Cohesion: An Adaptive Hybrid Memory Model for Accelerators

Authors:
John H. Kelm;Daniel R. Johnson;William Tuohy;Steven S. Lumetta;Sanjay J. Patel
Affiliations:
University of Illinois at Urbana-Champaign;University of Illinois at Urbana-Champaign;University of Illinois at Urbana-Champaign;University of Illinois at Urbana-Champaign;University of Illinois at Urbana-Champaign
Venue:
IEEE Micro
Year:
2011

Citing 0
Cited 2

Why on-chip cache coherence is here to stay

Communications of the ACM
Hardware-software coherence protocol for the coexistence of caches and local memories

SC '12 Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis

Quantified Score

Hi-index	0.02

Visualization

Abstract

Cohesion is a hybrid memory model that enables fine-grained temporal data reassignment between hardware- and software-managed coherence domains, allowing systems to support both. Cohesion can dynamically adapt to the sharing needs of both applications and runtimes requiring neither copy operations nor multiple address spaces.