chip architecture. He has published over 80 papers, and is a leader in the field.

From the Back Cover

Artificial intelligence (AI) research is rapidly developing, with AI applications predicted to profoundly affect all areas of industry in the near future. Consequently, there is an ever-growing demand for processor chips that support AI research and applications. Currently, chips offering this support are limited to GPUs, TPUs, and the Huawei Ascend AI chip, with limited detailed information on these processors.

The Ascend AI chip is a powerful and versatile AI processor that will see extensive deployment across the globe. AI Processor Architecture and Programming offers in-depth AI applications using Huawei’s Ascend chip, presenting and analyzing the unique performance and attributes of this processor. The title introduces the fundamental theory of AI, and the software and hardware architecture of the Ascend AI processor, as well as related tools and programming technology, and typical application cases. The book demonstrates internal software and hardware design principles, system tools and programming techniques for the processor. It also lays out those elements of AI programming technology needed by researchers developing AI applications. The book consists in six chapters, covering the theoretical fundamentals of AI and deep learning; the state of the industry, including the current state of Neural Network Processors, deep learning frameworks, and a deep learning compilation framework; the hardware architecture of the Ascend AI processor; programming methods and practices for developing the processor, and finally; detailed case studies on data and algorithms for AI. This title presents the hardware and software architecture of the Huawei Ascend AI processor, supporting the future development and deployment of powerful AI applications.

Edition details

• Publisher: Elsevier
• Publication date: 2020
• Language: English
• ISBN 10: 0128234881
• ISBN 13: 9780128234884
• Binding: Paperback
• Edition number: 1
• Number of pages: 308