Adding parity to the Linux ext3 file system

Authors:
Himani Apte;Meenali Rungta
Affiliations:
University of Wisconsin, Madison;University of Wisconsin, Madison
Venue:
ACM SIGOPS Operating Systems Review
Year:
2007

Citing 3
Cited 0

A case for redundant arrays of inexpensive disks (RAID)

SIGMOD '88 Proceedings of the 1988 ACM SIGMOD international conference on Management of data
Measurements of a distributed file system

SOSP '91 Proceedings of the thirteenth ACM symposium on Operating systems principles
Analysis and evolution of journaling file systems

ATEC '05 Proceedings of the annual conference on USENIX Annual Technical Conference

Quantified Score

Hi-index	0.00

Visualization

Abstract

Modern disks no longer operate in a simple "fail-stop" manner, yet commodity operating systems assume they do. We design and implement a parity based approach to improve the robustness of journaling file systems. We modify the existing ext3 file system for data and ordered journaling modes to incorporate parity and call it the "Parity File System". Using PFS, we are able to recover from a single latent sector error or silent block corruption within a given file. We show that the performance overhead for PFS compared to ext3 is minimal while the robustness is significantly improved.