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

Title: Evaluation of Simple Causal Message Logging for Large-Scale Fault Tolerant HPC Systems

Abstract

The era of petascale computing brought machines with hundreds of thousands of processors. The next generation of exascale supercomputers will make available clusters with millions of processors. In those machines, mean time between failures will range from a few minutes to few tens of minutes, making the crash of a processor the common case, instead of a rarity. Parallel applications running on those large machines will need to simultaneously survive crashes and maintain high productivity. To achieve that, fault tolerance techniques will have to go beyond checkpoint/restart, which requires all processors to roll back in case of a failure. Incorporating some form of message logging will provide a framework where only a subset of processors are rolled back after a crash. In this paper, we discuss why a simple causal message logging protocol seems a promising alternative to provide fault tolerance in large supercomputers. As opposed to pessimistic message logging, it has low latency overhead, especially in collective communication operations. Besides, it saves messages when more than one thread is running per processor. Finally, we demonstrate that a simple causal message logging protocol has a faster recovery and a low performance penalty when compared to checkpoint/restart. Running NAS Parallel Benchmarksmore » (CG, MG and BT) on 1024 processors, simple causal message logging has a latency overhead below 5%.« less

Authors:
; ;
Publication Date:
Research Org.:
Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
Sponsoring Org.:
USDOE
OSTI Identifier:
1021072
Report Number(s):
LLNL-CONF-471680
TRN: US201116%%985
DOE Contract Number:  
W-7405-ENG-48
Resource Type:
Conference
Resource Relation:
Conference: Presented at: Dependable Parallel, Distributed and Network-. Centric System, Anchorage, AK, United States, May 16 - May 16, 2011
Country of Publication:
United States
Language:
English
Subject:
97 MATHEMATICS, COMPUTING, AND INFORMATION SCIENCE; BENCHMARKS; COMMUNICATIONS; EVALUATION; PERFORMANCE; PRODUCTIVITY; SUPERCOMPUTERS; TOLERANCE

Citation Formats

Bronevetsky, G, Meneses, E, and Kale, L V. Evaluation of Simple Causal Message Logging for Large-Scale Fault Tolerant HPC Systems. United States: N. p., 2011. Web.
Bronevetsky, G, Meneses, E, & Kale, L V. Evaluation of Simple Causal Message Logging for Large-Scale Fault Tolerant HPC Systems. United States.
Bronevetsky, G, Meneses, E, and Kale, L V. 2011. "Evaluation of Simple Causal Message Logging for Large-Scale Fault Tolerant HPC Systems". United States. https://www.osti.gov/servlets/purl/1021072.
@article{osti_1021072,
title = {Evaluation of Simple Causal Message Logging for Large-Scale Fault Tolerant HPC Systems},
author = {Bronevetsky, G and Meneses, E and Kale, L V},
abstractNote = {The era of petascale computing brought machines with hundreds of thousands of processors. The next generation of exascale supercomputers will make available clusters with millions of processors. In those machines, mean time between failures will range from a few minutes to few tens of minutes, making the crash of a processor the common case, instead of a rarity. Parallel applications running on those large machines will need to simultaneously survive crashes and maintain high productivity. To achieve that, fault tolerance techniques will have to go beyond checkpoint/restart, which requires all processors to roll back in case of a failure. Incorporating some form of message logging will provide a framework where only a subset of processors are rolled back after a crash. In this paper, we discuss why a simple causal message logging protocol seems a promising alternative to provide fault tolerance in large supercomputers. As opposed to pessimistic message logging, it has low latency overhead, especially in collective communication operations. Besides, it saves messages when more than one thread is running per processor. Finally, we demonstrate that a simple causal message logging protocol has a faster recovery and a low performance penalty when compared to checkpoint/restart. Running NAS Parallel Benchmarks (CG, MG and BT) on 1024 processors, simple causal message logging has a latency overhead below 5%.},
doi = {},
url = {https://www.osti.gov/biblio/1021072}, journal = {},
number = ,
volume = ,
place = {United States},
year = {Fri Feb 25 00:00:00 EST 2011},
month = {Fri Feb 25 00:00:00 EST 2011}
}

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: