Dynamically managed data for CPU-GPU architectures

Thomas B. Jablin, James A. Jablin, Prakash Prabhu, Feng Liu, David I. August

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

59 Scopus citations

Abstract

GPUs are flexible parallel processors capable of accelerating real applications. To exploit them, programmers must ensure a consistent program state between the CPU and GPU memories by managing data. Manually managing data is tedious and error-prone. In prior work on automatic CPU-GPU data management, alias analysis quality limits performance, and type-inference quality limits applicability. This paper presents Dynamically Managed Data (DyManD), the first automatic system to manage complex and recursive data-structures without static analyses. By replacing static analyses with a dynamic run-time system, DyManD overcomes the performance limitations of alias analysis and enables management for complex and recursive data-structures. DyManD-enabled GPU parallelization matches the performance of prior work equipped with perfectly precise alias analysis for 27 programs and demonstrates improved applicability on programs not previously managed automatically.

Original languageEnglish (US)
Title of host publicationProceedings - International Symposium on Code Generation and Optimization, CGO 2012
Pages165-174
Number of pages10
DOIs
StatePublished - Jul 9 2012
Event10th International Symposium on Code Generation and Optimization, CGO 2012 - San Jose, CA, United States
Duration: Mar 31 2012Apr 4 2012

Publication series

NameProceedings - International Symposium on Code Generation and Optimization, CGO 2012

Other

Other10th International Symposium on Code Generation and Optimization, CGO 2012
CountryUnited States
CitySan Jose, CA
Period3/31/124/4/12

All Science Journal Classification (ASJC) codes

  • Software

Fingerprint Dive into the research topics of 'Dynamically managed data for CPU-GPU architectures'. Together they form a unique fingerprint.

  • Cite this

    Jablin, T. B., Jablin, J. A., Prabhu, P., Liu, F., & August, D. I. (2012). Dynamically managed data for CPU-GPU architectures. In Proceedings - International Symposium on Code Generation and Optimization, CGO 2012 (pp. 165-174). (Proceedings - International Symposium on Code Generation and Optimization, CGO 2012). https://doi.org/10.1145/2259016.2259038