skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: PaRSEC: A Software Framework for Performance and Productivity on Hybrid, Manycore Platforms

Abstract

As the era of computer architectures dominated by serial processors ends, the convergence of several unprecedented challenges suggests that closing the longstanding "application–architecture performance gap" will become more challenging than ever. To address this problem, the Parallel Runtime Scheduling and Execution Control (PaRSEC) project created a modular software framework that achieved two major objectives: first, it built a task-based runtime capable of delivering portable performance to a wide range of science and engineering applications at all levels of the platform pyramid, including the upcoming 100 Pflop/s systems and then exascale; and second, it supported and facilitated the work of developers in migrating their legacy codes and writing entirely new ones for the emerging hybrid and massively parallel manycore processor system designs. PaRSEC will support multiple domain-specific languages capable of increasing the developers' productivity while also providing the runtime with the constructs and flexibility necessary to exploit the maximal parallelism from parallel applications. Extensive preliminary research in dense linear algebra showed convincingly that a parameterized task graph representation that symbolically describes the algorithm content can achieve the project's twofold objective within that domain. The research also strongly suggested that this powerful method could be generalized to a far-wider variety of applications.

Authors:
 [1]
  1. Univ. of Tennessee, Knoxville, TN (United States)
Publication Date:
Research Org.:
Univ. of Tennessee, Knoxville, TN (United States)
Sponsoring Org.:
USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR) (SC-21)
OSTI Identifier:
1348914
Report Number(s):
DOE-UTK-10682
DOE Contract Number:  
SC0010682
Resource Type:
Technical Report
Country of Publication:
United States
Language:
English
Subject:
97 MATHEMATICS AND COMPUTING; task-based runtime; parallel runtime scheduling and execution control; application-architecture performance gap

Citation Formats

Dongarra, Jack. PaRSEC: A Software Framework for Performance and Productivity on Hybrid, Manycore Platforms. United States: N. p., 2016. Web. doi:10.2172/1348914.
Dongarra, Jack. PaRSEC: A Software Framework for Performance and Productivity on Hybrid, Manycore Platforms. United States. doi:10.2172/1348914.
Dongarra, Jack. Thu . "PaRSEC: A Software Framework for Performance and Productivity on Hybrid, Manycore Platforms". United States. doi:10.2172/1348914. https://www.osti.gov/servlets/purl/1348914.
@article{osti_1348914,
title = {PaRSEC: A Software Framework for Performance and Productivity on Hybrid, Manycore Platforms},
author = {Dongarra, Jack},
abstractNote = {As the era of computer architectures dominated by serial processors ends, the convergence of several unprecedented challenges suggests that closing the longstanding "application–architecture performance gap" will become more challenging than ever. To address this problem, the Parallel Runtime Scheduling and Execution Control (PaRSEC) project created a modular software framework that achieved two major objectives: first, it built a task-based runtime capable of delivering portable performance to a wide range of science and engineering applications at all levels of the platform pyramid, including the upcoming 100 Pflop/s systems and then exascale; and second, it supported and facilitated the work of developers in migrating their legacy codes and writing entirely new ones for the emerging hybrid and massively parallel manycore processor system designs. PaRSEC will support multiple domain-specific languages capable of increasing the developers' productivity while also providing the runtime with the constructs and flexibility necessary to exploit the maximal parallelism from parallel applications. Extensive preliminary research in dense linear algebra showed convincingly that a parameterized task graph representation that symbolically describes the algorithm content can achieve the project's twofold objective within that domain. The research also strongly suggested that this powerful method could be generalized to a far-wider variety of applications.},
doi = {10.2172/1348914},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Thu Jun 30 00:00:00 EDT 2016},
month = {Thu Jun 30 00:00:00 EDT 2016}
}

Technical Report:

Save / Share: