Abstract
Autotuning (AT) is a promising concept to minimize the often tedious manual effort of optimizing scientific applications for a specific target platform. Ideally, an AT approach can reliably identify the most efficient implementation variant(s) for a new platform or new characteristics of the input by applying suitable program transformations and analytic models. In this work, we introduce Offsite, an offline AT approach that automates this selection process at installation time by rating implementation variants based on an analytic performance model without requiring time-consuming runtime tests. From abstract multilevel description languages, Offsite automatically derives optimized, platform-specific and problem-specific code of possible variants and applies the performance model to these variants.
We apply Offsite to parallel numerical methods for ordinary differential equations (ODEs). In particular, we investigate tuning a specific class of explicit ODE solvers, PIRK methods, for four different initial value problems (IVPs) on three different shared-memory systems. Our experiments demonstrate that Offsite can reliably identify the set of most efficient implementation variants for different given test configurations (ODE solver, IVP, platform) and effectively handle important AT scenarios.
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
Change history
15 June 2020
The original version of chapters 17 and 24 were previously published non-open access. They have now been made open access under a CC BY 4.0 license and the copyright holder has been changed to ‘The Author(s).’ The book has also been updated with the change.
The chapters 19 and 25 were inadvertently published open access. This has been corrected and the chapters are now non-open access.
Notes
- 1.
YAML is a data serialization language; https://yaml.org.
- 2.
There are scalable ODE systems but also ODEs with a fixed size [10].
- 3.
For example files, we refer to https://github.com/RRZE-HPC/kerncraft.
- 4.
In this work, version 0.8.3 of the Kerncraft tool was used.
- 5.
The ECM prediction factors in the location of data in the memory hierarchy. As a simplified assumption—neglecting overlapping effects at cache borders—, this means that as long as data locations do not change, the ECM model yields the same value for a kernel independent from the actual n.
- 6.
Step-up time is the same for OffPre5 and OffPre10 as determining their set of considered variants is carried out by the same single database operation.
References
Ansel, J., et al.: OpenTuner: an extensible framework for program autotuning. In: Proceedings of the 23rd International Conference on Parallel Architecture and Compilation Techniques, PACT 2014, pp. 303–316. ACM, August 2014. https://doi.org/10.1145/2628071.2628092
Balaprakash, P., et al.: Autotuning in high-performance computing applications. Proc. IEEE 106(11), 2068–2083 (2018). https://doi.org/10.1109/JPROC.2018.2841200
Barthel, A., Günther, M., Pulch, R., Rentrop, P.: Numerical techniques for different time scales in electric circuit simulation. In: Breuer, M., Durst, F., Zenger, C. (eds.) High Performance Scientific and Engineering Computing, pp. 343–360. Springer, Heidelberg (2002). https://doi.org/10.1007/978-3-642-55919-8_38
Bilmes, J., Asanovic, K., Chin, C.W., Demmel, J.: Optimizing matrix multiply using PHiPAC: a portable, high-performance, ANSI C coding methodology. In: Proceedings 11th International Conference on Supercomputing, ICS 1997, pp. 340–347. ACM, July 1997. https://doi.org/10.1145/263580.263662
Calvo, M., Franco, J.M., Randez, L.: A new minimum storage runge-kutta scheme for computational acoustics. J. Comput. Phys. 201(1), 1–12 (2004). https://doi.org/10.1016/j.jcp.2004.05.012
Christen, M., Schenk, O., Burkhard, L.: PATUS: a code generation and autotuning framework for parallel iterative stencil computations on modern microarchitectures. In: 2011 IEEE International Parallel Distributed Processing Symposium, pp. 676–687, May 2015. https://doi.org/10.1109/IPDPS.2011.70
Das, S., Mullick, S.S., Suganthan, P.N.: Recent advances in differential evolution - an updated survey. Swarm Evol. Comput. 27, 1–30 (2016). https://doi.org/10.1016/j.swevo.2016.01.004
Denoyelle, N., Goglin, B., Ilic, A., Jeannot, E., Soussa, L.: Modeling non-uniform memory access on large compute nodes with the cache-aware roofline model. IEEE T. Parall. Distr. 30(6), 1374–1389 (2019). https://doi.org/10.1109/TPDS.2018.2883056
Gerndt, M., César, E., Benkner, S. (eds.): Automatic Tuning of HPC Applications - The Periscope Tuning Framework. Shaker Verlag (2015)
Hairer, E., Wanner, G.: Solving Ordinary Differential Equations II: Stiff and Differential-Algebraic Problems. Springer, Heidelberg (2002). 2nd rev. edn
