Aether: A scalable approach to logging

• Presented at: 36th International Conference on Very Large Data Bases, Singapore, September 13-17, 2010
• Published in: Proceedings of the 36th International Conference on Very Large Data Bases, vol. 3, num. 1
• Publication date: 2010

The shift to multi-core hardware brings new challenges to database systems, as the software parallelism determines performance. Even though database systems traditionally accommodate simultaneous requests, a multitude of synchronization barriers serialize execution. Write-ahead logging is a fundamental, omnipresent component in ARIES-style concurrency and recovery, and one of the most important yet-to-be addressed potential bottlenecks, especially in OLTP workloads with small and frequent changes to data. In this paper, we identify four logging-related impediments to database system scalability. Each issue challenges different level in the software architecture: (a) the high volume of small-sized I/O requests may saturate the disk, (b) transactions hold locks while waiting for the log flush, (c) extensive context switching overwhelms the OS scheduler with threads executing log I/Os, and (d) contention appears as transactions serialize accesses to in-memory log data structures. We demonstrate these problems and address them with techniques that, when combined, comprise a holistic, scalable approach to logging. Our solution achieves a 20%-69% speedup over a modern database system when running log-intensive workloads, such as the TPC-B and TATP benchmarks. Moreover, it achieves a log insert throughput of small-sized log records of over 1.8GB/s on a modern single socket server, an order of magnitude higher than the traditional way of accessing the log using a single mutex.

Note:

SYSTEMS PUBLICATION_SHORE_MT

Reference

• EPFL-CONF-149436
• URL: http://dl.acm.org/citation.cfm?id=1920841.1920928
• URL: http://www.vldb2010.org/

Record created on 2010-06-23, modified on 2017-05-12