SLAM : An Efficient Buffer Management Strategy using Spatial Locality of Spatial Data

SLAM : 공간 데이타의 공간적 근접성을 이용한 효율적인 버퍼관리기법

  • An, Jae-Yong (Dept. of Computer Science, Korea Advanced Institute of Science and Technology) ;
  • Min, Jun-Gi (Dept. of Computer Science, Korea Advanced Institute of Science and Technology) ;
  • Jeong, Jin-Wan (Dept. of Computer Science, Korea Advanced Institute of Science and Technology)
  • 안재용 (한국과학기술원 전산학과) ;
  • 민준기 (한국과학기술원 전산학과) ;
  • 정진완 (한국과학기술원 전산학과)
  • Published : 2002.10.01

Abstract

One of the major issues of DBMS is the buffer management. Because fetching data from the database disk is costly, the number of disk I/O's must be minimized in order to improve the DBMS performance. Although there have been many buffer management strategies to minimize the disk I/O, those strategies usually focused on just the temporal locality. Since there are the spatial locality as well as the temporal locality in the spatial database, strategies using only the temporal locality cannot achieve the optimal performance in the spatial database. In this paper, we propose a new buffer management strategy, the Spatial Locality Area Measure(SLAM) strategy, that considers not only the temporal locality but also the spatial locality. The SLAM buffer management strategy consists of two core structures, the SLM-tree and the M-LRU. We show the efficiency of the proposed strategy through experiments over various buffer sizes and reference frequencies.

데이타베이스 관리시스템에서 중요한 문제중의 하나는 효율적인 버퍼관리이다. 데이터베이스 관리시스템에서 객체를 디스크에서 읽어오는 작업은 많은 비용을 필요로 하기 때문에 시스템의 성능을 향상시키기 위해서는 디스크 I/O의 횟수를 최소화하는 것이 매우 중요하다. 지금까지 디스크 I/O 횟수를 줄이기 위한 많은 버퍼관리기법들이 제안되었지만, 그 기법들은 시간적 근접성만을 고려하기 때문에 공간적 근접성도 존재하는 공간데이타베이스 환경에서는 좋은 성능을 보여주지 못했다. 본 논문에서는 공간데이타베이스의 시간적 근접성과 공간적 근접성을 동시에 고려하는 새로운 버퍼관리기법인 Spatial Locality Area Measure(SLAM) 기법을 제안한다. 제안한 버퍼관리기법은 SLM-tree와 M-LRU, 두 개의 구조체로 구성되었으며 공간데이타베이스 환경에서의 다양한 버퍼크기와 참조빈도에 대한 실험에서 뛰어난 성능을 보여준다.

Keywords

References

  1. L. Ki-Joune and L. Robert 'The Spatial Locality and a Spatial Indexing Method by Dynamic Clustering in Hypermap System,' Advances in Spatial Databases, pp.207-223, 1990 https://doi.org/10.1007/3-540-54414-3_39
  2. Brinkhoff, T., Kriegel, H.-P., Seeger, B., 'Parallel processing of spatial joins using R-trees,' Data Engineering, Proceedings of the Twelfth International Conference on, pp.258-265, 1996 https://doi.org/10.1109/ICDE.1996.492114
  3. T. Brinkhoff, H. Hom, H. Kriegel, R. Schneider 'A Storage and Access Architecture for Efficient Query Processing in Spatial Database Systems,' Symposium on Large Spatial Databases, pp.357-376, 1993
  4. E. Bertino, L. Martino 'Object Oriented Database Management Systems: Concepts and Issues' IEEE Computer, Vol. 24, No.4, pp.31-47, April 1991 https://doi.org/10.1109/2.76261
  5. R. Breitl et al., 'The Gemstone Data Management System,' in Object-Oriented Concepts, Databases and Applications, W.Kim and F. Lochovsky eds., Addison-Wesley Readings, Mass., pp.283-308, 1989
  6. K. Wilkinson, P. Lyngbaek and W. Hassan, 'The Iris Architecture and Implementation,' IEEE Trans. Knowledge and Data Eng., Vol.2, No.1, pp.91-108, 1990 https://doi.org/10.1109/69.50906
  7. W. Effelsberg, T. Harder 'Principles of Database Buffer Management,' ACM Transactions on Database Systems, Vol. 9, No.4, pp.560-595, December 1984 https://doi.org/10.1145/1994.2022
  8. H. Chou and D.J.DeWitt 'An Evaluation of Buffer Management Strategies for Relational Database Systems,' Proceedings of the 11th VLDB Conference, pp.127-141, August 1985
  9. E.J. O'Neil, P. E. O'Neil, G. Weikum 'The LRU-K Page Replacement Algorithm For Database Disk Buffering,' ACM SIGMOD Conference, pp. 297-306, May, 1993 https://doi.org/10.1145/170035.170081
  10. T. Johnson and D. Shasha '2Q: A Low Overhead High Performance Buffer Management Replacement Algorithm,' Proceedings of the 20th VLDB Conference, pp.439-450, 1994
  11. R. Guting 'An Introduction to Spatial Database Systems,' the VLDB Journal, Vol. 3, No.4, pp, 357-399, October 1994 https://doi.org/10.1007/BF01231602
  12. D. J. DeWitt and D. Mater 'A Study of Three Alternative Workstation-Server Architectures for Object-Oriented Database Systems,' Proceedings of the 16th VLDB Conference, pp.107-121, 1990
  13. Tiger/lines precensus files: 1994 technical documentation, Technical report, U.S. Breau of Census, 1994
  14. T. Brinkhoff and H. Kriegel 'The Impact of Global Clustering on Spatial Database System,' Proceedings of the 20th VLDB Conference, pp.168-179, 1994