Hammer, J., Eitzinger, J., Hager, G., Wellein, G.: Kerncraft: a tool for analytic performance modeling of loop kernels. In: Niethammer, C., Gracia, J., Hilbrich, T., Knüpfer, A., Resch, M.M., Nagel, W.E. (eds.) Tools for High Performance Computing 2016, pp. 1–22. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-56702-0_1
Hofmann, J., Alappat, C., Hager, G., Fey, D., Wellein, G.: Bridging the Architecture Gap: Abstracting Performance-Relevant Properties of Modern Server Processors (2019). https://arxiv.org/abs/1907.00048, Preprint
van der Houwen, P.J., Sommeijer, B.P.: Parallel iteration of high-order runge-kutta methods with stepsize control. J. Comput. Appl. Math. 29(1), 111–127 (1990). https://doi.org/10.1016/0377-0427(90)90200-J
Mazzia, F., Magherini, C.: Test Set for Initial Value Problem Solvers, Release 2.4, February 2008. https://archimede.dm.uniba.it/~testset/
Mendis, C., Renda, A., Amarasinghe, S., Carbin, M.: Ithemal: accurate, portable and fast basic block throughput estimation using deep neural networks. In: Proceedings of the 36th International Conference on Machine Learning. Proceedings of the Machine Learning Research, vol. 97, pp. 4505–4515. PMLR, June 2019
Pfaffe, P., Grosser, T., Tillmann, M.: Efficient hierarchical online-autotuning: a case study on polyhedral accelerator mapping. In: Proceedings of the ACM International Conference on Supercomputing, ICS 2019, pp. 354–366. ACM, New York (2019). https://doi.org/10.1145/3330345.3330377
Rasch, A., Gorlatch, S.: ATF: a generic directive-based auto-tuning framework. Concurr. Comput. Pract. Exper. 31(5), e4423 (2019). https://doi.org/10.1002/cpe.4423
Scherg, M., Seiferth, J., Korch, M., Rauber, T.: Performance prediction of explicit ODE methods on multi-core cluster systems. In: Proceedings of the 2019 ACM/SPEC International Conference on Performance Engineering, ICPE 2019, pp. 139–150. ACM (2019). https://doi.org/10.1145/3297663.3310306
Seiferth, J., Alappat, C., Korch, M., Rauber, T.: Applicability of the ECM performance model to explicit ode methods on current multi-core processors. In: Yokota, R., Weiland, M., Keyes, D., Trinitis, C. (eds.) ISC High Performance 2018. LNCS, vol. 10876, pp. 163–183. Springer, Cham (2018). https://doi.org/10.1007/978-3-319-92040-5_9
Shudler, S., Vrabec, J., Wolf, F.: Understanding the scalability of molecular simulation using empirical performance modeling. In: Bhatele, A., Boehme, D., Levine, J.A., Malony, A.D., Schulz, M. (eds.) ESPT/VPA 2017-2018. LNCS, vol. 11027, pp. 125–143. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-17872-7_8
Stengel, H., Treibig, J., Hager, G., Wellein, G.: Quantifying performance bottlenecks of stencil computations using the execution-cache-memory model. In: Proceedings of the 29th ACM International Conference on Supercomputing, pp. 207–216. ICS 2015. ACM (2015). https://doi.org/10.1145/2751205.2751240
Tiwari, A., Hollingsworth, J.K.: Online adaptive code generation and tuning. In: Proceedings of the 2011 IEEE International Parallel Distributed Processing Symposium, IPDPS 2011, pp. 879–892. IEEE, May 2011. https://doi.org/10.1109/IPDPS.2011.86
Whaley, R.C., Petitet, A., Dongarra, J.: Automated empirical optimizations of software and the ATLAS project. Parallel Comput. 27(1), 3–35 (2001). https://doi.org/10.1016/S0167-8191(00)00087-9
Williams, S., Waterman, A., Patterson, D.: Roofline: an insightful visual performance model for multicore architectures. Commun. ACM 52(4), 65–76 (2009). https://doi.org/10.1145/1498765.1498785
Yount, C., Tobin, J., Breuer, A., Duran, A.: YASK - yet another stencil kernel: a framework for HPC stencil code-generation and tuning. In: 2016 6th International Workshop on Domain-Specific Languages and High-Level Frameworks for High Performance Computing, WOLFHPC, pp. 30–39. IEEE, November 2016. https://doi.org/10.1109/WOLFHPC.2016.08
Acknowledgments
This work is supported by the German Ministry of Science and Education (BMBF) under project number 01IH16012A. Furthermore, we thank the Erlangen Regional Computing Center for granting access to their IvyBridge and Skylake systems.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Seiferth, J., Korch, M., Rauber, T. (2020). Offsite Autotuning Approach. In: Sadayappan, P., Chamberlain, B.L., Juckeland, G., Ltaief, H. (eds) High Performance Computing. ISC High Performance 2020. Lecture Notes in Computer Science(), vol 12151. Springer, Cham. https://doi.org/10.1007/978-3-030-50743-5_19
Download citation
DOI: https://doi.org/10.1007/978-3-030-50743-5_19
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-50742-8
Online ISBN: 978-3-030-50743-5
eBook Packages: Computer ScienceComputer Science (R0)