Can SSDs help reduce random i/os in hash joins?

Authors:
Liang Huai Yang;Mingchao Liu;Yifan Pan;Weihua Gong;Simon Stannus
Affiliations:
School of Computer Science & Technology, Zhejiang Univ. of Technology, Hangzhou, China;School of Computer Science & Technology, Zhejiang Univ. of Technology, Hangzhou, China;School of Computer Science & Technology, Zhejiang Univ. of Technology, Hangzhou, China;School of Computer Science & Technology, Zhejiang Univ. of Technology, Hangzhou, China;School of Computing and Information Systems, University of Tasmania, Tasmania, Australia
Venue:
WAIM'13 Proceedings of the 14th international conference on Web-Age Information Management
Year:
2013

Citing 5
Cited 0

Towards Eliminating Random I/O in Hash Joins

ICDE '96 Proceedings of the Twelfth International Conference on Data Engineering
Fast scans and joins using flash drives

Proceedings of the 4th international workshop on Data management on new hardware
Query processing techniques for solid state drives

Proceedings of the 2009 ACM SIGMOD International Conference on Management of data
DigestJoin: Exploiting Fast Random Reads for Flash-Based Joins

MDM '09 Proceedings of the 2009 Tenth International Conference on Mobile Data Management: Systems, Services and Middleware
Data management over flash memory

Proceedings of the 2011 ACM SIGMOD International Conference on Management of data

Quantified Score

Hi-index	0.00

Visualization

Abstract

A hybrid drive combines the features of SSDs and HDDs in the same unit by using SSD as the intermediate cache. In light of its promising features, we propose a new algorithm called CGHJ(Cached Grace Hash Join) for hybrid drives, which reduces hard disk random I/Os that occur in the partitioning phase of traditional Grace hash joins by caching the segments of buckets in the SSD and migrating these segments of each partition contiguously to the HDD. Experiment results show that CGHJ can greatly reduce random disk I/Os in the case of small joining working spaces or larger relations and improve hash join performance.