The VLDB Journal

, Volume 26, Issue 1, pp 31–54 | Cite as

Resource bricolage and resource selection for parallel database systems

  • Jiexing Li
  • Jeffrey F. Naughton
  • Rimma V. Nehme
Special Issue Paper

Abstract

Running parallel database systems in an environment with heterogeneous resources has become increasingly common, due to cluster evolution and increasing interest in moving applications into public clouds. Performance differences among machines in the same cluster pose new challenges for parallel database systems. First, for database systems running in a heterogeneous cluster, the default uniform data partitioning strategy may overload some of the slow machines, while at the same time it may underutilize the more powerful machines. Since the processing time of a parallel query is determined by the slowest machine, such an allocation strategy may result in a significant query performance degradation. Second, since machines might have varying resources or performance, different choices of machines may lead to different costs or performance for executing the same workload. By carefully selecting the most suitable machines for running a workload, we may achieve better performance with the same budget, or we may meet the same performance requirements with a lower cost. We address these challenges by introducing techniques we call resource bricolage and resource selection that improve database performance in heterogeneous environments. Our approaches quantify the performance differences among machines with various resources as they process workloads with diverse resource requirements. For the purpose of better resource utilization, we formalize the problem of minimizing workload execution time and view it as an optimization problem, and then, we employ linear programming to obtain a recommended data partitioning scheme. For the purpose of better resource selection, we formalize two problems: One minimizes the total workload execution time with a given budget, and the other minimizes the total budget with a given performance target. We then employ different mixed-integer programs to search for the optimal resource selection decisions. We verify the effectiveness of both resource bricolage and resource selection techniques with an extensive experimental study.

Keywords

Resource bricolage Resource selection Parallel database systems Heterogeneous clusters Performance prediction Data partitioning 

