Spintronic Stochastic Dataflow Computing

reportActive / Technical Report | Accesssion Number: AD1229897 | Open PDF

Abstract:

This program addressed the memory bottleneck problem in traditional Von Neumann computing architecture. This particularly challenging problem limits advances in artificial intelligence applications because they have an insatiable need for memory. This effort focused on two novel approaches to overcome the bottleneck: new magnetic memory technology and a stochastic computing framework.

Author(s):
Suhail, Haris ; Yang, Jiyue ; He, Haoran ; Jacob, Vinod K. ; Graening, Alexander ; Gupta, Puneet ; Wang, Kang L. ; Pamarti, Sudhakar
Author Organization(s):
University of California Los Angeles
Funding Organization(s):
Air Force Research Laboratory, Sensors Directorate, Wright-Patterson AFB, OH
Document Type:
Technical Report/Final Report
Publication Date:
2024 Jun 12
Pagination:
73

Security Markings

DOCUMENT & CONTEXTUAL SUMMARY

Distribution Code:
A - Approved For Public Release
Distribution Statement: Public Release.
Copyright: Not Copyrighted

RECORD

Collection: TRECMS
Subject Terms
Keyword(s):
Von Neumann architectures, memory bottleneck, magnetic memory, stochastic computing framework, meram(magnoelectric random access memory), magnetic tunnel junction device
Subject Categories:
Mathematical and Computer Sciences; Mathematical and Computer Sciences
Creation Date:
2024 Jun 18
Update Date:
2024 Jun 18