Advertisement

Cost-efficient parallel processing of irregularly structured problems in cloud computing environments

  • Jens Haussmann
  • Wolfgang Blochinger
  • Wolfgang Kuechlin
Article
  • 110 Downloads

Abstract

In this paper, we deal with optimizing the monetary costs of executing parallel applications in cloud-based environments. Specifically, we investigate on how scalability characteristics of parallel applications impact the total costs of computations. We focus on a specific class of irregularly structured problems, where the scalability typically depends on the input data. Consequently, dynamic optimization methods are required for minimizing the costs of computation. For quantifying the total monetary costs of individual parallel computations, the paper presents a cost model that considers the costs for the parallel infrastructure employed as well as the costs caused by delayed results. We discuss a method for dynamically finding the number of processors for which the total costs based on our cost model are minimal. Our extensive experimental evaluation gives detailed insights into the performance characteristics of our approach.

Keywords

High performance distributed computing Cloud computing Parallel computing Cost model Irregularly structured problems 

References

  1. 1.
  2. 2.
    Anbar, A., Narayana, V.K., El-Ghazawi, T.: Distributed shared memory programming in the cloud. In: Proceedings of the 12th International Symposium on Cluster, Cloud and Grid Computing, CCGRID, pp. 707–708. IEEE/ACM, Washington, DC, USA (2012).  https://doi.org/10.1109/CCGrid.2012.48
  3. 3.
  4. 4.
    Berenbrink, P., Friedetzky, T., Goldberg, L.A.: The natural work-stealing algorithm is stable. In: Proceedings of the International Conference on Cluster Computing, pp. 178–187. IEEE (2001).  https://doi.org/10.1109/SFCS.2001.959892
  5. 5.
    Blochinger, W.: Towards robustness in parallel SAT solving. In: PARCO, pp. 301–308 (2005)Google Scholar
  6. 6.
    Blochinger, W., Sinz, C., Küchlin, W.: Parallel propositional satisfiability checking with distributed dynamic learning. Parallel Comput. 29(7), 969–994 (2003).  https://doi.org/10.1016/S0167-8191(03)00068-1 CrossRefGoogle Scholar
  7. 7.
    Blochinger, W., Dangelmayr, C., Schulz, S.: Aspect-oriented parallel discrete optimization on the cohesion desktop grid platform. In: Proceedings of the 6th Symposium on Cluster Computing and the Grid, CCGRID, pp. 49–56. IEEE (2006).  https://doi.org/10.1109/CCGRID.2006.20
  8. 8.
    Bonet, B., Geffner, H.: Planning as heuristic search. Artif. Intell. 129(1–2), 5–33 (2001).  https://doi.org/10.1016/S0004-3702(01)00108-4 MathSciNetCrossRefzbMATHGoogle Scholar
  9. 9.
    Carreno, E.D., Diener, M., Cruz, E.H.M., Navaux, P.O.A.: Automatic communication optimization of parallel applications in public clouds. In: Proceedings of the 16th International Symposium on Cluster, Cloud and Grid Computing, CCGRID, pp. 1–10. IEEE/ACM (2016).  https://doi.org/10.1109/ccgrid.2016.59
  10. 10.
    Da Rosa Righi, R., Rodrigues, V.F., Da Costa, C.A., Galante, G., De Bona, L.C.E., Ferreto, T.: AutoElastic: automatic resource elasticity for high performance applications in the cloud. IEEE Trans. Cloud Comput. 4(1), 6–19 (2016).  https://doi.org/10.1109/TCC.2015.2424876 CrossRefGoogle Scholar
