1 / 9

Enhancing Query Evaluation Through Compressed Projection Index Uncompression on CUDA

This project explores the use of compressed projection indexes to improve query evaluation efficiency on GPUs. Traditional approaches often face bottlenecks during data transfer from the CPU to the GPU, leading to increased latency. By implementing Run-Length Encoding (RLE) for data compression alongside a parallel prefix sum algorithm for decompression, we aim to minimize data transfer sizes while leveraging thread-level parallelism on the GPU. This could significantly enhance performance for query evaluation on large datasets.

red
Télécharger la présentation

Enhancing Query Evaluation Through Compressed Projection Index Uncompression on CUDA

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Uncompressing a Projection Index in CUDA Eduardo Gutarra Velez

  2. Introduction & Motivation • The projection index supports thread-level parallelism and therefore could potentially make good use of a GPU. • However, most of the time spent when doing query evaluation on projection indexes, is spent in transferring data from the CPU to the GPU • A common approach to improve on this problem is to reduce the size of the data that needs to be transferred. • Compression could be a good way to reduce the size of data.

  3. The Project • A compressed projection index will be used. • The compression method is RLE (Run Length Encoding) • For this to be effective the following assumptions must be made: • The data in the projection index is previously sorted • The projection index is created on a column that is not unique.

  4. The Project • The Index will be transferred compressed to the GPU • It will then be uncompressed in the GPU using a parallel prefix sum algorithm. CPU GPU • A3B1C7 • A3B1C7 • AAABCCCCCCC

  5. RLE (Run Length Encoding)

  6. Prefix Sum (Scan)

  7. How Prefix Sum will be applied • Many uses for prefix sum but our use will be to uncompress a previously compressed index, that will be sent to memory. • A3B1C7 • AAABCCCCCCC • Source:

  8. References • Gosink, L., Kesheng Wu, E. Wes Bethel, John D. Owens, Kenneth I. Joy: Data Parallel Bin-Based Indexing for Answering Queries on Multi-core Architectures. SSDBM 2009: 110-129 • Gosink, L., E. Wes Bethel, John D. Owens, Kenneth I. Joy. Bin-Hash Indexing: A Parallel GPU-Based Method For Fast Query Processing. IDAV (2008) • Wu, K., Otoo, E., Shoshani, A.: On the performance of bitmap indices for high cardinality attributes. In: Proc. of VLDB, pp. 24–35 (2004) • O’Neil, P.E., Quass, D.: Improved query performance with variant indexes. In: Proc. of SIGMOD, pp. 38–49 (1997)

More Related