An Overview of Computation-in-Memory (CIM) Architectures
摘要
This chapter presents a comprehensive classification of Computing-In-Memory (CIM) architectures based on three criteria, namely, computation location, level of parallelism, and used memory technology. This classification does not only provide an overview of existing CIM architectures with their pros and cons but also unifies the terminology that uniquely identifies these architectures. The chapter also discusses the steps in CIM design flow from the application level all the way down to the device technology and show the importance of system level to device technology level co-optimization.