PredCom: A Predictive Approach to Collecting Approximated Communication Traces

Miwa, Shinobu; Laguna, Ignacio; Schulz, Martin

doi:10.1109/tpds.2020.3011121

Title: PredCom: A Predictive Approach to Collecting Approximated Communication Traces

Journal Article · Wed Jul 22 00:00:00 EDT 2020 · IEEE Transactions on Parallel and Distributed Systems

DOI:https://doi.org/10.1109/tpds.2020.3011121· OSTI ID:1769152

^[1];

^[2]; Schulz, Martin ^[3]

Univ. of Electro-Communications, Tokyo (Japan)
Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
Technical Univ. of Munich (Germany)

Communication traces collected from MPI applications are an important source of information for performance optimization as they can help analysts determine communication patterns and identify inefficiencies. However, their collection, especially at scale, is time consuming, since it usually requires running the complete target application on a large number of nodes. In this work, we present PredCom, a tool-chain to generate a predictive communication proxy based on information gathered from a few small scale runs, which allows us to extract approximate communication traces with an accuracy high enough for most analysis goals. For this, we combine LLVM passes on the original source code (to capture static program structure) with parameter prediction (to capture dynamic and scaling behavior). This approach drastically reduces the time needed for collecting the communication traces, even for traces on large numbers of MPI processes. Here, we demonstrate that PredCom generates communication traces of various applications up to 1612x faster with an accuracy loss of 0.11 on average compared to the original large-scale traces, and we show that the generated traces can be used to optimize process placement.

View Accepted Manuscript (DOE)

Cite

Export

Save

Research Organization:: Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)

Sponsoring Organization:: USDOE National Nuclear Security Administration (NNSA); Japan Science and Technology Agency (JST); Kayamori Foundation of Informational Science Advancement; Japan Society for the Promotion of Science (JSPS)

Grant/Contract Number:: AC52-07NA27344; JP20H04193

OSTI ID:: 1769152

Report Number(s):: LLNL-JRNL-748663; 933780

Journal Information:: IEEE Transactions on Parallel and Distributed Systems, Vol. 32, Issue 1; ISSN 1045-9219

Publisher:: IEEECopyright Statement

Country of Publication:: United States

Language:: English

References (33)