  11. 11.
    Da Rosa Righi, R., Rodrigues, V.F., Rostirolla, G., Da Costa, C.A., Roloff, E., Navaux, P.O.A.: A lightweight plug-and-play elasticity service for self-organizing resource provisioning on parallel applications. Fut. Gen. Comput. Syst. 78(1), 176–190 (2018).  https://doi.org/10.1016/j.future.2017.02.023 CrossRefGoogle Scholar
  12. 12.
    Deelman, E., Singh, G., Livny, M., Berriman, B., Good, J.: The cost of doing science on the cloud: the montage example. In: Proceedings of the Conference on Supercomputing, SC ’09, p. 50. ACM/IEEE (2008).  https://doi.org/10.1109/SC.2008.5217932
  13. 13.
    Dorigo, M., Caro, G.D., Gambardella, L.M.: Ant algorithms for discrete optimization. Artif. Life 5(2), 137–172 (1999).  https://doi.org/10.1162/106454699568728 CrossRefGoogle Scholar
  14. 14.
    Edelkamp, S., Schroedl, S.: Heuristic Search: Theory and Applications. Elsevier, Amsterdam (2012).  https://doi.org/10.1016/B978-0-12-372512-7.00019-5 CrossRefGoogle Scholar
  15. 15.
    Ekanayake, J., Jackson, J., Lu, W., Barga, R., Balkir, A.S.: A scalable communication runtime for clouds. In: Proceedings of the 4th International Conference on Cloud Computing, CLOUD, pp. 211–218. IEEE (2011).  https://doi.org/10.1109/CLOUD.2011.21
  16. 16.
    Evangelinos, C., Hill, C.N.: Cloud computing for parallel scientific HPC applications: feasibility of running coupled atmosphere-ocean climate models on Amazons EC2. In: 1st Workshop on Cloud Computing and its Applications, CCA, vol. 2, pp. 2–34 (2008)Google Scholar
  17. 17.
    Fokaefs, M., Barna, C., Litoiu, M.: Economics-driven resource scalability on the cloud. In: Proceedings of the 11th International Symposium on Software Engineering for Adaptive and Self-Managing Systems, SEAMS, pp. 129–139. ACM Press, New York, USA (2016).Google Scholar
  18. 18.
    Genaud, S., Gossa, J.: Cost-wait trade-offs in client-side resource provisioning with elastic clouds. In: Proceedings of the 4th International Conference on Cloud Computing, CLOUD, pp. 1–8. IEEE, Washington, DC, USA (2011).  https://doi.org/10.1109/CLOUD.2011.23
  19. 19.
    Gholkar, N., Mueller, F., Rountree, B.: A power-aware cost model for HPC procurement. In: Proceedings of the 30th International Symposium on Parallel and Distributed Processing, IPDPSW, pp. 1110–1113. IEEE (2016).  https://doi.org/10.1109/IPDPSW.2016.35
  20. 20.
    Google Cloud Platform—Compute Engine (2014). https://cloud.google.com/compute/
  21. 21.
    Grama, A., Kumar, V.: State of the art in parallel search techniques for discrete optimization problems. IEEE Trans. Knowl. Data Eng. 11(1), 28–35 (1999).  https://doi.org/10.1109/69.755612 CrossRefGoogle Scholar
  22. 22.
    Grama, A., Kumar, V., Karypis, G., Gupta, A.: Introduction to Parallel Computing, 2nd edn. Addison-Wesley Longman Publishing Co., Inc, Boston (2003)zbMATHGoogle Scholar
  23. 23.
    Gupta, A., Milojicic, D.: Evaluation of HPC applications on cloud. In: Proceedings of the 6th Open Cirrus Summit, OCS, pp. 22–26. IEEE (2011).  https://doi.org/10.1109/OCS.2011.10
  24. 24.
    Hansen, E.A., Zilberstein, S.: LAO*: a heuristic search algorithm that finds solutions with loops. Artif. Intell. 129(1–2), 35–62 (2001).  https://doi.org/10.1016/S0004-3702(01)00106-0 MathSciNetCrossRefzbMATHGoogle Scholar
