Talati N, Ben-Hur R, Wald N, Haj-Ali A, Prabahar JR, Kvatinsky S (2020)
Publication Language: English
Publication Type: Book chapter / Article in edited volumes
Publication year: 2020
Publisher: Springer Singapore
Edited Volumes: Applications of Emerging Memory Technology
Series: Springer Series in Advanced Microelectronics
Book Volume: 63
Pages Range: 191-213
ISBN: 978-981-13-8378-6
URI: https://link.springer.com/chapter/10.1007/978-981-13-8379-3_8
DOI: 10.1007/978-981-13-8379-3_8
Data transfer between processing and memory units in modern comput-
ing systems is their main performance and energy-efficiency bottleneck, commonly
known as the von Neumann bottleneck. Prior research attempts to alleviate the prob-
lem by moving the computing units closer to the memory that has had limited success
since data transfer is still required. In this chapter, we present mMPU memristive
memory processing unit, which relies on a memristive memory to perform computa-
tion using the memory cells, and therefore directly tackles the von Neumann bottle-
neck. In mMPU, the operation is controlled by a modified controller and peripheral
circuit without changing the structure of the memory cells and arrays. As the basic
logic element, we present Memristor-Aided loGIC (MAGIC), a technique to com-
pute logical functions using memristors within the memory array. We further show
how to extend basic MAGIC primitives to execute any arbitrary Boolean function
and demonstrate the microarchitecture of the memory. This process is required to enable data computing using MAGIC. Finally, we show how to build the computing
system using mMPU, which performs computation using MAGIC to enable a real
processing-in-memory machine.
APA:
Talati, N., Ben-Hur, R., Wald, N., Haj-Ali, A., Prabahar, J.R., & Kvatinsky, S. (2020). mMPU—A Real Processing-in-Memory Architecture to Combat the von Neumann Bottleneck. In Manan suri (Eds.), Applications of Emerging Memory Technology. (pp. 191-213). Springer Singapore.
MLA:
Talati, Nishil, et al. "mMPU—A Real Processing-in-Memory Architecture to Combat the von Neumann Bottleneck." Applications of Emerging Memory Technology. Ed. Manan suri, Springer Singapore, 2020. 191-213.
BibTeX: Download