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

Title: Simulation-based Performance Analysis and Tuning for a Two-level Directly Connected System

Abstract

Hardware and software co-design is becoming increasingly important due to complexities in supercomputing architectures. Simulating applications before there is access to the real hardware can assist machine architects in making better design decisions that can optimize application performance. At the same time, the application and runtime can be optimized and tuned beforehand. BigSim is a simulation-based performance prediction framework designed for these purposes. It can be used to perform packet-level network simulations of parallel applications using existing parallel machines. In this paper, we demonstrate the utility of BigSim in analyzing and optimizing parallel application performance for future systems based on the PERCS network. We present simulation studies using benchmarks and real applications expected to run on future supercomputers. Future petascale systems will have more than 100,000 cores, and we present simulations at that scale.

Authors:
; ; ; ; ; ; ;
Publication Date:
Research Org.:
Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
Sponsoring Org.:
USDOE
OSTI Identifier:
1035952
Report Number(s):
LLNL-CONF-500821
TRN: US201205%%510
DOE Contract Number:  
W-7405-ENG-48
Resource Type:
Conference
Resource Relation:
Conference: Presented at: The 17th IEEE International Conference on Parallel and Distributed Systems, Tainan, Taiwan, Dec 07 - Dec 09, 2011
Country of Publication:
United States
Language:
English
Subject:
97 MATHEMATICAL METHODS AND COMPUTING; ARCHITECTS; BENCHMARKS; DESIGN; FORECASTING; PERFORMANCE; SIMULATION; SUPERCOMPUTERS; TUNING

Citation Formats

Totoni, E, Bhatele, A, Bohm, E J, Jain, N, Mendes, C L, Mokos, R M, Zheng, G, and Kale, L V. Simulation-based Performance Analysis and Tuning for a Two-level Directly Connected System. United States: N. p., 2011. Web.
Totoni, E, Bhatele, A, Bohm, E J, Jain, N, Mendes, C L, Mokos, R M, Zheng, G, & Kale, L V. Simulation-based Performance Analysis and Tuning for a Two-level Directly Connected System. United States.
Totoni, E, Bhatele, A, Bohm, E J, Jain, N, Mendes, C L, Mokos, R M, Zheng, G, and Kale, L V. Mon . "Simulation-based Performance Analysis and Tuning for a Two-level Directly Connected System". United States. https://www.osti.gov/servlets/purl/1035952.
@article{osti_1035952,
title = {Simulation-based Performance Analysis and Tuning for a Two-level Directly Connected System},
author = {Totoni, E and Bhatele, A and Bohm, E J and Jain, N and Mendes, C L and Mokos, R M and Zheng, G and Kale, L V},
abstractNote = {Hardware and software co-design is becoming increasingly important due to complexities in supercomputing architectures. Simulating applications before there is access to the real hardware can assist machine architects in making better design decisions that can optimize application performance. At the same time, the application and runtime can be optimized and tuned beforehand. BigSim is a simulation-based performance prediction framework designed for these purposes. It can be used to perform packet-level network simulations of parallel applications using existing parallel machines. In this paper, we demonstrate the utility of BigSim in analyzing and optimizing parallel application performance for future systems based on the PERCS network. We present simulation studies using benchmarks and real applications expected to run on future supercomputers. Future petascale systems will have more than 100,000 cores, and we present simulations at that scale.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {2011},
month = {9}
}

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: