Seltzer, M., Bostic, K., McKusick, M., and Staelin, C. An Implementation of a Log-Structured File System for UNIX. In Proc. of the 1993. Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
First 50 documents Next 50
PersonalRAID: Mobile Storage for Distributed and.. - Sobti, Garg.. (2002) (2 citations) (Correct)
....may complement the mobile storage device to address this limitation. 3.8. 3 Limitations of the Log Structured Organization The potential disadvantages of the log structured organization are the possible destruction of read locality and the cost of segment cleaning (or disk garbage collection) [13, 14]. PersonalRAID mainly targets personal computing workloads, which are often bursty and leave ample idle time for cleaning. The cleaning overhead can be further reduced by using techniques like freeblock scheduling [7] and by buying bigger disks and keeping the disk utilization low. The base LLD ....
Seltzer, M., Bostic, K., McKusick, M., and Staelin, C. An Implementation of a LogStructured File System for UNIX. In Proc. of the 1993.
Bridging the Information Gap in Storage Protocol Stacks - Denehy, Arpaci-Dusseau.. (2002) (9 citations) (Correct)
....in part or full to file systems built above, and provides on line performance and failure characteristics. We call this layer the Exposed RAID layer (ERAID) To take advantage of the information provided by ERAID, we introduce an Informed LFS (ILFS) an enhancement of a log structured file system [30, 37]. By combining the performance and failure information presented by ERAID along with file system specific knowledge, ILFS is more flexible and manageable than a traditional file system, and can deliver higher performance and availability as well. For example, adding a disk to ILFS on line is ....
....of our implementation techniques are general and could be applied to other file systems, and hope to investigate doing so in the future. Those interested in general LFS file system performance issues should consult the work of Rosenblum and Ousterhout [30] or subsequent research by Seltzer et al. [37, 38]. All of our software was developed within the context of the NetBSD 1.5 operating system. ERAID was implemented as a set of hooks on the lower level blockdriver calls, and is described in more detail in Section 4. ILFS was implemented by extending the NetBSD version of LFS, which is based on the ....
[Article contains additional citation context not shown here]
M. Seltzer, K. Bostic, M. K. McKusick, and C. Staelin. An Implementation of a Log-Structured File System for UNIX. In Proceedings of the 1993.
Mixed-Media File Systems - Bosch (1999) (9 citations) (Correct)
....UNIX device drivers. By making VFS available under Clockwise, each of the incorporated file systems is also available under Clockwise. Currently the following file systems are incorporated in VFS: McKusick s FFS, Linux EXT2FS, a Memory FS (MFS) a Union FS (union) and an implementation of BSD LFS [89]. NetBSD is a public domain implementation of the Berkeley BSD4.4 system. 82 Standard VFS posix lfs ffs ext2 fs NFS 2 NFS 2 posix lfs ffs ext2 fs Clockwise if Clockwise if Figure 5.3: Virtual File System organization. Porting VFS to Clockwise Nemesis Porting VFS to ....
Margo Seltzer, Keith Bostic, Marshall Kirk McKusick, and Carl Staelin. An Implementation of a Log-Structured File System for UNIX. USENIX Technical Conference Proceedings (San Diego, CA), pages 307--26. USENIX, Winter 1993.
Track-aligned Extents: Matching Access Patterns to.. - Schindler, Griffin, .. (2002) (9 citations) (Correct)
....Fortunately, doing so is straightforward. In an LFS, the segment usage table records information about each segment. In the SpriteLFS implementation [33] this table is kept as an in memory kernel structure and is stored in the checkpoint region of the file system. The BSD LFS implementation [36] stores this table in a special file called the IFILE. Because of its frequent use, this file is almost always in the file system s cache. Variable sized segments can be supported by augmenting the per segment information in the segment usage table with a starting location (the LBN) and length. ....
Margo Seltzer, Keith Bostic, Marshall Kirk McKusick, and Carl Staelin. An implementation of a log-structured file system for UNIX. Winter USENIX Technical Conference (San Diego, CA, 25 29 January 1993.
The Architecture of eNVy, A Non-Volatile, Main Memory Storage System - Wu (1994) (1 citation) (Correct)
....An application can roll back a transaction simply by copying data back from Flash. In order to 42 implement this feature, the controller has to keep track of the location of the shadow copies and protect them from being cleaned. This function has been studied in the context of log file systems [14] which use similar data movement primitives. A final hardware optimization is to use the comparators normally used to verify write data as a general purpose search engine. Since 256 bytes can be examined each cycle, this kind of search can be over an order of magnitude faster than a sequential ....
