Back to Search Start Over

Database Cracking: Fancy Scan, not Poor Man’s Sort!

Authors :
Pirk, H. (Holger)
Petraki, E. (Eleni)
Idreos, S. (Stratos)
Manegold, S. (Stefan)
Kersten, M.L. (Martin)
Pirk, H. (Holger)
Petraki, E. (Eleni)
Idreos, S. (Stratos)
Manegold, S. (Stefan)
Kersten, M.L. (Martin)
Source :
Proceedings of the Tenth International Workshop on Data Management on New Hardware
Publication Year :
2014

Abstract

Database Cracking is an appealing approach to adaptive indexing: on every range-selection query, the data is partitioned using the supplied predicates as pivots. The core of database cracking is, thus, pivoted partitioning. While pivoted partitioning, like scanning, requires a single pass through the data it tends to have much higher costs due to lower CPU efficiency. In this paper, we conduct an in-depth study of the reasons for the low CPU efficiency of pivoted partitioning. Based on the findings, we develop an optimized version with significantly higher (single-threaded) CPU efficiency. We also develop a number of multi-threaded implementations that are effectively bound by memory bandwidth. Combining all of these optimizations we achieve an implementation that has costs close to or better than an ordinary scan on a variety of systems ranging from low-end (cheaper than $300) desktop machines to high-end (above $60,000) servers.

Details

Database :
OAIster
Journal :
Proceedings of the Tenth International Workshop on Data Management on New Hardware
Notes :
application/pdf, English
Publication Type :
Electronic Resource
Accession number :
edsoai.on1251886931
Document Type :
Electronic Resource
Full Text :
https://doi.org/10.1145.2619228.2619232