skip to main content

SciTech ConnectSciTech Connect

Title: Locality-Driven Dynamic GPU Cache Bypassing

This paper presents novel cache optimizations for massively parallel, throughput-oriented architectures like GPUs. Based on the reuse characteristics of GPU workloads, we propose a design that integrates such efficient locality filtering capability into the decoupled tag store of the existing L1 D-cache through simple and cost-effective hardware extensions.
Authors:
; ; ; ; ;
Publication Date:
OSTI Identifier:
1194296
Report Number(s):
PNNL-SA-109271
KJ0402000
DOE Contract Number:
AC05-76RL01830
Resource Type:
Conference
Resource Relation:
Conference: Proceedings of the 29th ACM on International Conference on Supercomputing (ICS 2015), June 8-11, 2015, Newport Beach, California, 66-77
Publisher:
ACM , New York, NY, United States(US).
Research Org:
Pacific Northwest National Laboratory (PNNL), Richland, WA (US)
Sponsoring Org:
USDOE
Country of Publication:
United States
Language:
English
Subject:
architecture optimization; reuse; performance; energy; locality; cache