Failure-Atomic msync(): A Simple and Efficient Mechanism for Preserving the Integrity of Durable Data

DMCA

Failure-Atomic msync(): A Simple and Efficient Mechanism for Preserving the Integrity of Durable Data

Cached

Download Links

by Stan Park , Terence Kelly , Kai Shen

Citations:

7 - 1 self

BibTeX

@MISC{Park_failure-atomicmsync():,
    author = {Stan Park and Terence Kelly and Kai Shen},
    title = {Failure-Atomic msync(): A Simple and Efficient Mechanism for Preserving the Integrity of Durable Data},
    year = {}
}

OpenURL

Abstract

Preserving the integrity of application data across updates is difficult if power outages and system crashes may occur during updates. Existing approaches such as relational databases and transactional key-value stores restrict programming flexibility by mandating narrow data access interfaces. We have designed, implemented, and evaluated an approach that strengthens the semantics of a standard operating system primitive while maintaining conceptual simplicity and supporting highly flexible programming: Failureatomic msync() commits changes to a memory-mapped file atomically, even in the presence of failures. Our Linux implementation of failure-atomic msync() has preserved application data integrity across hundreds of whole-machine power interruptions and exhibits good microbenchmark performance

Keyphrases

failure-atomic msync durable data efficient mechanism power outage application data integrity transactional key-value store exhibit good microbenchmark performance failureatomic msync linux implementation flexible programming application data memory-mapped file whole-machine power interruption narrow data access interface relational database system crash conceptual simplicity

DMCA