Cited By
View all- Bouhrour SPepin TJaeger J(2022)Towards leveraging collective performance with the support of MPI 4.0 features in MPCParallel Computing10.1016/j.parco.2021.102860109:COnline publication date: 1-Mar-2022
Implementation and performance evaluation of MPI persistent collectives in MPC: a case study
Pages 51 - 60
Abstract
Persistent collective communications have recently been voted in the MPI standard, opening the door to many optimizations to reduce collectives cost, in particular for recurring operations. Indeed persistent semantics contains an initialization phase called only once for a specific collective. It can be used to collect building costs necessary to the collective, to avoid paying them each time the operation is performed.
We propose an overview of the implementation of the persistent collectives in the MPC MPI runtime. We first present a naïve implementation for MPI runtimes already providing nonblocking collectives. Then, we improve this first implementation with two levels of caching optimizations. We present the performance results of the naïve and optimized versions and discuss their impact on different collective algorithms. We observe performance improvement compared to the naïve version on a repetitive benchmark, up to a 3x speedup for the reduce collective.
References
[1]
Alexandra Carpen-Amarie, Sascha Hunold, and Jesper Larsson Träff. 2017. On expected and observed communication performance with MPI derived datatypes. Parallel Comput. 69, 98 – 117. https://doi.org/10.1016/j.parco.2017.08.006
[2]
Alexandre Denis, Julien Jaeger, Emmanuel Jeannot, Marc Pérache, and Hugo Taboada. 2018. Dynamic Placement of Progress Thread for Overlapping MPI Non-blocking Collectives on Manycore Processor. In Euro-Par 2018: Parallel Processing - 24th International Conference on Parallel and Distributed Computing, Turin, Italy, August 27-31, 2018, Proceedings(Lecture Notes in Computer Science), Marco Aldinucci, Luca Padovani, and Massimo Torquati(Eds.), Vol. 11014. Springer, 616–627. https://doi.org/10.1007/978-3-319-96983-1_44
[3]
Alexandre Denis, Julien Jaeger, Emmanuel Jeannot, Marc Pérache, and Hugo Taboada. 2019. Study on progress threads placement and dedicated cores for overlapping MPI nonblocking collectives on manycore processor. Int. J. High Perform. Comput. Appl. 33, 6 (2019). https://doi.org/10.1177/1094342019860184
[4]
MPI Forum. 1994. MPI: A message-passing interface standard, version 1.1.
[5]
MPI Forum. 2015. MPI: A message-passing interface standard, version 3.1.
[6]
Brice Goglin, Emmanuel Jeannot, Farouk Mansouri, and Guillaume Mercier. 2018. Hardware topology management in MPI applications through hierarchical communicators. Parallel Comput. 76(2018), 70 – 90. https://doi.org/10.1016/j.parco.2018.05.006
[7]
Khalid Hasanov and Alexey Lastovetsky. 2017. Hierarchical redesign of classic MPI reduction algorithms. The Journal of Supercomputing 73, 2 (2017), 713–725.
[8]
Masayuki Hatanaka, Atsushi Hori, and Yutaka Ishikawa. 2013. Optimization of MPI Persistent Communication. In Proceedings of the 20th European MPI Users’ Group Meeting(EuroMPI ’13). Association for Computing Machinery, New York, NY, USA, 79–84. https://doi.org/10.1145/2488551.2488566
[9]
Masayuki Hatanaka, Masamichi Takagi, Atsushi Hori, and Yutaka Ishikawa. 2017. Offloaded MPI Persistent Collectives Using Persistent Generalized Request Interface. In Proceedings of the 24th European MPI Users’ Group Meeting(EuroMPI ’17). Association for Computing Machinery, New York, NY, USA, Article Article 5, 10 pages. https://doi.org/10.1145/3127024.3127029
[10]
Torsten Hoefler and Andrew Lumsdaine. 2006. Design, Implementation, and Usage of LibNBC. School of Informatics.
