Processing-in-Memory for AI: From Circuits to Systems

This book provides an introduction to processing-in-memory (PIM) technology, discussing its architectures, circuits, and potential applications in AI and big data systems. It highlights the challenges of AI hardware and presents feasible PIM solutions for various memory types.

Format: Hardback
Length: 165 pages
Publication date: 10 July 2022
Publisher: Springer Nature Switzerland AG

This book is a comprehensive guide to processing-in-memory (PIM) technology, covering everything from its architectural designs to the actual implementation of circuits on various memory types. It explores how PIM can be a viable computer architecture in the age of AI and big data, highlighting its potential benefits and challenges. The authors provide a thorough summary of the difficulties encountered in AI hardware systems, as well as the limitations and considerations surrounding PIM. They then outline approaches to derive system-level requirements for a practical and feasible PIM solution.

The presentation primarily focuses on viable PIM solutions that can be implemented and used in real-world systems. It covers architectures, circuits, and implementation cases for each major memory type, including SRAM, DRAM, and ReRAM. By delving into these details, the book aims to provide readers with a deep understanding of PIM technology and its potential applications in the field of AI and data processing.

Overall, this book is an invaluable resource for researchers, engineers, and practitioners interested in exploring the possibilities of PIM technology in the modern computing landscape. It offers a comprehensive and up-to-date overview of the field, covering both theoretical aspects and practical implementation considerations.

Weight: 436g
Dimension: 235 x 155 (mm)
ISBN-13: 9783030987800
Edition number: 1st ed. 2023