  25. 25.
    Henderson, D.R.: The Concise Encyclopedia of Economics. Liberty Fund, Indianapolis (2007)Google Scholar
  26. 26.
    Jackson, K., Ramakrishnan, L., Muriki, K., Canon, S., Cholia, S., Shalf, J., Wasserman, H.J., Wright, N.J.: Performance analysis of high performance computing applications on the Amazon Web Services Cloud. In: Proceedings of the 2nd International Conference on Cloud Computing Technology and Science, CLOUDCOM, pp. 159–168. IEEE (2010).  https://doi.org/10.1109/CloudCom.2010.69
  27. 27.
    Janjic, V., Hammond, K.: Using load information in work-stealing on distributed systems with non-uniform communication latencies. In: Proceedings of the 18th European Conference on Parallel Processing, Euro-Par, pp. 155–166. Springer (2012).  https://doi.org/10.1007/978-3-642-32820-6_17
  28. 28.
    Kang, D.K., Kim, S.H., Youn, C.H., Chen, M.: Cost adaptive workflow scheduling in cloud computing. In: Proceedings of the 8th International Conference on Ubiquitous Information Management and Communication, ICUIMC, pp. 1–8. ACM, New York, USA (2014).  https://doi.org/10.1145/2557977.2558079
  29. 29.
    Kiefer, J.: Sequential minimax search for a maximum. Proc. Am. Math. Soc. 4(3), 502–506 (1953).  https://doi.org/10.2307/2032161 MathSciNetCrossRefzbMATHGoogle Scholar
  30. 30.
    Mao, M., Humphrey, M.: Auto-scaling to minimize cost and meet application deadlines in cloud workflows. In: Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis, SC, pp. 1–12. IEEE (2011).  https://doi.org/10.1145/2063384.2063449
  31. 31.
    Mao, M., Humphrey, M.: Scaling and scheduling to maximize application performance within budget constraints in cloud workflows. In: Proceedings of the 27th International Symposium on Parallel and Distributed Processing, IPDPS, pp. 67–78. IEEE (2013).  https://doi.org/10.1109/IPDPS.2013.61
  32. 32.
    Mao, M., Li, J., Humphrey, M.: Cloud auto-scaling with deadline and budget constraints. In: Proceedings of the 11th International Conference on Grid Computing, GRID, pp. 41–48. IEEE/ACM (2010).  https://doi.org/10.1109/GRID.2010.5697966
  33. 33.
    Marques-Silva, J.P., Sakallah, K.A.: Dynamic search-space pruning techniques in path sensitization. In: Proceedings of the 31st Design Automation Conference, DAC, pp. 705–711. ACM, New York, USA (1994).  https://doi.org/10.1109/DAC.1994.204192
  34. 34.
    Marques-Silva, J.P., Sakallah, K.A.: Boolean satisifiability in electronic design automation. In: Proceedings of the 37th Design Automation Conference, DAC, pp. 675–680. ACM Press, New York, USA (2000).  https://doi.org/10.1145/337292.337611
  35. 35.
    Mauch, V., Kunze, M., Hillenbrand, M.: High performance cloud computing. Fut. Gen. Comput. Syst. 29(6), 1408–1416 (2013).  https://doi.org/10.1016/j.future.2012.03.011 CrossRefGoogle Scholar
  36. 36.
    Messac, A., Puemi-Sukam, C., Melachrinoudis, E.: Aggregate objective functions and pareto frontiers: required relationships and practical implications. Optim. Eng. 1(2), 171–188 (2000).  https://doi.org/10.1023/A:1010035730904 MathSciNetCrossRefzbMATHGoogle Scholar
  37. 37.
    Min, S., Iancu, C., Yelick, K.: Hierarchical work stealing on manycore clusters. In: Proceedings of the 5th Conference on Partitioned Global Address Space Programming Models, PGAS, pp. 1–10 (2011).Google Scholar
