LA-LLC: Inter-Core Locality-Aware Last-Level Cache to Exploit Many-to-Many Traffic in GPGPUs
Sign Up to like & getrecommendations! Published in 2017 at "IEEE Computer Architecture Letters"
DOI: 10.1109/lca.2016.2611663
Abstract: The reply network is a severe performance bottleneck in General Purpose Graphic Processing Units (GPGPUs), as the communication path from memory controllers (MC) to cores is often congested. In this paper, we find that instead… read more here.
Keywords: inter core; many traffic; core locality; core ... See more keywords