Stash: Have your scratchpad and cache it too

Rakesh Komuravelli, Matthew D. Sinclair, Johnathan Alsop, Muhammad Huzaifa, Maria Kotsifakou, Prakalp Srivastava, Sarita V Adve, Vikram Sadanand Adve

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Heterogeneous systems employ specialization for energy efficiency. Since data movement is expected to be a dominant consumer of energy, these systems employ specialized memories (e.g., scratchpads and FIFOs) for better efficiency for targeted data. These memory structures, however, tend to exist in local address spaces, incurring significant performance and energy penalties due to inefficient data movement between the global and private spaces. We propose an efficient heterogeneous memory system where specialized memory components are tightly coupled in a unified and coherent address space. This paper applies these ideas to a system with CPUs and GPUs with scratchpads and caches. We introduce a new memory organization, stash, that combines the benefits of caches and scratchpads without incurring their downsides. Like a scratchpad, the stash is directly addressed (without tags and TLB accesses) and provides compact storage. Like a cache, the stash is globally addressable and visible, providing implicit data movement and increased data reuse. We show that the stash provides better performance and energy than a cache and a scratchpad, while enabling new use cases for heterogeneous systems. For 4 microbenchmarks, which exploit new use cases (e.g., reuse across GPU compute kernels), compared to scratchpads and caches, the stash reduces execution cycles by an average of 27% and 13% respectively and energy by an average of 53% and 35%. For 7 current GPU applications, which are not designed to exploit the new features of the stash, compared to scratchpads and caches, the stash reduces cycles by 10% and 12% on average (max 22% and 31%) respectively, and energy by 16% and 32% on average (max 30% and 51%).

Original languageEnglish (US)
Title of host publicationISCA 2015 - 42nd Annual International Symposium on Computer Architecture, Conference Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages707-719
Number of pages13
ISBN (Electronic)9781450334020
DOIs
StatePublished - Jun 13 2015
Event42nd Annual International Symposium on Computer Architecture, ISCA 2015 - Portland, United States
Duration: Jun 13 2015Jun 17 2015

Publication series

NameProceedings - International Symposium on Computer Architecture
Volume13-17-June-2015
ISSN (Print)1063-6897

Other

Other42nd Annual International Symposium on Computer Architecture, ISCA 2015
CountryUnited States
CityPortland
Period6/13/156/17/15

ASJC Scopus subject areas

  • Hardware and Architecture

Fingerprint Dive into the research topics of 'Stash: Have your scratchpad and cache it too'. Together they form a unique fingerprint.

Cite this