ALS 2000 Abstract
Scalability and Failure Recovery in a Linux Cluster File System
Kenneth W. Preslan, Andrew Barry, Jonathan Brassow,
Michael Declerck, A.J. Lewis, Adam Manthei,
Ben Marzinski, Erling Nygaard, Seth Van Oort,
David Teigland, Mike Tilstra, Steven Whitehouse,
and Matthew O'Keefe , Sistina Software, Inc.
In this paper we describe how we implemented journaling and recovery in the Global File System (GFS), a
shared-disk, cluster file system for Linux. We also present our latest performance results for a 16-way Linux cluster.
Traditional local file systems support a persistent name space by creating a mapping between blocks found on disk
drives and a set of files, file names, and directories. These file systems view devices as local: devices are not shared so
there is no need in the file system to enforce device sharing semantics. Instead, the focus is on aggressively caching and
aggregating file system operations to improve performance by reducing the number of actual disk accesses required for
each file system operation.
- View the full text of this paper in
HTML form, and
- If you need the latest Adobe Acrobat Reader, you can download it from Adobe's site.
- To become a USENIX Member, please see our Membership Information.