References

  1. 1.
    Agrawal, S., Narasayya, V., Yang, B.: Integrating vertical and horizontal partitioning into automated physical database design. In: SIGMOD (2004)Google Scholar
  2. 2.
    Ahmad, F., Chakradhar, S.T., Raghunathan, A., Vijaykumar. T.N.: Tarazu: Optimizing mapreduce on heterogeneous clusters. In: ASPLOS (2012)Google Scholar
  3. 3.
    Ahmad, M., Aboulnaga, A., Babu, S., Munagala, K.: Modeling and exploiting query interactions in database systems. In: CIKM (2008)Google Scholar
  4. 4.
    Akdere, M., Çetintemel, U., Riondato, M., Upfal, E., Zdonik, S.B.: Learning-based query performance modeling and prediction. In: ICDE (2012)Google Scholar
  5. 5.
    Chaudhuri, S., Kaushik, R., Ramamurthy, R.: When can we trust progress estimators for SQL queries? In: SIGMOD (2005)Google Scholar
  6. 6.
    Chaudhuri, S., Narasayya, V., Ramamurthy, R.: Estimating progress of execution for SQL queries. In: SIGMOD (2004)Google Scholar
  7. 7.
    Copeland, G., Alexander, W., Boughter, E., Keller, T.: Data placement in Bubba. In: SIGMOD Record (1988)Google Scholar
  8. 8.
    Curino, C., Jones, E., Zhang, Y., Madden, S.: Schism: a workload-driven approach to database replication and partitioning. In: PVLDB (2010)Google Scholar
  9. 9.
    Dantzig, G.B., Thapa, M.N.: Linear Programming 1: Introduction. Springer, Berlin (1997)MATHGoogle Scholar
  10. 10.
    Delimitrou, C., and Kozyrakis, C.: Paragon: Qos-aware scheduling for heterogeneous datacenters. In: ASPLOS (2013)Google Scholar
  11. 11.
    DeWitt, D.J., Naughton, J.F., Schneider, D.A., Seshadri, S.: Practical skew handling in parallel joins. In: VLDB (1992)Google Scholar
  12. 12.
    Duggan, J., Cetintemel, U., Papaemmanouil, O., and Upfal, E.: Performance prediction for concurrent database workloads. In: SIGMOD (2011)Google Scholar
  13. 13.
    Farley, B., Juels, A., Varadarajan, V., Ristenpart, T., Bowers, K.D., Swift, M.M.: More for your money: exploiting performance heterogeneity in public clouds. In: SoCC (2012)Google Scholar
  14. 14.
    Ganapathi, A., Kuno, H., Dayal, U., Wiener, J.L., Fox, A., Jordan, M., Patterson, D.: Predicting multiple metrics for queries: better decisions enabled by machine learning. In: ICDE (2009)Google Scholar
  15. 15.
    Ghandeharizadeh, S., DeWitt, D.J., Qureshi, W.: A performance analysis of alternative multi-attribute declustering strategies. In: SIGMOD (1992)Google Scholar
  16. 16.
    Johnson, D.S.: The NP-completeness column: An ongoing guide. J. Algorithms. 6(3), 434–451 (1985)MathSciNetCrossRefMATHGoogle Scholar
  17. 17.
    König, A.C., Ding, B., Chaudhuri, S., Narasayya, V.: A statistical approach towards robust progress estimation. In: PVLDB (2012)Google Scholar
  18. 18.
    Li, J., Nehme, R.V., Naughton, J.F.: GSLPI: A cost-based query progress indicator. In ICDE (2012)Google Scholar
  19. 19.
    Li, J., Nehme, R.V., Naughton, J.F., Toward progress indicators on steroids for big data systems. In CIDR (2013)Google Scholar
  20. 20.
    Luo, G., Naughton, J.F., Ellmann, C.J., Watzke, M.W.: Toward a progress indicator for database queries. In: SIGMOD (2004)Google Scholar
  21. 21.
    Luo, G., Naughton, J.F., Ellmann, C.J., Watzke, M.W.: Increasing the accuracy and coverage of SQL progress indicators. In: ICDE (2005)Google Scholar
  22. 22.
    Mangot, D.: EC2 Variability: The Numbers Revealed. http://tech.mangot.com/roller/dave/entry/ec2_variability_the_numbers_revealed (2009)
  23. 23.
    Mehta, M., DeWitt, D.J.: Data placement in shared-nothing parallel database systems. VLDB J. 6(1), 53–72 (1997)CrossRefGoogle Scholar
  24. 24.
    Mishra, C., Koudas, N.: A lightweight online framework for query progress indicators. In: ICDE (2007)Google Scholar
  25. 25.
    Morton, K., Balazinska, M., Grossman, D.: Paratimer: a progress indicator for MapReduce DAGs. In: SIGMOD (2010)Google Scholar
  26. 26.
    Morton, K., Friesen, A., Balazinska, M., Grossman, D.: Estimating the progress of MapReduce pipelines. In: ICDE (2010)Google Scholar
  27. 27.
    Nehme, R., Bruno, N.: Automated partitioning design in parallel database systems. In: SIGMOD (2011)Google Scholar
  28. 28.
    Ou, Z., Zhuang, H., Lukyanenko, A., Nurminen, J.K., Hui, P., Mazalov, V., Yla-Jaaski, A.: Is the same instance type created equal? Exploiting heterogeneity of public clouds. IEEE Trans. Cloud Comput. 1(2), 201–214 (2013)CrossRefGoogle Scholar
  29. 29.
    Ou, Z., Zhuang, H., Nurminen, J.K., Ylä-Jääski, A., Hui, P.: Exploiting hardware heterogeneity within the same instance type of Amazon EC2. In: HotCloud (2012)Google Scholar
  30. 30.
    Pavlo, A., Curino, C., Zdonik, S.: Skew-aware automatic database partitioning in shared-nothing, parallel OLTP systems. In: SIGMOD (2012)Google Scholar
  31. 31.
    Rao, J., Zhang, C., Megiddo, N., Lohman, G.: Automating physical database design in a parallel database. In: SIGMOD (2002)Google Scholar
  32. 32.
    Reiss, C., Tumanov, A., Ganger, G. R., Katz, R. H., Kozuch, M. A.: Heterogeneity and dynamicity of clouds at scale: Google trace analysis. In: SoCC (2012)Google Scholar
  33. 33.
    Schad, J., Dittrich, J., Quiané-Ruiz, J.-A., Runtime measurements in the cloud: observing, analyzing, and reducing variance. In: PVLDB (2010)Google Scholar
  34. 34.
  35. 35.
    Wang, G., Ng, T.S.E.: The impact of virtualization on network performance of amazon EC2 data center. In: INFOCOM (2010)Google Scholar
  36. 36.
    Wu, W., Chi, Y., Hacígümüş, H., Naughton, J.F.: Towards predicting query execution time for concurrent and dynamic database workloads. In: PVLDB (2013)Google Scholar
  37. 37.
    Wu, W., Chi, Y., Zhu, S., Tatemura, J., Hacigümüs, H., Naughton, J.F.: Predicting query execution time: are optimizer cost models really unusable? In: ICDE (2013)Google Scholar
  38. 38.
    Xu, Y., Kostamaa, P.: Efficient outer join data skew handling in parallel DBMS. In: PVLDB (2009)Google Scholar
  39. 39.
    Xu, Y., Kostamaa, P., Zhou, X., Chen, L.: Handling data skew in parallel joins in shared-nothing systems. In: SIGMOD (2008)Google Scholar
  40. 40.
    Zaharia, M., Konwinski, A., Joseph, A.D., Katz, R., Stoica, I. Improving mapreduce performance in heterogeneous environments. In: OSDI (2008)Google Scholar
  41. 41.
    Zhuang, H., Liu, X., Ou, Z., Aberer, K.: Impact of instance seeking strategies on resource allocation in cloud data centers. In: CLOUD (2013)Google Scholar
  42. 42.
    Zou, T., Bras, R.L., Salles, M.V., Demers, A., Gehrke, J.: ClouDiA: a deployment advisor for public clouds. In: PVLDB (2013)Google Scholar
  43. 43.
    Zou, T., Wang, G., Salles, M.V., Bindel, D., Demers, A., Gehrke, J., White, W.: Making time-stepped applications tick in the cloud. In: SOCC (2011)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2016

Authors and Affiliations

  • Jiexing Li
    • 1
  • Jeffrey F. Naughton
    • 2
  • Rimma V. Nehme
    • 3
  1. 1.Google IncMountain ViewUSA
  2. 2.Department of Computer SciencesUniversity of Wisconsin, MadisonMadisonUSA
  3. 3.Microsoft Jim Gray Systems LabMadisonUSA

Personalised recommendations