LoT: Dynamic declustering of TSB-tree nodes for parallel access to temporal data
In this paper, we consider the problem of exploiting I/O parallelism for efficient access to transaction-time temporal databases. As temporal databases maintain historical versions of records in addition to current ones, we consider range queries in both time dimension and key dimension. Multiple disks can be used to read sets of disk blocks in parallel, thereby improving the performance of such queries substantially.
The problem is to find an optimal declustering algorithm for spreading record versions across disks. The solution depends on the index structure used. We have adopted the time split B-tree, as it provides efficient support for time range and key range queries. Our declustering method coined LoT (Local Balancing for TSB-trees) aims to decluster runs of logically consecutive leaf nodes of a TSB-tree onto separate disks. The method is dynamic in the sense that it computes the disk address of a new node at its creation time, based on the disk addresses of the nodes in its neighborhood.
LoT is an extension of the local balancing algorithm presented in [SL91]. It considers different sets of disks for historical and current nodes, and uses a two-dimensional distance metric between TSB-tree leaf nodes. As historical nodes of TSB-trees are no longer subject to splits, the coordinates of new nodes in the time-key space are restricted. This is exploited in LoT for achieving good declustering for both time range and key range queries. We derive performance guarantees for LoT in terms of the speedup for range queries. Simulation results show the response time speedup of LoT compared to a scheme that assigns nodes to disks in a random manner.
KeywordsLeaf Node Range Query Current Node Index Term Index Node
Unable to display preview. Download preview PDF.
- [AE93]Abdel-Ghaffar, K., El Abbadi, A.: Optimal disk allocation for partial match queries, ACM TODS, Vol. 18 No. 1, 1993Google Scholar
- [Ch94]Chen, P.M., Lee, E.K., Gibson, G.A., Katz, R.H., Patterson, D.A., RAID: High-Performance, Reliable Secondary Storage, ACM Computing Surveys Vol.26 No.2,1994Google Scholar
- [DG92]De Witt, D.J., Gray, J.: Parallel Database Systems: The Future of High Performance Database Systems, Communications of the ACM Vol. 35 No.6, 1992, pp. 85–98Google Scholar
- [DS82]Du, H.C., Sobolewski, J.S., Disk Allocation for Cartesian Product Files on Multiple Disk Systems, ACM Transactions on Database Systems Vol. 7 No. 1, 1982, pp. 82–101Google Scholar
- [EWK93]Elmasri, R., Wuu, G.T.J., Kouramajian, V.: The Time Index and the Monotonic B+-tree, in: [Tan93]Google Scholar
- [FB93]Faloutsos, C., Bhagwat, P., Declustering Using Fractals, 2nd International Conference on Parallel and Distributed Information Systems, San Diego, 1993Google Scholar
- [FM91]Faloutsos, C., Metaxas, D., Disk Allocation Methods Using Error Correcting Codes, IEEE Transactions on Computers Vol. 40 No.8, 1991, pp. 907–914Google Scholar
- [GD94]Ghandeharizadeh, S., DeWitt, D.J.: MAGIC.: A Multiattribute Declustering Mechanism for Multiprocessor Database Machines, IEEE Transactions on Parallel and Distributed Systems Vol. 5 No.5, 1994, pp. 509–524Google Scholar
- [HLY93]Hua, K., Lee, C., Young, H.C.: Data Partitioning for Multicomputer Database Systems: A Cell-Based Approach, Information Systems Vol. 18 No.5, 1993, pp. 329–342Google Scholar
- [HS94]Himatsingka, B., Srivastava, J.: Performance Evaluation of Grid Based Multi-Attribute Record Declustering Methods, 10th IEEE International Conference on Data Engineering, Houston, 1994Google Scholar
- [KEC94]Kouramajian, V, Elmasri, R., Chaudry, A.: Declustering Techniques for Parallelizing Temporal Access Structures, Proc. Data Engineering, 1994Google Scholar
- [KF92]Kamel, I., Faloutsos, C.: Parallel R-trees, Proc. SIGMOD, 1992Google Scholar
- [KP88]Kim, M.H., Pramanik, S., Optimal File Distribution for Partial Match Queries, ACM SIGMOD International Conference on Management of Data, Chicago, 1988Google Scholar
- [Kol93]Kolovson, C.P.: Indexing Techniques for Historical Databases, in [Tan93]Google Scholar
- [LS89]Lomet, D., Salzberg, B.: Access Methods for Multiversion Data, Proc. SIGMOD, 1989Google Scholar
- [LS90]Lomet, D., Salzberg, B.: The Performance of a Multiversion Access Method, Proc. SIGMOD, 1990Google Scholar
- [LS93]Lomet, D., Salzberg, B.: Transaction-Time Databases, in: [Tan93]Google Scholar
- [LSR92]Li, J., Srivastava, J., Rotem, D.: CMD: A multidimensional declustering method for parallel database systems. Proc. VLDB, 1992Google Scholar
- [MS94]Matsliach, G., Shmueli, O.: A Combined Method for Maintaining Large Indices in Multiprocessor Multidisk Environments, IEEE Transactions on Knowledge and Data Engineering Vol. 6 No.3, 1994, pp. 479–496Google Scholar
- [Mo94]Mohan, C., Pirahesh, H., Tang, W.G., Wang, Y.: Parallelism in Relational Database Management Systems, IBM Systems Journal Vol. 33 No.2, 1994, pp. 349–371Google Scholar
- [PK90]Pramanik, S., Kim, M.H.: Parallel Precessing of Large Node B-trees, IEEE Transactions on Computers, Vol. 39, 9, 1990Google Scholar
- [SL91]Seeger, B., Larson, P.A.: Multi-Disk B-trees, Proc. SIGMOD, 1991Google Scholar
- [SNH94]Srivastava, J., Niccum, T.M., Himatsingka, B., Data Declustering in PADMA: A PArallel Database MAnager, Bulletin of the IEEE TC on Data Engineering Vol. 17 No.3, September 1994Google Scholar
- [ST94]B. Salzberg, V.J. Tsotras, A Comparison of Access Methods for Time Evolving Data, Technical Report NU-CCS-94-21, Northeastern University, Boston, 1994Google Scholar
- [SWZ94]Scheuermann, P., Weikum, G., Zabback, P., “Disk Cooling” in Parallel Disk Systems, Bulletin of the IEEE TC on Data Engineering Vol.17 No.3 September 1994Google Scholar
- [Tan93]Tansel et al.: Temporal Databases, Benjamin Cummings Publishing, 1993Google Scholar
- [ZSC94]Zhou, Y, Shekhar, S., Coyle, M.: Disk Allocation Methods for Parallelizing Grid Files, 10th IEEE International Conference on Data Engineering, Houston, 1994Google Scholar