[11]
Torsten Hoefler and Timo Schneider. 2012. Optimization Principles for Collective Neighborhood Communications. International Conference for High Performance Computing, Networking, Storage and Analysis, SC, 98:1–98:10. https://doi.org/10.1109/SC.2012.86
[12]
Qiao Kang, Jesper Larsson Träff, Reda Al-Bahrani, Ankit Agrawal, Alok Choudhary, and Wei-keng Liao. 2018. Full-Duplex Inter-Group All-to-All Broadcast Algorithms with Optimal Bandwidth. In Proceedings of the 25th European MPI Users’ Group Meeting(EuroMPI ’18). Association for Computing Machinery, New York, NY, USA, Article Article 1, 10 pages. https://doi.org/10.1145/3236367.3236374
[13]
S. Kumar, P. Heidelberger, D. Chen, and M. Hines. 2010. Optimization of applications with non-blocking neighborhood collectives via multisends on the Blue Gene/P supercomputer. In 2010 IEEE International Symposium on Parallel Distributed Processing (IPDPS). 1–11. https://doi.org/10.1109/IPDPS.2010.5470407
[14]
Jesper Larsson Träff, Alexandra Carpen-Amarie, Sascha Hunold, and Antoine Rougier. 2016. Message-Combining Algorithms for Isomorphic, Sparse Collective Communication. arXiv e-prints, Article arXiv:1606.07676, arXiv:1606.07676 pages. arxiv:cs.DC/1606.07676
[15]
Bradley Morgan, Daniel J. Holmes, Anthony Skjellum, Purushotham Bangalore, and Srinivas Sridharan. 2017. Planning for Performance: Persistent Collective Operations for MPI. In Proceedings of the 24th European MPI Users’ Group Meeting(EuroMPI ’17). Association for Computing Machinery, New York, NY, USA, Article Article 4, 11 pages. https://doi.org/10.1145/3127024.3127028
[16]
Marc Pérache, Patrick Carribault, and Hervé Jourdren. 2009. MPC-MPI: An MPI Implementation Reducing the Overall Memory Consumption. In Recent Advances in Parallel Virtual Machine and Message Passing Interface, Proceedings of the 16th European PVM/MPI Users’ Group Meeting (EuroPVM/MPI 2009), Matti Ropo, Jan Westerholm, and Jack Dongarra (Eds.). Lecture Notes in Computer Science, Vol. 5759. Springer Berlin Heidelberg, 94–103. https://doi.org/10.1007/978-3-642-03770-2_16
[17]
Martin Ruefenacht, Mark Bull, and Stephen Booth. 2017. Generalisation of recursive doubling for AllReduce: Now with simulation. Parallel Comput. 69(2017), 24 – 44. https://doi.org/10.1016/j.parco.2017.08.004
[18]
Anthony Skjellum. 1998. High Performance MPI: Extending the Message Passing Interface for Higher Performance and Higher Predictability. In International Conference on Parallel and Distributed Processing Techniques and Applications(PDPTA’98). 25–32.
[19]
Jesper Larsson Träff and Sascha Hunold. 2019. Cartesian Collective Communication. In Proceedings of the 48th International Conference on Parallel Processing(ICPP 2019). Association for Computing Machinery, New York, NY, USA, Article 48, 11 pages. https://doi.org/10.1145/3337821.3337848
[20]
Jesper Larsson Träff, Antoine Rougier, and Sascha Hunold. 2014. Implementing a Classic: Zero-Copy All-to-All Communication with Mpi Datatypes. In Proceedings of the 28th ACM International Conference on Supercomputing(ICS’14). Association for Computing Machinery, New York, NY, USA, 135–144. https://doi.org/10.1145/2597652.2597662
Recommendations
Offloaded MPI persistent collectives using persistent generalized request interface
EuroMPI '17: Proceedings of the 24th European MPI Users' Group MeetingThis paper proposes a library with a persistent generalized request interface for the implementation of persistent communication operations. This interface allows developers to add persistent communication functions to the existing MPI library. We ...
Implementation and performance analysis of non-blocking collective operations for MPI
SC '07: Proceedings of the 2007 ACM/IEEE conference on SupercomputingCollective operations and non-blocking point-to-point operations have always been part of MPI. Although non-blocking collective operations are an obvious extension to MPI, there have been no comprehensive studies of this functionality. In this paper we ...
Verifying Performance Guidelines for MPI Collectives at Scale
SC-W '23: Proceedings of the SC '23 Workshops of The International Conference on High Performance Computing, Network, Storage, and AnalysisMPI collective communication operations are crucial for high-performance computing, making the efficient implementation of collective algorithms essential for optimal application performance. While most MPI libraries provide several algorithms for a ...
Comments
Information & Contributors
Information
Published In
September 2020
88 pages
ISBN:9781450388801
DOI:10.1145/3416315
Copyright © 2020 ACM.
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than the author(s) must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected].
Publisher
Association for Computing Machinery
New York, NY, United States
Publication History
Published: 07 October 2020
Check for updates
Qualifiers
- Research-article
- Research
- Refereed limited
Conference
EuroMPI/USA '20
Acceptance Rates
Overall Acceptance Rate 66 of 139 submissions, 47%
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- View Citations1Total Citations
- 103Total Downloads
- Downloads (Last 12 months)22
- Downloads (Last 6 weeks)0
Reflects downloads up to 10 Nov 2024
Other Metrics
Citations
Cited By
View all- Bouhrour SPepin TJaeger J(2022)Towards leveraging collective performance with the support of MPI 4.0 features in MPCParallel Computing10.1016/j.parco.2021.102860109:COnline publication date: 1-Mar-2022
View Options
Get Access
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign inFull Access
View options
View or Download as a PDF file.
PDFeReader
View online with eReader.
eReaderHTML Format
View this article in HTML Format.
HTML Format