Skip to main content

Dynamically managed data for CPU-GPU architectures

Author(s): Jablin, Thomas B; Jablin, James A; Prabhu, Prakash; Liu, Feng; August, David I

To refer to this page use:
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.
Publication Date: Mar-2012
Citation: Jablin, Thomas B., James A. Jablin, Prakash Prabhu, Feng Liu, and David I. August. "Dynamically managed data for CPU-GPU architectures." Proceedings of the Tenth International Symposium on Code Generation and Optimization (2012): pp. 165-174. doi:10.1145/2259016.2259038
DOI: 10.1145/2259016.2259038
ISSN: 2164-2397
Pages: 165 - 174
Type of Material: Conference Article
Journal/Proceeding Title: Proceedings of the Tenth International Symposium on Code Generation and Optimization
Version: Author's manuscript

Items in OAR@Princeton are protected by copyright, with all rights reserved, unless otherwise indicated.