© The Institution of Engineering and Technology
A new snapshot file system for storage-class memory is presented. The file system reduces the overhead of snapshots, in terms of both space and the number of writes, by using in-place writes. Experimental results with real workloads show that the proposed file system uses 20–65% less space than the well-known copy-on-write snapshot technique, while providing the same reliability.
References
-
-
1)
-
Btrfs, http://btrfs.wiki.kernel.org.
-
2)
-
ZFS, http://opensolaris.org/os/community/zfs/.
-
3)
-
Hitz, D., Lau, J., Malcom, M.: `File system design for an NFS file server appliance', USENIX Winter Conf., 1994, p. 235–246.
-
4)
-
Venkataraman, S., Tolia, N., Ranganathan, P., Campbell, R.: `Consistent and durable data structures for non-volatile byte-addressable memory', Proc. USENIX Conf. on File and Storage Technologies, 2011.
-
5)
-
Freitas, R., Wilcke, W., Kurdi, B., Burr, G.: `Storage class memories', Tutorial at 7th USENIX Conf. on File and Storage Technologies, 2009.
-
6)
-
Condit, J., Nightingale, E.B., Frost, C., Ipek, E., Lee, B., Burger, D., Coetzee, D.: `Better I/O through byte-addressable, persistent memory', Proc. ACM Symp. on Operating Systems Principles, 2009.
http://iet.metastore.ingenta.com/content/journals/10.1049/el.2012.1016
Related content
content/journals/10.1049/el.2012.1016
pub_keyword,iet_inspecKeyword,pub_concept
6
6