Print Email Facebook Twitter Efficient organization of digital periphery to support integer datatype for memristor-based cim Title Efficient organization of digital periphery to support integer datatype for memristor-based cim Author Zahedi, M.Z. (TU Delft Computer Engineering) Mayahinia, M. (TU Delft Computer Engineering) Abu Lebdeh, M.F.M. (TU Delft Computer Engineering) Wong, J.S.S.M. (TU Delft Computer Engineering) Hamdioui, S. (TU Delft Quantum & Computer Engineering) Contributor O'Conner, L. (editor) Department Quantum & Computer Engineering Date 2020 Abstract Von Neumann-based architectures suffer from costly communication between CPU and memory. This communication imposes several orders of magnitude more power and performance overheads compared to the arithmetic operations performed by the processor. This overhead becomes critical for applications that require processing a large amount of data. Computation-in-Memory (CIM) leveraging memristor devices in the crossbar structure offers a potential solution to tackle this challenge. However, support for the integer data type is lacking in CIM approaches as most solutions operate on a single/few bits only. This paper proposes a new organization of the periphery (next to memristor crossbar) to compute matrix-matrix multiplication (MMM) at the tile level. More precisely, the analog additions performed in the crossbar is complemented with additions performed in the digital periphery. In this mixed analog-digital system, digital additions are performed in a way that only the minimum size of adders are required-this is to reduce the latency of the digital periphery as much as possible. In addition, the design is customized to the number of ADCs as well as datatype sizes to support different possible scenarios. The results show that our organization reduces energy and latency up to 50x and 3x, respectively, compared to the reference design. To reference this document use: http://resolver.tudelft.nl/uuid:1fb95514-5b5d-457a-be3e-65b7803f1951 DOI https://doi.org/10.1109/ISVLSI49217.2020.00047 Publisher IEEE, Piscataway ISBN 978-1-7281-5776-4 Source 2020 IEEE Computer Society Annual Symposium on VLSI (ISVLSI): Proceedings Event 19th IEEE Computer Society Annual Symposium on VLSI, ISVLSI 2020, 2020-07-06 → 2020-07-08, Limassol, Cyprus Series 2020 IEEE COMPUTER SOCIETY ANNUAL SYMPOSIUM ON VLSI (ISVLSI 2020), 2159-3469 Bibliographical note Accepted author manuscript Part of collection Institutional Repository Document type conference paper Rights © 2020 M.Z. Zahedi, M. Mayahinia, M.F.M. Abu Lebdeh, J.S.S.M. Wong, S. Hamdioui Files PDF Efficient_organization_of ... ed_CIM.pdf 2.6 MB Close viewer /islandora/object/uuid:1fb95514-5b5d-457a-be3e-65b7803f1951/datastream/OBJ/view