  38. 38.
    Nguyen, T.V.A., Bimonte, S., D’Orazio, L., Darmont, J.: Cost models for view materialization in the cloud. In: Proceedings of the 15th Joint EDBT/ICDT Workshop, pp. 47–54. ACM, New York, USA (2012).  https://doi.org/10.1145/2320765.2320788
  39. 39.
    OpenStack OpenSource Community: OpenStack Project (2013). https://www.openstack.org/
  40. 40.
  41. 41.
    Quintin, J.N., Wagner, F.: Hierarchical work-stealing. In: Proceedings of the 16th European Conference on Parallel Processing, Euro-Par, pp. 217–229. Springer (2010).  https://doi.org/10.1007/978-3-642-15277-1_21
  42. 42.
    Rajan, D., Thain, D.: Designing self-tuning split-map-merge applications for high cost-efficiency in the cloud. IEEE Trans. Cloud Comput. 5(2), 303–316 (2017).  https://doi.org/10.1109/TCC.2015.2415780 CrossRefGoogle Scholar
  43. 43.
    Righi, RdR, Rodrigues, V.F., da Costa, C.A., Kreutz, D., Heiss, H.U.: Towards cloud-based asynchronous elasticity for iterative HPC applications. J. Phys. Conf. Ser. 649, 012006 (2015).  https://doi.org/10.1088/1742-6596/649/1/012006 CrossRefGoogle Scholar
  44. 44.
    Schulz, S., Blochinger, W.: Parallel SAT solving on peer-to-peer desktop grids. J. Grid Comput. 8(3), 443–471 (2010).  https://doi.org/10.1007/s10723-010-9160-1 CrossRefGoogle Scholar
  45. 45.
    Sharma, U., Shenoy, P., Sahu, S., Shaikh, A.: A cost-aware elasticity provisioning system for the cloud. In: Proceedings of the 31st International Conference on Distributed Computing Systems, pp. 559–570. IEEE (2011).  https://doi.org/10.1109/ICDCS.2011.59
  46. 46.
    Sun, Y., Wang, C.L.: Solving irregularly structured problems based on distributed object model. Parallel Comput. 29(11–12), 1539–1562 (2003).  https://doi.org/10.1016/j.parco.2003.05.006 CrossRefGoogle Scholar
  47. 47.
    TSPLIB95—Library of Sample Instances for the TSP: https://www.iwr.uni-heidelberg.de/groups/comopt/software/TSPLIB95/
  48. 48.
    van Nieuwpoort, R.V., Kielmann, T., Bal, H.E.: Efficient load balancing for wide-area divide-and-conquer applications. In: Proceedings of the 8th ACM SIGPLAN Symposium on Principles and Practices of Parallel Programming, PPoPP, vol. 36, pp. 34–43. ACM (2001).  https://doi.org/10.1145/568014.379563
  49. 49.
    Viana, V., De Oliveira, D., Mattoso, M.: Towards a cost model for scheduling scientific workflows activities in cloud environments. In: Proceedings of the World Congress on Services, SERVICES, pp. 216–219. IEEE (2011).  https://doi.org/10.1109/SERVICES.2011.52
  50. 50.
    Vu, T.T., Derbel, B.: Link-heterogeneous work stealing. In: Proceedings of the 14th International Symposium on Cluster, Cloud, and Grid Computing, CCGrid, pp. 354–363. IEEE/ACM (2014).  https://doi.org/10.1109/CCGrid.2014.85
  51. 51.
    Wu, C., Buyya, R.: Cloud Data Centers and Cost Modeling—A Complete Guide to Planning, Designing and Building a Cloud Data Center, 1st edn. Elsevier, San Francisco (2015)Google Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2018

Authors and Affiliations

  1. 1.Parallel and Distributed Computing GroupReutlingen UniversityReutlingenGermany
  2. 2.Symbolic Computation GroupUniversity of TuebingenTuebingenGermany

Personalised recommendations