File Download
  Links for fulltext
     (May Require Subscription)
  • Find via Find It@HKUL
Supplementary

Article: Parallel H-Tree Based Data Cubing on Graphics Processors

TitleParallel H-Tree Based Data Cubing on Graphics Processors
Authors
KeywordsSIMD
GPGPUs
Data cubing
H-tree
OLAP
Issue Date2012
PublisherScience in China Press. The Journal's web site is located at http://www.ijsi.org/ch/index.aspx
Citation
International Journal of Software and Informatics, 2012, v. 6 n. 1, p. 61-87 How to Cite?
AbstractAbstract:Graphics processing units (GPUs) have an SIMD architecture and have been widely used recently as powerful general-purpose co-processors for the CPU. In this paper, we investigate efficient GPU-based data cubing because the most frequent operation in data cube computation is aggregation, which is an expensive operation well suited for SIMD parallel processors. H-tree is a hyper-linked tree structure used in both top-k H-cubing and the stream cube. Fast H-tree construction, update and real-time query response are crucial in many OLAP applications. We design highly efficient GPU-based parallel algorithms for these H-tree based data cube operations. This has been made possible by taking effective methods, such as parallel primitives for segmented data and efficient memory access patterns, to achieve load balance on the GPU while hiding memory access latency. As a result, our GPU algorithms can often achieve more than an order of magnitude speedup when compared with their sequential counterparts on a single CPU. To the best of our knowledge, this is the first attempt to develop parallel data cubing algorithms on graphics processors.
Persistent Identifierhttp://hdl.handle.net/10722/160526
ISSN

 

DC FieldValueLanguage
dc.contributor.authorWang, Ben_US
dc.contributor.authorYu, Yen_US
dc.date.accessioned2012-08-16T06:13:27Z-
dc.date.available2012-08-16T06:13:27Z-
dc.date.issued2012en_US
dc.identifier.citationInternational Journal of Software and Informatics, 2012, v. 6 n. 1, p. 61-87en_US
dc.identifier.issn1673-7288-
dc.identifier.urihttp://hdl.handle.net/10722/160526-
dc.description.abstractAbstract:Graphics processing units (GPUs) have an SIMD architecture and have been widely used recently as powerful general-purpose co-processors for the CPU. In this paper, we investigate efficient GPU-based data cubing because the most frequent operation in data cube computation is aggregation, which is an expensive operation well suited for SIMD parallel processors. H-tree is a hyper-linked tree structure used in both top-k H-cubing and the stream cube. Fast H-tree construction, update and real-time query response are crucial in many OLAP applications. We design highly efficient GPU-based parallel algorithms for these H-tree based data cube operations. This has been made possible by taking effective methods, such as parallel primitives for segmented data and efficient memory access patterns, to achieve load balance on the GPU while hiding memory access latency. As a result, our GPU algorithms can often achieve more than an order of magnitude speedup when compared with their sequential counterparts on a single CPU. To the best of our knowledge, this is the first attempt to develop parallel data cubing algorithms on graphics processors.-
dc.languageengen_US
dc.publisherScience in China Press. The Journal's web site is located at http://www.ijsi.org/ch/index.aspx-
dc.relation.ispartofInternational Journal of Software and Informaticsen_US
dc.subjectSIMD-
dc.subjectGPGPUs-
dc.subjectData cubing-
dc.subjectH-tree-
dc.subjectOLAP-
dc.titleParallel H-Tree Based Data Cubing on Graphics Processorsen_US
dc.typeArticleen_US
dc.identifier.emailYu, Y: yzyu@cs.hku.hken_US
dc.identifier.authorityYu, Y=rp01415en_US
dc.description.naturelink_to_OA_fulltext-
dc.identifier.hkuros202192en_US
dc.identifier.volume6en_US
dc.identifier.issue1-
dc.identifier.spage61en_US
dc.identifier.epage87en_US
dc.publisher.placeChina-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats