Analysing Metric Data Structures Thinking of an Efficient GPU Implementation
Similarity search is becoming a field of interest because it can be applied to different areas in science and engineering. In real applications, when large volumes of data are processing, query response time can be quite high. In this case, it is necessary to apply mechanisms to significantly reduce the average query response time. For that purpose, modern GPU/Multi-GPU systems offer a very impressive cost/performance ratio. In this paper, the authors make a comparative study of the most popular pivot selection methods in order to stablish a set of attractive features from the point of view of future GPU implementations.
KeywordsClustering-based methods Comparative study Data structures Metric spaces Pivot-based methods Range queries Similarity search.
This work has been supported by the Ministerio de Ciencia e Innovación, project SATSIM (Ref: CGL2010-20787-C02-02), Spain and Research Center, University of Magallanes, Chile. Also, this work has been partially supported by CAPAP-H3 Network (TIN2010-12011-E).
- 1.Chávez E, Navarro G, Baeza-Yates R, Marroquín JL (2001) Searching in metric spaces. ACM Comput Surv 33(3):273–321Google Scholar
- 4.Ciaccia P, Patella M, Zezula P (1997) M-tree : an efficient access method for similarity search in metric spaces. In: Proceedings of 23rd international conference on VLDB, 426–435Google Scholar
- 5.Brin S (1995) Near neighbor search in large metric spaces. In: Proceedings of the 21st VLDB conference, Morgan Kaufmann Publishers, 574–584, 1995Google Scholar
- 10.Pedreira O, Brisaboa NR (2007) Spatial selection of sparse pivots for similarity search in metric spaces. In: Proceedings of the 33rd conference on current trends in theory and practice of computer science (SOFSEM (2007) LNCS, vol 4362. Czech Republic, Springer, Harrachov, pp 434–445Google Scholar
- 11.Uribe-Paredes R, Cazorla D, Sánchez JL, Arias E (2012) A comparative study of different metric structures: thinking on gpu implementations. In: Proceedings of the world congress on engineering (2012) WCE 2012. Lecture notes in engineering and computer science, England, London, pp 312–317Google Scholar
- 12.Chávez E, Marroquín J, Baeza-Yates R (1999) Spaghettis: an array based algorithm for similarity queries in metri spaces. In: Proceedings of 6th international symposium on String Processing and Information Retrieval (SPIRE’99), IEEE CS Press, pp 38–46Google Scholar
- 14.Hetland M (2009) The basic principles of metric indexing. In: Coello CA, Dehuri S, Ghosh, S (eds) Swarm intelligence for multi-objective problems in data mining, Studies in computational intelligence, vol 242. Springer Berlin, pp 199–232Google Scholar