Abstract
This paper allows L. Lamport hyperplane method modified for improvement of the temporal data locality. Gauss-Seidel algorithm optimized by modified hyperplane method is faster than non-optimized in 2.5 times. This algorithm was paralleled by the technique of data placement with overlaps and we have got the speedup in 28 times on 16 processors in comparison with the non-optimized sequential algorithm.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Lamport, L.: The parallel execution of DO loops. Commun. ACM 17(2), 83–93 (1974)
Fernandez, A., Llaberia, J.M., Valero-Garcia, M.: Loop transformation using nonunimodal matrices. IEEE Trans. Parallel Distrib. Syst. 6(8), 832–840 (1995)
Kowarschik, M., Weiß, C.: An overview of cache optimization techniques and cache-aware numerical algorithms. In: Meyer, U., Sanders, P., Sibeyn, J. (eds.) Algorithms for Memory Hierarchies. LNCS, vol. 2625, pp. 213–232. Springer, Heidelberg (2003). doi:10.1007/3-540-36574-5_10
Wolfe, M.: More iteration space tiling. In: Proceedings of the 1989 ACM/IEEE conference on Supercomputing (Supercomputing 1989), pp. 655–664. ACM, New York (1989). https://doi.org/10.1145/76263.76337
Bastoul, C., Feautrier, P.: Improving data locality by chunking. In: Hedin, G. (ed.) CC 2003. LNCS, vol. 2622, pp. 320–334. Springer, Heidelberg (2003). doi:10.1007/3-540-36579-6_23
Likhoded, N.A.: Generalized tiling. Doklady NAN Belarusi, T. 55, N. 1, pp. 16–21 (2011). (in Russian)
Yurushkin, M.V.: Double block placement of data in RAM for solving the problem of matrix multiplication. Programmnaya inzheneriya, pp. 132–139 (2016). (in Russian)
Gervich, L.R., Steinberg, B.Y., Yurushkin, M.V.: Development of parallel programs with optimizing the use of memory structures. 120 p. Southern Federal University, Rostov-on-Don (2014). (in Russian)
Lam, S.M.: A data locality optimizing algorithm. In: Proceedings of the ACM SIGPLAN 1991 Conference on Programming Language Design and Implementation, pp. 30–44. ACM, New York (1991). ISBN:0-89791-428-7
Goto, K.: Anatomy of high-performance matrix multiplication. ACM Trans. Math. Softw. 34(3), 1–25 (2008)
Perepelkina, A.Y., Levchenko, V.D.: DiamondTorre algorithm for high-performance wave modeling. Keldysh Institute preprints, vol. 018, 20 p. (2015)
Graham, S.L., Snir, M., Patterson, C.A.: Getting Up To Speed: The Future Of Supercomputing, p. 289. National Academies Press, Washington (2005)
Abu-Khalil, J., Guda, S., Steinberg, B.: Porting Parallel Programs Without Loss of Efficiency. Open Syst. DBMS J. 23(4) (2015)
Steinberg, B.J., Abu-Khalil, J.M., Adigeyev, M.G., Bout, A.A., Kermanov, A.V., Pshenichnyy, E.A., Ramanchauskayte, G.V., Kroshkina, A.P., Gutnikov, A.V., Ponomareva, N.S., Panich, A.E., Shkurat, T.P.: A package of fast tools for genomic sequence analysis. Int. J. Math. Models Methods Appl. Sci. 10, 42–50 (2016). ISSN:1998-0140
Gervich, L.R., Kravchenko, E.N., Steinberg, B.Y., Yurushkin, M.V.: Automatic program parallelization with block data distribution. Sib. Zh. Vychisl. Mat. 18(1), 41–53 (2015)
Gervich, L.R., Steinberg, B.Y., Yurushkin, M.V.: ExaScale Systems Programming. Open Syst. J. 21(8) (2013)
Simonov, A.S.: High-speed Angara network: opportunities and prospects. PaVT 2016. http://omega.sp.susu.ru/books/conference/PaVT2016/talks/Simonov.pdf. (in Russian)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2017 Springer International Publishing AG
About this paper
Cite this paper
Ammaev, S.G., Gervich, L.R., Steinberg, B.Y. (2017). Combining Parallelization with Overlaps and Optimization of Cache Memory Usage. In: Malyshkin, V. (eds) Parallel Computing Technologies. PaCT 2017. Lecture Notes in Computer Science(), vol 10421. Springer, Cham. https://doi.org/10.1007/978-3-319-62932-2_24
Download citation
DOI: https://doi.org/10.1007/978-3-319-62932-2_24
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-62931-5
Online ISBN: 978-3-319-62932-2
eBook Packages: Computer ScienceComputer Science (R0)