ScalaExtrap: trace-based communication extrapolation for spmd programs Wu, Xing; Mueller, Frank Proceedings of the 16th ACM symposium on Principles and practice of parallel programming - PPoPP '11 https://doi.org/10.1145/1941553.1941569	conference	January 2011
ScalaBenchGen: Auto-Generation of Communication Benchmarks Traces Wu, Xing; Deshpande, Vivek; Mueller, Frank 2012 IEEE International Symposium on Parallel & Distributed Processing (IPDPS), 2012 IEEE 26th International Parallel and Distributed Processing Symposium https://doi.org/10.1109/IPDPS.2012.114	conference	May 2012
The Tau Parallel Performance System Shende, Sameer S.; Malony, Allen D. The International Journal of High Performance Computing Applications, Vol. 20, Issue 2 https://doi.org/10.1177/1094342006064482	journal	May 2006
A compiler-based communication analysis approach for multiprocessor systems Proceedings 20th IEEE International Parallel & Distributed Processing Symposium https://doi.org/10.1109/IPDPS.2006.1639322	conference	January 2006
Transforming MPI source code based on communication patterns Preissl, Robert; Schulz, Martin; Kranzlmüller, Dieter Future Generation Computer Systems, Vol. 26, Issue 1 https://doi.org/10.1016/j.future.2009.05.017	journal	January 2010
Program Slicing Weiser, Mark IEEE Transactions on Software Engineering, Vol. SE-10, Issue 4 https://doi.org/10.1109/TSE.1984.5010248	journal	July 1984
Statistical scalability analysis of communication operations in distributed applications Vetter, Jeffrey S.; McCracken, Michael O. Proceedings of the eighth ACM SIGPLAN symposium on Principles and practices of parallel programming - PPoPP '01 https://doi.org/10.1145/379539.379590	conference	January 2001
Points-to analysis in almost linear time Steensgaard, Bjarne Proceedings of the 23rd ACM SIGPLAN-SIGACT symposium on Principles of programming languages - POPL '96 https://doi.org/10.1145/237721.237727	conference	January 1996
Exascaling Your Library Shudler, Sergei; Calotoiu, Alexandru; Hoefler, Torsten Proceedings of the 29th ACM on International Conference on Supercomputing https://doi.org/10.1145/2751205.2751216	conference	June 2015
Automatic generation of executable communication specifications from parallel applications Wu, Xing; Mueller, Frank; Pakin, Scott Proceedings of the international conference on Supercomputing - ICS '11 https://doi.org/10.1145/1995896.1995901	conference	January 2011
The Scalasca performance toolset architecture Geimer, Markus; Wolf, Felix; Wylie, Brian J. N. Concurrency and Computation: Practice and Experience https://doi.org/10.1002/cpe.1556	journal	January 2010
Flow-sensitive pointer analysis for millions of lines of code Hardekopf, Ben; Lin, Calvin International Symposium on Code Generation and Optimization (CGO 2011) https://doi.org/10.1109/CGO.2011.5764696	conference	April 2011
Interprocedural slicing using dependence graphs Horwitz, Susan; Reps, Thomas; Binkley, David ACM Transactions on Programming Languages and Systems, Vol. 12, Issue 1 https://doi.org/10.1145/77606.77608	journal	January 1990
Recovering logical structure from Charm++ event traces Isaacs, Katherine E.; Bhatele, Abhinav; Lifflander, Jonathan Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis https://doi.org/10.1145/2807591.2807634	conference	November 2015
Combing the Communication Hairball: Visualizing Parallel Execution Traces using Logical Time Isaacs, Katherine E.; Bremer, Peer-Timo; Jusufi, Ilir IEEE Transactions on Visualization and Computer Graphics, Vol. 20, Issue 12 https://doi.org/10.1109/TVCG.2014.2346456	journal	December 2014
ScalaTrace: Scalable compression and replay of communication traces for high-performance computing Noeth, Michael; Ratn, Prasun; Mueller, Frank Journal of Parallel and Distributed Computing, Vol. 69, Issue 8 https://doi.org/10.1016/j.jpdc.2008.09.001	journal	August 2009
PEMOGEN: automatic adaptive performance modeling during program runtime Bhattacharyya, Arnamoy; Hoefler, Torsten Proceedings of the 23rd international conference on Parallel architectures and compilation - PACT '14 https://doi.org/10.1145/2628071.2628100	conference	January 2014
Profile-based power shifting in interconnection networks with on/off links Miwa, Shinobu; Nakamura, Hiroshi Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis https://doi.org/10.1145/2807591.2807639	conference	November 2015
A regression-based approach to scalability prediction Barnes, Bradley J.; Rountree, Barry; Lowenthal, David K. Proceedings of the 22nd annual international conference on Supercomputing - ICS '08 https://doi.org/10.1145/1375527.1375580	conference	January 2008
Fast Multi-parameter Performance Modeling Calotoiu, Alexandru; Beckinsale, David; Earl, Christopher W. 2016 IEEE International Conference on Cluster Computing (CLUSTER) https://doi.org/10.1109/CLUSTER.2016.57	conference	September 2016
How The Flang Frontend Works Osmialowski, Paul Proceedings of the Fourth Workshop on the LLVM Compiler Infrastructure in HPC https://doi.org/10.1145/3148173.3148183	conference	November 2017
Mpipp Chen, Hu; Chen, Wenguang; Huang, Jian Proceedings of the 20th annual international conference on Supercomputing - ICS '06 https://doi.org/10.1145/1183401.1183451	conference	January 2006
Using automated performance modeling to find scalability bugs in complex codes Calotoiu, Alexandru; Hoefler, Torsten; Poke, Marius Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis on - SC '13 https://doi.org/10.1145/2503210.2503277	conference	January 2013
Scalable performance analysis of exascale MPI programs through signature-based clustering algorithms Bahmani, Amir; Mueller, Frank Proceedings of the 28th ACM international conference on Supercomputing - ICS '14 https://doi.org/10.1145/2597652.2597676	conference	January 2014
Ordering Traces Logically to Identify Lateness in Message Passing Programs Isaacs, Katherine E.; Gamblin, Todd; Bhatele, Abhinav IEEE Transactions on Parallel and Distributed Systems, Vol. 27, Issue 3 https://doi.org/10.1109/TPDS.2015.2417531	journal	March 2016
Parallel sparse flow-sensitive points-to analysis Zhao, Jisheng; Burke, Michael G.; Sarkar, Vivek Proceedings of the 27th International Conference on Compiler Construction https://doi.org/10.1145/3178372.3179517	conference	February 2018
MPIWiz Xue, Ruini; Liu, Xuezheng; Wu, Ming ACM SIGPLAN Notices, Vol. 44, Issue 4 https://doi.org/10.1145/1594835.1504213	journal	February 2009
Points-to analysis with efficient strong updates Lhoták, Ondrej; Chung, Kwok-Chiang Andrew Proceedings of the 38th annual ACM SIGPLAN-SIGACT symposium on Principles of programming languages - POPL '11 https://doi.org/10.1145/1926385.1926389	conference	January 2011
Persistent pointer information Xiao, Xiao; Zhang, Qirun; Zhou, Jinguo ACM SIGPLAN Notices, Vol. 49, Issue 6 https://doi.org/10.1145/2666356.2594314	journal	June 2014
Automatic pool allocation Lattner, Chris; Adve, Vikram Proceedings of the 2005 ACM SIGPLAN conference on Programming language design and implementation - PLDI '05 https://doi.org/10.1145/1065010.1065027	conference	January 2005
Fact Zhai, Jidong; Sheng, Tianwei; He, Jiangzhou Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis - SC '09 https://doi.org/10.1145/1654059.1654087	conference	January 2009
Quality of service profiling Misailovic, Sasa; Sidiroglou, Stelios; Hoffmann, Henry Proceedings of the 32nd ACM/IEEE International Conference on Software Engineering - ICSE '10 https://doi.org/10.1145/1806799.1806808	conference	January 2010
Level by level Yu, Hongtao; Xue, Jingling; Huo, Wei Proceedings of the 8th annual IEEE/ ACM international symposium on Code generation and optimization - CGO '10 https://doi.org/10.1145/1772954.1772985	conference	January 2010

Similar Records

Scalable I/O Tracing and Analysis

Conference · Thu Jan 01 00:00:00 EST 2009 · OSTI ID:1769152

Vijayakumar, Karthik; Mueller, Frank; Ma, Xiaosong; +1 more

Probabilistic Communication and I/O Tracing with Deterministic Replay at Scale

Conference · Sat Jan 01 00:00:00 EST 2011 · OSTI ID:1769152

Wu, Xing; Vijayakumar, Karthik; Mueller, Frank; +2 more

ScalaTrace: Tracing, Analysis and Modeling of HPC Codes at Scale

Conference · Wed Mar 31 00:00:00 EDT 2010 · OSTI ID:1769152

Mueller, F; Wu, X; Schulz, M; +2 more

Related Subjects

97 MATHEMATICS AND COMPUTING
Communication traces
MPI
LLVM

Title: PredCom: A Predictive Approach to Collecting Approximated Communication Traces

Citation Formats

References (33)

Similar Records

Related Subjects