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

Title: A High-Performance Event Service for HPC Applications

Abstract

Event services based on publish-subscribe architectures are well established components of distributed computing applications. Recently, an event service has been proposed as part of the Common Component Architecture (CCA) for high-performance computing applications. In this paper we describe our experiences investigating implementation options for the CCA event service that exploit interprocess communications mechanisms commonly used on HPC platforms. The aim of our work is to create an event service that supports the well-known engineering advantages of publish-subscribe architecture, and provides performance levels approaching those achievable using more primitive message-passing mechanisms such as MPI.

Authors:
; ; ;
Publication Date:
Research Org.:
Pacific Northwest National Lab. (PNNL), Richland, WA (United States)
Sponsoring Org.:
USDOE
OSTI Identifier:
912717
Report Number(s):
PNNL-SA-53530
KJ0101030; TRN: US200801%%987
DOE Contract Number:
AC05-76RL01830
Resource Type:
Conference
Resource Relation:
Conference: Proceedings of the 3rd International Workshop on Software Engineering for High Performance Computing Applications, 5 pages
Country of Publication:
United States
Language:
English
Subject:
99 GENERAL AND MISCELLANEOUS//MATHEMATICS, COMPUTING, AND INFORMATION SCIENCE; COMPUTER ARCHITECTURE; COMMUNICATIONS; IMPLEMENTATION; PERFORMANCE; COMPUTER NETWORKS; DATA TRANSMISSION; Common Component Architecture, event service, Aggregate Remote Memory Copy Interface (ARMCI)

Citation Formats

Gorton, Ian, Chavarría-Miranda, Daniel, Krishnan, Manoj Kumar, and Nieplocha, Jarek. A High-Performance Event Service for HPC Applications. United States: N. p., 2007. Web. doi:10.1109/SE-HPC.2007.1.
Gorton, Ian, Chavarría-Miranda, Daniel, Krishnan, Manoj Kumar, & Nieplocha, Jarek. A High-Performance Event Service for HPC Applications. United States. doi:10.1109/SE-HPC.2007.1.
Gorton, Ian, Chavarría-Miranda, Daniel, Krishnan, Manoj Kumar, and Nieplocha, Jarek. Sun . "A High-Performance Event Service for HPC Applications". United States. doi:10.1109/SE-HPC.2007.1.
@article{osti_912717,
title = {A High-Performance Event Service for HPC Applications},
author = {Gorton, Ian and Chavarría-Miranda, Daniel and Krishnan, Manoj Kumar and Nieplocha, Jarek},
abstractNote = {Event services based on publish-subscribe architectures are well established components of distributed computing applications. Recently, an event service has been proposed as part of the Common Component Architecture (CCA) for high-performance computing applications. In this paper we describe our experiences investigating implementation options for the CCA event service that exploit interprocess communications mechanisms commonly used on HPC platforms. The aim of our work is to create an event service that supports the well-known engineering advantages of publish-subscribe architecture, and provides performance levels approaching those achievable using more primitive message-passing mechanisms such as MPI.},
doi = {10.1109/SE-HPC.2007.1},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Sun May 20 00:00:00 EDT 2007},
month = {Sun May 20 00:00:00 EDT 2007}
}

Conference:
Other availability
Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share:
  • The pipeline software architecture pattern is commonly used in many application domains to structure a software system. A pipeline comprises a sequence of processing steps that progressively transform data to some desired outputs. As pipeline-based systems are required to handle increasingly large volumes of data and provide high throughput services, simple scripting-based technologies that have traditionally been used for constructing pipelines do not scale. In this paper we describe the MeDICI Integration Framework (MIF), which is specifically designed for building flexible, efficient and scalable pipelines that exploit distributed services as elements of the pipeline. We explain the core runtime andmore » development infrastructures that MIF provides, and demonstrate how MIF has been used in two complex applications to improve performance and modifiability.« less
  • Preparing codes for next generation supercomputer systems is anticipated to require significant changes to the optimization strategies employed in established HPC applications. In this paper, we present our experience in applying term-rewriting transformations for the optimization of such applications. We have designed application-specific term-rewriting transformations to improve the scalability, enhance locality and reduce the communication overhead of the Self-Consistent Field (SCF) computational chemistry benchmark. We present the rationale for the use of term-rewriting in this manner, the design of our transformations, and the much enhanced performance of the resulting code.
  • Abstract not provided.
  • Accelerators are used in about 13% of the current Top500 List. Supercomputers leveraging accelerators grew by a factor of 2.2x in 2012 and are expected to completely dominate the Top500 by 2015. Though most of these deployments use NVIDIA GPGPU accelerators, Intel’s Xeon Phi architecture will likely grow in popularity in the coming years. Unfortunately, there are few studies analyzing the performance and energy efficiency of systems leveraging the Intel Xeon Phi. We extend our systemic measurement methodology to isolate system power by component including accelerators. We use this methodology to present a detailed study of the performance-energy tradeoffs ofmore » the Xeon Phi architecture.« less
  • Abstract not provided.