Optimizing Sorting and Top-k Selection Steps in Permutation Based Indexing on GPUs
Permutation-based indexing is one of the most popular techniques for the approximate nearest-neighbor search problem in high-dimensional spaces. Due to the exponential increase of multimedia data, the time required to index this data has become a serious constraint of current techniques. One of the possible steps towards faster index construction is the utilization of massively parallel platforms such as the GPGPU architectures. In this paper, we have focused on two particular steps of permutation index construction – the selection of top-k nearest pivot points and sorting these pivots according to their respective distances. Even though these steps are integrated into a more complex algorithm, we address them selectively since they may be employed individually for different indexing techniques or query processing algorithms in multimedia databases. We also provide a discussion of alternative approaches that we have tested but which have proved less efficient on present hardware.
KeywordsIndexing Permutation GPU Top-k Sorting Bitonic sort
Unable to display preview. Download preview PDF.
- 1.NVIDIA: Maxwell GPU Architecture. http://developer.nvidia.com/maxwell-compute-architecture
- 2.Batcher, K.E.: Sorting networks and their applications. In: Proceedings of the Spring Joint Computer Conference, April 30-May 2, 1968, pp. 307–314. ACM (1968)Google Scholar
- 4.Jagadish, H.V., Mendelzon, A.O., Milo, T.: Similarity-based queries. In: Proceedings of the Fourteenth ACM SIGACT-SIGMOD-SIGART Symposium on Principles of Database Systems, pp. 36–45 (1995)Google Scholar
- 5.Knuth, D.E.: Sorting and Searching. Addison-Wesley (2003)Google Scholar
- 6.Kruliš, M., Lokoč, J., Skopal, T.: Efficient extraction of feature signatures using Multi-GPU architecture. In: Li, S., El Saddik, A., Wang, M., Mei, T., Sebe, N., Yan, S., Hong, R., Gurrin, C. (eds.) MMM 2013, Part II. LNCS, vol. 7733, pp. 446–456. Springer, Heidelberg (2013) CrossRefGoogle Scholar
- 7.Krulis, M., Osipyan, H., Marchand-Maillet, S.: Permutation based indexing for high dimensional data on GPU architectures. In: 13th International Workshop on Content-Based Multimedia Indexing (CBMI). IEEE (2015)Google Scholar
- 8.Mohamed, H., Osipyan, H., Marchand-Maillet, S.: Multi-core (CPU and GPU) for permutation-based indexing. In: Traina, A.J.M., Traina Jr., C., Cordeiro, R.L.F. (eds.) SISAP 2014. LNCS, vol. 8821, pp. 277–288. Springer, Heidelberg (2014) Google Scholar
- 11.Samet, H.: Foundations of Multidimensional and Metric Data Structures (The Morgan Kaufmann Series in Computer Graphics and Geometric Modeling). Morgan Kaufmann Publishers Inc., San Francisco (2005) Google Scholar