File Download
  Links for fulltext
     (May Require Subscription)
Supplementary

Conference Paper: ByteSlice: pushing the envelop of main memory data processing with a new storage layout

TitleByteSlice: pushing the envelop of main memory data processing with a new storage layout
Authors
KeywordsMain memory
Column store
Storage layout
SIMD
OLAP
Issue Date2015
PublisherThe Association for Computing Machinery.
Citation
The 2015 ACM SIGMOD/PODS Conference (SIGMOD’15), Melbourne, VC., Australia, 31 May-4 June 2015. In Conference Proceedings, 2015, p. 31-46 How to Cite?
AbstractScan and lookup are two core operations in main memory column stores. A scan operation scans a column and returns a result bit vector that indicates which records satisfy a filter. Once a column scan is completed, the result bit vector is converted into a list of record numbers, which is then used to look up values from other columns of interest for a query. Recently there are several inmemory data layout proposals that aim to improve the performance of in-memory data processing. However, these solutions all stand at either end of a trade-off — each is either good in lookup performance or good in scan performance, but not both. In this paper we present ByteSlice, a new main memory storage layout that supports both highly efficient scans and lookups. ByteSlice is a bytelevel columnar layout that fully leverages SIMD data-parallelism. Micro-benchmark experiments show that ByteSlice achieves a data scan speed at less than 0.5 processor cycle per column value — a new limit of main memory data scan, without sacrificing lookup performance. Our experiments on TPC-H data and real data show that ByteSlice offers significant performance improvement over all state-of-the-art approaches.
DescriptionAccepted Research Papers
Persistent Identifierhttp://hdl.handle.net/10722/214761
ISBN

 

DC FieldValueLanguage
dc.contributor.authorFeng, Z-
dc.contributor.authorLo, E-
dc.contributor.authorKao, B-
dc.contributor.authorXu, W-
dc.date.accessioned2015-08-21T11:54:28Z-
dc.date.available2015-08-21T11:54:28Z-
dc.date.issued2015-
dc.identifier.citationThe 2015 ACM SIGMOD/PODS Conference (SIGMOD’15), Melbourne, VC., Australia, 31 May-4 June 2015. In Conference Proceedings, 2015, p. 31-46-
dc.identifier.isbn978-1-4503-2758-9-
dc.identifier.urihttp://hdl.handle.net/10722/214761-
dc.descriptionAccepted Research Papers-
dc.description.abstractScan and lookup are two core operations in main memory column stores. A scan operation scans a column and returns a result bit vector that indicates which records satisfy a filter. Once a column scan is completed, the result bit vector is converted into a list of record numbers, which is then used to look up values from other columns of interest for a query. Recently there are several inmemory data layout proposals that aim to improve the performance of in-memory data processing. However, these solutions all stand at either end of a trade-off — each is either good in lookup performance or good in scan performance, but not both. In this paper we present ByteSlice, a new main memory storage layout that supports both highly efficient scans and lookups. ByteSlice is a bytelevel columnar layout that fully leverages SIMD data-parallelism. Micro-benchmark experiments show that ByteSlice achieves a data scan speed at less than 0.5 processor cycle per column value — a new limit of main memory data scan, without sacrificing lookup performance. Our experiments on TPC-H data and real data show that ByteSlice offers significant performance improvement over all state-of-the-art approaches.-
dc.languageeng-
dc.publisherThe Association for Computing Machinery.-
dc.relation.ispartofProceedings of the 2015 ACM SIGMOD International Conference on Management of Data (SIGMOD’15)-
dc.subjectMain memory-
dc.subjectColumn store-
dc.subjectStorage layout-
dc.subjectSIMD-
dc.subjectOLAP-
dc.titleByteSlice: pushing the envelop of main memory data processing with a new storage layout-
dc.typeConference_Paper-
dc.identifier.emailKao, B: kao@cs.hku.hk-
dc.identifier.authorityKao, B=rp00123-
dc.description.naturelink_to_OA_fulltext-
dc.identifier.doi10.1145/2723372.2747642-
dc.identifier.hkuros249901-
dc.identifier.spage31-
dc.identifier.epage46-
dc.publisher.placeUnited States-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats