The Community for Technology Leaders
2014 IEEE International Conference On Cluster Computing (CLUSTER) (2014)
Madrid, Spain
Sept. 22, 2014 to Sept. 26, 2014
ISBN: 978-1-4799-5548-0
pp: 19-28
Jonathan Lifflander , Department of Computer Science, University of Illinois Urbana-Champaign, USA
Esteban Meneses , Center for Simulation and Modeling, University of Pittsburgh, USA
Harshitha Menon , Department of Computer Science, University of Illinois at Urbana-Champaign, USA
Phil Miller , Department of Computer Science, University of Illinois at Urbana-Champaign, USA
Sriram Krishnamoorthy , Advanced Computing, Mathematics, and Data Division, Pacific Northwest National Laboratory, USA
Laxmikant V. Kale , Department of Computer Science, University of Illinois at Urbana-Champaign, USA
ABSTRACT
Deterministic replay of a parallel application is commonly used for discovering bugs or to recover from a hard fault with message-logging fault tolerance. For message passing programs, a major source of overhead during forward execution is recording the order in which messages are sent and received. During replay, this ordering must be used to deterministically reproduce the execution. Previous work in replay algorithms often makes minimal assumptions about the programming model and application to maintain generality. However, in many applications, only a partial order must be recorded due to determinism intrinsic in the program, ordering constraints imposed by the execution model, and events that are commutative (their relative execution order during replay does not need to be reproduced exactly). In this paper, we present a novel algebraic framework for reasoning about the minimum dependencies required to represent the partial order for different orderings and interleavings. By exploiting this framework, we improve on an existing scalable message-logging fault tolerance scheme that uses a total order. The improved scheme scales to 131,072 cores on an IBM BlueGene/P with up to 2× lower overhead.
INDEX TERMS
Fault tolerance, Fault tolerant systems, Programming, Debugging, Program processors, Electronic mail, Benchmark testing,execution model, replay, partial-order dependencies, fault tolerance, message logging, determinism
CITATION
Jonathan Lifflander, Esteban Meneses, Harshitha Menon, Phil Miller, Sriram Krishnamoorthy, Laxmikant V. Kale, "Scalable replay with partial-order dependencies for message-logging fault tolerance", 2014 IEEE International Conference On Cluster Computing (CLUSTER), vol. 00, no. , pp. 19-28, 2014, doi:10.1109/CLUSTER.2014.6968739
90 ms
(Ver 3.3 (11022016))