An architecture for recycling intermediates in a column-store

ABSTRACT

Automatically recycling (intermediate) results is a grand challenge for state-of-the-art databases to improve both query response time and throughput. Tuples are loaded and streamed through a tuple-at-a-time processing pipeline avoiding materialization of intermediates as much as possible. This limits the opportunities for reuse of overlapping computations to DBA-defined materialized views and function/result cache tuning.

In contrast, the operator-at-a-time execution paradigm produces fully materialized results in each step of the query plan. To avoid resource contention, these intermediates are evicted as soon as possible.

In this paper we study an architecture that harvests the by-products of the operator-at-a-time paradigm in a column store system using a lightweight mechanism, the recycler. The key challenge then becomes selection of the policies to admit intermediates to the resource pool, their retention period, and the eviction strategy when facing resource limitations.

The proposed recycling architecture has been implemented in an open-source system. An experimental analysis against the TPC-H ad-hoc decision support benchmark and a complex, real-world application (SkyServer) demonstrates its effectiveness in terms of self-organizing behavior and its significant performance gains. The results indicate the potentials of recycling intermediates and charters a route for further development of database kernels.

References

1. S. Agrawal, S.Chaudhuri, and V. R. Narasayya. Automated Selection of Materialized Views and Indexes in SQL Databases. In VLDB, 2000.
2. P. A. Boncz, M. L. Kersten, and S. Manegold. Breaking the Memory Wall in MonetDB. Commun. ACM, 51(12), 2008.
3. C. Bornhövd, M. Altinel, C. Mohan, H. Pirahesh, and B. Reinwald. Adaptive Database Caching with DBCache. IEEE Data Eng. Bull., 27(2):11--18, 2004.
4. N. Bruno and S. Chaudhuri. Physical Design Refinement: The 'Merge-Reduce' Approach. ACM Trans. Database Syst., 32(4), 2007.
5. C.-M. Chen and N. Roussopoulos. The Implementation and Performance Evaluation of the ADMS Query Optimizer: Integrating Query Result Caching and Matching. In EDBT, pages 323--336, 1994.
6. C.-H. Choi, J. X. Yu, and H. Lu. Dynamic Materialized View Management Based on Predicates. In APWeb, pages 583--594, 2003.
7. R. Cornacchia, S. Heman, M. Zukowski, A. P. de Vries, and P. A. Boncz. Flexible and Efficient IR Using Array Databases. VLDB J., 17(1):151---168, 2008.
8. J. Goldstein and P.-A. Larson. Optimizing Queries Using Materialized Views: A practical, scalable solution. In SIGMOD Conference, pages 331--342, 2001.
9. G. Graefe. Volcano -- An Extensible and Parallel Query Evaluation System. IEEE Trans. Knowl. Data Eng., 6(1):120--135, 1994.
10. M. Ivanova, M. L. Kersten, and N. Nes. Self-organizing Strategies for a Column-store Database. In Proc. EDBT, pages 157--168, 2008.
11. M. Ivanova, N. Nes, R. Goncalves, and M. L. Kersten. MonetDB/SQL Meets SkyServer: the Challenges of a Scientific Database. In Proc. SSDBM, Banff, Canada, July 2007.
12. Y. Kotidis and N. Roussopoulos. A Case for Dynamic View Management. ACM Trans. Database Syst., 26(4):388--423, 2001.
13. P.-Å. Larson, J. Goldstein, and J. Zhou. MTCache: Transparent Mid-Tier Database Caching in SQL Server. In ICDE, pages 177--189, 2004.
14. G. Luo. Partial Materialized Views. In ICDE, pages 756--765, 2007.
15. G. Luo and P. S. Yu. Content-based Filtering for Efficient Online Materialized View Maintenance. In CIKM, pages 163--172, 2008.
16. H. Mistry, P. Roy, S. Sudarshan, and K. Ramamritham. Materialized View Selection and Maintenance Using Multi-Query Optimization. In SIGMOD Conference, pages 307--318, 2001.
17. MonetDB, http://monetdb.cwi.nl/, 2008.
18. T. Phan and W.-S. Li. Dynamic Materialization of Query Views for Data Warehouse Workloads. In ICDE, pages 436--445, 2008.
19. P. Roy, S. Seshadri, S. Sudarshan, and S. Bhobe. Efficient and Extensible Algorithms for Multi Query Optimization. In SIGMOD Conference, pages 249--260, 2000.
20. Sloan Digital Sky Survey / SkyServer, 2008.
21. A. S. Szalay, J. Gray, et al. The SDSS SkyServer: Public Access to the Sloan Digital Sky Server Data. In SIGMOD, pages 570--581, 2002.
22. K.-L. Tan, S.-T. Goh, and B. C. Ooi. Cache-on-Demand: Recycling with Certainty. In ICDE, pages 633--640, 2001.
23. Transaction Processing Performance Council. TPC Benchmark H, Revision 2.6.2, 2008.
24. J. Zhou, P.-Å. Larson, J. C. Freytag, and W. Lehner. Efficient Exploitation of Similar Subexpressions for Query Processing. In SIGMOD Conference, pages 533--544, 2007.
25. J. Zhou, P.-Å. Larson, J. Goldstein, and L. Ding. Dynamic Materialized Views. In ICDE, pages 526--535, 2007.
26. M. Zukowski, S. Héman, N. Nes, and P. Boncz. Super-Scalar RAM-CPU Cache Compression. In Proc. ICDE, Atlanta, GA, USA, 2006.

Index Terms

• Information systems

  • Data management systems

    • Database management system engines

      • Database query processing

• Theory of computation

  • Theory and algorithms for application domains

    • Database theory

      • Database query processing and optimization (theory)