Idioma:

Column Sketches: A Scan Accelerator for Rapid and Robust Predicate Evaluation

Hentschel, Brian ; Kester, Michael S. ; Idreos, Stratos

Proceedings of the 2018 International Conference on Management of Data, 2018, p.857-872

New York, NY, USA: ACM

Texto completo disponível

Citações Citado por

Enviar para

Título:
Column Sketches: A Scan Accelerator for Rapid and Robust Predicate Evaluation
Autor: Hentschel, Brian ; Kester, Michael S. ; Idreos, Stratos
Assuntos: Information systems -- Information storage systems -- Record storage systems -- Record layout alternatives -- Fixed length attributes ; Information systems -- Information storage systems -- Record storage systems -- Record layout alternatives -- Variable length attributes ; Information systems -- Information storage systems -- Record storage systems -- Record storage alternatives -- Indexed file organization ; Information systems -- Information storage systems -- Record storage systems -- Relational storage -- Column based storage ; Information systems -- Information storage systems -- Record storage systems -- Relational storage -- Compression strategies ; Information systems -- Information systems applications -- Decision support systems -- Data analytics ; Information systems -- Information systems applications -- Decision support systems -- Data warehouses ; Information systems -- Information systems applications -- Decision support systems -- Online analytical processing
É parte de: Proceedings of the 2018 International Conference on Management of Data, 2018, p.857-872
Descrição: While numerous indexing and storage schemes have been developed to address the core functionality of predicate evaluation in data systems, they all require specific workload properties (query selectivity, data distribution, data clustering) to provide good performance and fail in other cases. We present a new class of indexing scheme, termed a Column Sketch, which improves the performance of predicate evaluation independently of workload properties. Column Sketches work primarily through the use of lossy compression schemes which are designed so that the index ingests data quickly, evaluates any query performantly, and has small memory footprint. A Column Sketch works by applying this lossy compression on a value-by-value basis, mapping base data to a representation of smaller fixed width codes. Queries are evaluated affirmatively or negatively for the vast majority of values using the compressed data, and only if needed check the base data for the remaining values. Column Sketches work over column, row, and hybrid storage layouts. We demonstrate that by using a Column Sketch, the select operator in modern analytic systems attains better CPU efficiency and less data movement than state-of-the-art storage and indexing schemes. Compared to standard scans, Column Sketches provide an improvement of 3x-6x for numerical attributes and 2.7x for categorical attributes. Compared to state-of-the-art scan accelerators such as Column Imprints and BitWeaving, Column Sketches perform 1.4 - 4.8× better.
Editor: New York, NY, USA: ACM
Idioma: Inglês

Voltar para lista de resultados

Realização: Logos de Redes Sociais:

Column Sketches: A Scan Accelerator for Rapid and Robust Predicate Evaluation

Hentschel, Brian ; Kester, Michael S. ; Idreos, Stratos

New York, NY, USA: ACM

Buscando em bases de dados remotas. Favor aguardar.