Merge Network for a Non-Von Neumann Accumulate Accelerator in a 3D Chip - Anirudh Jain - ICRC 2018

267 views
Download
  • Share
Create Account or Sign In to post comments
#IEEE #ICRC #Rebooting Computing #2018 #conference #event #computing #technology #research #energy costs #computation #logicmemory #nonvon Neumann #network architecture #graph analytics #sparse data #energy efficiency

Anirudh Jain, Georgia Tech, presents his group’s research findings on tackling the issue of the high energy costs of moving data for computation. Considering logic-memory integration that helps mitigate the von Neumann bottleneck, Jain explains how this has enabled a new class of architectures that help accelerate graph analytics and operations on sparse data streams.

Anirudh Jain, Georgia Tech, presents his group’s research findings on tackling the issue of the high energy costs of moving data for computation. Considering logic-memory integration that helps mitigate the von Neumann bottleneck, Jain explains how this has enabled a new class of architectures that help accelerate graph analytics and operations on sparse data streams.

Advertisment

Advertisment