Abstract
Legacy MPI applications are an important and economically valuable category of parallel software that rely on the MPI-1, MPI-2 (and, more recently, MPI-3) standards to achieve performance and portability. Many of these applications have been developed or ported to MPI over the past two decades, with the implicit (dual) goal of achieving acceptably high performance and scalability, and a high level of portability between diverse parallel architectures. However they were often created implicitly using MPI in ways that exploited how a particular underlying MPI behaved at the time (such as those with polling progress and poor implementation of some operations). Thus, they did not necessarily take advantage of the full potential for describing latent concurrency or for loosening the coupling of the application thread from the message scheduling and transfer.
This paper presents a first transformation tool, Petal, that identifies calls to legacy MPI primitives. Petal is implemented on top of the ROSE source-to-source infrastructure and automates the analysis and transformation of existing codes to utilize non-blocking MPI and persistent MPI primitives. We use control flow and pointer alias analysis to overlap communication and computation. The transformed code is capable of supporting better application bypass, yielding better overlapping of communication, computation, and I/O. We present the design of the tool and its evaluation on available benchmarks.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Provided the underlying MPI does not poll excessively to make progress or for message completion, the messages are long enough, and there is sufficient memory bandwidth for both communication and computation.
- 2.
At least on high quality implementations of MPI.
- 3.
This indicates that mvapich may not optimize the code path for persistent send and/or receive.
References
The NAS parallel benchmarks. https://www.nas.nasa.gov/publications/npb.html
Persistent MPI communication. http://www.mpi-forum.org/docs/mpi-1.1/mpi-11-html/node51.html. Accessed on 28 June 2015
The ROSE source-to-source compiler. http://rosecompiler.org
Aananthakrishnan, S., Bronevetsky, G., Gopalakrishnan, G.: Hybrid approach for data-flow analysis of mpi programs. In: Proceedings of the 27th International ACM Conference on International Conference on Supercomputing, ICS 2013, pp. 455–456. ACM, New York (2013)
Danalis, A., Pollock, L., Swany, M.: Automatic MPI application transformation with asphalt. In: Parallel and Distributed Processing Symposium, IPDpPS 2007, pp. 1–8, IEEE International, March 2007
Das, D., Gupta, M., Ravindran, R., Shivani, W., Sivakeshava, P., Uppal, R.: Compiler-controlled extraction of computation-communication overlap in MPI applications. In: IEEE International Symposium on Parallel and Distributed Processing, IPDpPS 2008, pp. 1–8, April 2008
Frexus: mpi-2d-plate (2013). http://project.github.com. Accessed on 1 September 2015
Gropp, W., Lusk, E., Thakur, R.: Using MPI-2: Advanced Features of the Message-Passing Interface. MIT Press, Cambridge (1999)
Haque, M., Yi, Q., Dinan, J., Balaji, P.: Enhancing performance portability of MPI applications through annotation-based transformations. In: 2013 42nd International Conference on Parallel Processing (ICPP), pp. 631–640, October 2013
Hoefler, T.: New and old features in MPI-3.0: The past, the standard, and the future, April 2012
Karlin, I., Keasler, J., Neely, R.: Lulesh 2.0 updates and changes. Technical report LLNL-TR-641973, August 2013
Forum, Message Passing Interface: MPI: A message-passing interface standard. Technical report, Knoxville (1994)
Message Passing Interface Forum: MPI: A message-passing interface standard version 3.1, June 2015
Nguyen, T., Cicotti, P., Bylaska, E., Quinlan, D., Baden, S.B.: Bamboo: translating MPI applications to a latency-tolerant, data-driven form. In: Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis, SC 2012, pp. 39:1–39:11. IEEE Computer Society Press, Los Alamitos (2012)
Pellegrini, S., Hoefler, T., Fahringer, T.: Exact dependence analysis for increased communication overlap. In: Träff, J.L., Benkner, S., Dongarra, J.J. (eds.) EuroMPI 2012. LNCS, vol. 7490, pp. 89–99. Springer, Heidelberg (2012)
Pirkelbauer, P., Dechev, D., Stroustrup, B.: Source code rejuvenation is not refactoring. In: van Leeuwen, J., Muscholl, A., Peleg, D., Pokorný, J., Rumpe, B. (eds.) SOFSEM 2010. LNCS, vol. 5901, pp. 639–650. Springer, Heidelberg (2010)
Schordan, M., Quinlan, D.: A source-to-source architecture for user-defined optimizations. In: Böszörményi, L., Schojer, P. (eds.) JMLC 2003. LNCS, vol. 2789, pp. 214–223. Springer, Heidelberg (2003)
Shao, S., Jones, A., Melhem, R.: A compiler-based communication analysis approach for multiprocessor systems. In: 20th International Parallel and Distributed Processing Symposium, IPDpPS 2006, p. 10, April 2006
Steensgaard, B.: Points-to analysis in almost linear time. In: Proceedings of the 23rd ACM SIGPLAN-SIGACT Symposium on Principles of Programming Languages, POPL 1996, pp. 32–41. ACM, New York (1996)
Towns, J., Cockerill, T., Dahan, M., Foster, I., Gaither, K., Grimshaw, A., Hazlewood, V., Lathrop, S., Lifka, D., Peterson, G., Roskies, R., Scott, J., Wilkins-Diehr, N.: XSEDE: accelerating scientific discovery. Comput. Sci. Eng. 16(5), 62–74 (2014)
Acknowledgements
This work used the Extreme Science and Engineering Discovery Environment (XSEDE), which is supported by National Science Foundation grant number ACI-1053575.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2016 Springer International Publishing Switzerland
About this paper
Cite this paper
Ahmed, H., Skjellum, A., Pirkelbauer, P. (2016). Petal Tool for Analyzing and Transforming Legacy MPI Applications. In: Shen, X., Mueller, F., Tuck, J. (eds) Languages and Compilers for Parallel Computing. LCPC 2015. Lecture Notes in Computer Science(), vol 9519. Springer, Cham. https://doi.org/10.1007/978-3-319-29778-1_10
Download citation
DOI: https://doi.org/10.1007/978-3-319-29778-1_10
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-29777-4
Online ISBN: 978-3-319-29778-1
eBook Packages: Computer ScienceComputer Science (R0)