| Item type |
SIG Technical Reports(1) |
| 公開日 |
2016-02-22 |
| タイトル |
|
|
タイトル |
Recursive Incremental Computation for Efficient Window Aggregate over Array Database |
| タイトル |
|
|
言語 |
en |
|
タイトル |
Recursive Incremental Computation for Efficient Window Aggregate over Array Database |
| 言語 |
|
|
言語 |
eng |
| キーワード |
|
|
主題Scheme |
Other |
|
主題 |
データベースとKVS |
| 資源タイプ |
|
|
資源タイプ識別子 |
http://purl.org/coar/resource_type/c_18gh |
|
資源タイプ |
technical report |
| 著者所属 |
|
|
|
Graduate School of Systems and Information Engineering, University of Tsukuba／CREST JST CREST |
| 著者所属 |
|
|
|
Faculty of Engineering, Information and Systems, University of Tsukuba／CREST JST CREST |
| 著者所属 |
|
|
|
Faculty of Engineering, Information and Systems, University of Tsukuba／CREST JST CREST |
| 著者所属(英) |
|
|
|
en |
|
|
Graduate School of Systems and Information Engineering, University of Tsukuba / CREST JST CREST |
| 著者所属(英) |
|
|
|
en |
|
|
Faculty of Engineering, Information and Systems, University of Tsukuba / CREST JST CREST |
| 著者所属(英) |
|
|
|
en |
|
|
Faculty of Engineering, Information and Systems, University of Tsukuba / CREST JST CREST |
| 著者名 |
Li, Jiang
Hideyuki, Kawashima
Osamu, Tatebe
|
| 著者名(英) |
Li, Jiang
Hideyuki, Kawashima
Osamu, Tatebe
|
| 論文抄録 |
|
|
内容記述タイプ |
Other |
|
内容記述 |
An array database is effective for managing and analyzing multi-dimensional scientific big data, and the window aggregate is an important operator in array databases. This paper proposes an efficient method that exploits the scheme of incremental computation and accelerates the execution of window aggregate considerably. Six types of aggregates are improved using different design of buffer tools to eliminate redundant computation. Our proposed recursive incremental computation method completely eliminates all redundant computation. Proposed method is fully implemented in SciDB. Evaluation is conducted on real scientific data as NASA MODIS data. The proposed method achieves performance improvements of 10x for the real application in earth science, comparing with SciDB's built-in window operator. The results align with our time-complexity analysis results. |
| 論文抄録(英) |
|
|
内容記述タイプ |
Other |
|
内容記述 |
An array database is effective for managing and analyzing multi-dimensional scientific big data, and the window aggregate is an important operator in array databases. This paper proposes an efficient method that exploits the scheme of incremental computation and accelerates the execution of window aggregate considerably. Six types of aggregates are improved using different design of buffer tools to eliminate redundant computation. Our proposed recursive incremental computation method completely eliminates all redundant computation. Proposed method is fully implemented in SciDB. Evaluation is conducted on real scientific data as NASA MODIS data. The proposed method achieves performance improvements of 10x for the real application in earth science, comparing with SciDB's built-in window operator. The results align with our time-complexity analysis results. |
| 書誌レコードID |
|
|
収録物識別子タイプ |
NCID |
|
収録物識別子 |
AN10444176 |
| 書誌情報 |
研究報告システムソフトウェアとオペレーティング・システム（OS）
巻 2016-OS-136,
号 5,
p. 1-15,
発行日 2016-02-22
|
| ISSN |
|
|
収録物識別子タイプ |
ISSN |
|
収録物識別子 |
2188-8795 |
| Notice |
|
|
|
SIG Technical Reports are nonrefereed and hence may later appear in any journals, conferences, symposia, etc. |
| 出版者 |
|
|
言語 |
ja |
|
出版者 |
情報処理学会 |
IPSJ-OS16136005.pdf (1.4 MB)
