Offloading Bloom Filter Operations to Network Processor for Parallel Query Processing in Cluster of Workstations
Workstation clusters have high performance interconnects with programmable network processors, which facilitate interesting opportunities to offload certain application specific computation on them and hence enhance the performance of the parallel application. Our earlier work in this direction achieves enhanced performance and balanced utilization of resources by exploiting the programmable features of the network interface in parallel database query execution. In this paper, we extend our earlier work for studying parallel query execution with Bloom filters. We propose and evaluate a scheme to offload the Bloom filter operations to the network processor. Further we explore offloading certain tuple processing activities on to the network processor by adopting a network interface attached disk scheme. The above schemes yield a speedup of up to 1.13 over the base scheme with Bloom filter where all processing is done by the host processor and achieve balanced utilization of resources. In the presence of a disk buffer cache, which reduces both the disk and I/O traffic, offloading schemes improve the speedup to 1.24.
KeywordsNetwork Interface Bloom Filter Query Execution Host Processor Network Processor
Unable to display preview. Download preview PDF.
- 3.Chen, M.-S., Lo, M.-L., Yu, P.S., Young, H.C.: Using Segmented Right-Deep Trees for the Execution of Pipelined Hash Joins. In: Proc. of 18th Very Large Data Bases, pp. 15–26 (August 1992)Google Scholar
- 5.Jenkins, B. (1996), http://burtleburtle.net/bob/c/lookup2.c
- 7.Mackert, L.F., Lohman, G.M.: R* Optimizer Validation and Performance Evaluation for Distributed Queries. In: Proc. of 12th Intl. Conf. on Very Large Data Bases, pp. 149–159 (August 1986)Google Scholar
- 9.PCI-SIG Home (2004), http://www.pcisig.com/
- 10.Santhosh Kumar, V., Thazhuthaveetil, M.J., Govindarajan, R.: Exploiting Programmable Network Interfaces for Parallel Query Execution in Workstation Clusters. TR-HPC-10/2005, LHPC, SERC, IISc (2005), http://hpc.serc.iisc.ernet.in/Publications/gvsk2005.ps
- 11.Schneider, D., DeWitt, D.: A performance evaluation of four parallel join algorithms in a shared-nothing multiprocessor environment. In: Proc. of ACM SIGMOD Conference (June 1989)Google Scholar
- 12.Schneider, D., DeWitt, D.: Tradeoffs in processing complex queries via hashing in multiprocessor database machines. In: Proc. of 16th Intl. Conf. on Very Large Data Bases (August 1990)Google Scholar
- 13.Seitz, C.L.: Myrinet Technology Roadmap. Myrinet Users Group Conf. (May 2002)Google Scholar
- 14.Karamcheti, V., Chien, A.: Software Overhead in Messaging Layers: Where Does the Time Go? In: Proc. of 6th Intl. Conf. on Architectural Support for Programming Languages and Operating Systems (October 1994)Google Scholar
- 15.Tamura, T., Oguchi, M., Kitsuregawa, M.: Parallel Database Processing on a 100 Node PC Cluster: Cases for Decision Support Query Processing and Data Mining. In: Proc. of Supercomputing (November 1997)Google Scholar
- 16.TPC BenchmarkTM H (Decision Support) Standard Specification Revision 1.3.0. Transaction Processing Performance Council(TPC) (1999)Google Scholar
- 17.Zuberek, W.M.: Modeling using Timed Petri Nets - event-driven simulation, Technical Report No. 9602, Dept. of Computer Science, Memorial Univ. of Newfoundland, St. John’s, Canada (1996), ftp://ftp.cs.mun.ca/pub/techreports/tr-9602.ps.Z