ETRI-Knowledge Sharing Plaform

KOREAN
논문 검색
Type SCI
Year ~ Keyword

Detail

Journal Article Efficient Parallel Join Processing Exploiting SIMD in Multi-Thread Environments
Cited 3 time in scopus Download 5 time Share share facebook twitter linkedin kakaostory
Authors
Gilseok Hong, Seonghyeon Kang, Chang soo Kim, Jun-Ki Min
Issue Date
2018-03
Citation
IEICE Transactions on Information and Systems, v.E101-D, no.3, pp.659-667
ISSN
1745-1361
Publisher
일본, 전자정보통신학회 (IEICE)
Language
English
Type
Journal Article
DOI
https://dx.doi.org/10.1587/transinf.2017EDP7300
Project Code
17HS7600, Development of an Unified Data Engineering Technology for Large-scale Transaction Processing and Real-time Complex Analytics, Chang Soo Kim
Abstract
In this paper, we study parallel join processing to improve the performance of the merge phase of sort-merge join by integrating all parallelism provided by mainstream CPUs. Modern CPUs support SIMD instruction sets with wider SIMD registers which allows to process multiple data items per each instruction. Thus, we devise an efficient parallel join algorithm, called Parallel Merge Join with SIMD instructions (PMJS). In our proposed algorithm, we utilize data parallelism by exploiting SIMD instructions. And we also accelerate the performance by avoiding the usage of conditional branch instructions. Furthermore, to take advantage of the multiple cores, our proposed algorithm is threaded in multi-thread environments. In our multi-thread algorithm, to distribute workload evenly to each thread, we devise an efficient workload balancing algorithm based on the kernel density estimator which allows to estimate the workload of each thread accurately.
KSP Keywords
Instruction sets, Join algorithm, Join processing, Multi-thread, Multiple data, SIMD instruction, Sort-merge Join, Workload Balancing, balancing algorithm, data parallelism, kernel density estimator