Learn More
Solid-state devices (SSDs) have the potential to replace traditional hard disk drives (HDDs) as the de facto storage medium. Unfortunately, there are several decades of spinning-media assumptions embedded in the software stack as an " unwritten contract " [20]. In this paper, we revisit these system-level assumptions in light of SSDs and find that several(More)
We present a study of the effects of disk and memory corruption on file system data integrity. Our analysis fo-cuses on Sun's ZFS, a modern commercial offering with numerous reliability mechanisms. Through careful and thorough fault injection, we show that ZFS is robust to a wide range of disk faults. We further demonstrate that ZFS is less resilient to(More)
We introduce Membrane, a set of changes to the operating system to support restartable file systems. Membrane allows an operating system to tolerate a broad class of file system failures, and does so while remaining transparent to running applications; upon failure, the file system restarts, its state is restored, and pending application requests are(More)
We present Coerced Cache Eviction (CCE), a new method to force writes to disk in the presence of a disk cache that does not properly obey write-cache configuration or flush requests. We demonstrate the utility of CCE by building a new journaling mode within the Linux ext3 file system. When mounted in this discreet mode, ext3 uses CCEs to ensure that writes(More)
The lowly state of the art for file system checking and repair does not match what is needed to keep important data available for users. Current file system checkers, such as e2fsck, are complex pieces of imperfect code written in low-level languages. We introduce SQCK, a file system checker based on a declarative query language ; declarative queries are a(More)
The file system is one of the most critical components of the operating system. Almost all applications running in the operating system require file systems to be available for their proper operation. Though file-system availability is critical in many cases, very little work has been done on tolerating file system crashes. In this paper, we propose(More)
details). Gailly said, " I feel I have received more from the OS community than I gave, " and Adler then said " Ditto, " in a couple of the shortest acceptance speeches ever. long history in CS research, with many contributions in systems, including the first mention of clusters. Both men also spoke of Popek's dedication to his students. Kuen-ning explained(More)
The reliability of a file system considerably depends upon how it deals with on-disk data corruption. A file system should ideally be able to detect and recover from all kinds of data corruptions on disk. ZFS is a new filesys-tem that arrives almost a generation after the introduction of other desktop filesystems like ext and NTFS and makes strong claims(More)
  • 1