Skip to main navigation Skip to search Skip to main content

Toward Energy-Efficient Sparse Matrix-Vector Multiplication with near STT-MRAM Computing Architecture

  • Beihang University
  • Southeast University, Nanjing
  • Vimicro Corporation
  • Truth Memory Coporation

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Sparse Matrix-Vector Multiplication (SpMV) is one of the vital computational primitives used in modern workloads. SpMV performs memory access, leading to unnecessary data transmission, massive data access, and redundant multiplicative accumulators. Therefore, we propose the near spin-transfer torque magnetic random access memory (STT-MRAM) processing architecture from three optimization perspectives. These optimizations include (1) the NMP controller receives the instruction through the AXI4 bus to implement the SpMV operation in the following steps, identifies valid data, and encodes the index depending on the kernel size, (2) the NMP controller uses high-level synthesis dataflow in the shared buffer for achieving better performance throughput while do not consume bus bandwidth, and (3) the configurable MACs are implemented in the NMP core without matching step entirely during the multiplication. Using these optimizations, the NMP architecture can access the pipelined STT-MRAM (read bandwidth is 26.7GB/s). The experimental simulation results show that this design achieves up to 66x and 28x speedup compared with state-of-the-art ones and 69x speedup without sparse optimization.

Original languageEnglish
Title of host publicationASP-DAC 2023 - 28th Asia and South Pacific Design Automation Conference, Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages222-227
Number of pages6
ISBN (Electronic)9781450397834
DOIs
StatePublished - 16 Jan 2023
Event28th Asia and South Pacific Design Automation Conference, ASP-DAC 2023 - Tokyo, Japan
Duration: 16 Jan 202319 Jan 2023

Publication series

NameProceedings of the Asia and South Pacific Design Automation Conference, ASP-DAC

Conference

Conference28th Asia and South Pacific Design Automation Conference, ASP-DAC 2023
Country/TerritoryJapan
CityTokyo
Period16/01/2319/01/23

Keywords

  • Energy Efficient
  • Near Memory Processing
  • STT-MRAM
  • SpMV

Fingerprint

Dive into the research topics of 'Toward Energy-Efficient Sparse Matrix-Vector Multiplication with near STT-MRAM Computing Architecture'. Together they form a unique fingerprint.

Cite this