.... this work is to provide better operating systems support for mobile computers rather than higher performance (although that may be possible too) Finally, the problem of cleaning and how to minimize write costs for data reda marion has been studied in the context of log structured file systems [13, 14]. eNVy s 47 cleaning algorithm is closely related to their algorithms, but has more freedom in what it can do because random accesses in eNVy are much faster than the disks the file systems are built on. Conclusion Flash memory is becoming an increasingly important storage medium. Its high ....
Seltzer, M., Bostic, K., McKusick, M. K., Staelin, C. An Implementation of a Log Structured File System for UNIX. P'oceedigs of the 1993.
eNVy: A Non-Volatile, Main Memory Storage System - Wu, Zwaenepoel (1994) (17 citations) (Correct)
....copy. An application can roll back a transaction simply by copying data back from Flash. In order to implement this feature, the controller has to keep track of the location of the shadow copies and protect them from being cleaned. This function has been studied in the context of log file systems [14] which use similar data movement primitives. 7 Related Work There has been other work done in areas relevant to eNVy. Researchers at MITL [2] argue that Flash has interesting benefits as a memory mapped storage medium, and suggest a similar copy on write scheme handled by software in the ....
....memory mapped storage medium, and suggest a similar copy on write scheme handled by software in the operating system rather than a hardware controller although with lower performance. Others have studied the problem of how to minimize write costs for data reclamation in log structured file systems [13, 14]. The benefits possible from having fast access to stable storage were demonstrated by advocates of non volatile write caches [1, 3, 12] Finally, exten sive work on in memory data structures and other aspects of retrieval of data in large memory arrays have been done by researchers in the main ....
M. Seltzer, K. Bostic, M. K. McKusick, and C. Staelin. An implementation of a log structured file system for unix. In Proceedings of the 1993.
Recovery Oriented Computing (ROC): Motivation.. - Patterson, Brown, .. (2002) (5 citations) (Correct)
....for fast post crash recovery of large services like Microsoft Exchange. In the research community, recovery enhancing techniques have emerged serendipitously from work that was originally performancefocused, as in the development of journaling, logging and soft update file systems [Rosenblum92] Seltzer93] Hitz95] Recovery oriented work in the OS community is rarer but present. Much of it focuses on the ability to restart quickly after failures. An early example is Sprites Recovery Box , in which the OS uses a protected area of non volatile memory to store crucial state needed for fast recovery ....
M. Seltzer, K. Bostic, M. K. McKusick, and C. Staelin. An Implementation of a LogStructured File System for UNIX. Proc. of the Winter
TDB: A Database System for Digital Rights Management - Vingralek, Maheshwari, Shapiro (2001) (1 citation) (Correct)
....updated chunks to the tail of the log destroys clustering of related chunks. In fact, log structured storage trades off read performance for a better write performance. # Poor performance when cleaning. Performance studies based on TPC B showed a 40 performance reduction due to log cleaning [31]. Because DRM databases are relatively small, the working data set can be cached in memory, minimizing the cost of unclustered reads. The primary performance bottleneck then becomes writes, for which log structured systems are optimized. Because DRM database systems are embedded in devices used ....
M. Seltzer, K. Bostic, M. McKusick, and C. Staelin. An implementation of a log-structured file system for UNIX. In Proceedings of the
Logging versus Soft Updates: Asynchronous.. - Seltzer, Ganger.. Self-citation (Seltzer Mckusick) (Correct)
....(small I Os and seeks between accesses to small files) so much of the literature has focused on addressing these first two issues in the context of writes. The first of these barriers (small I O sizes) has largely been overcome by clustering sequential reads and writes to the same file [Peacock88, McVoy91, Seltzer93]. This solves the problem for large files. Log structured file systems optimize all writes and avoid synchronous meta data updates [Rosenblum92] They also improve read performance when the read pattern matches the write pattern. The Co locating FFS [Ganger97] solves the inter file access problem ....
....case, it is necessary for LFS to make sure it can recover the file system to a consistent state. The original LFS implementation [Rosenblum92] solved this problem by adding small log entries to the beginning of segments, applying a logging approach to the problem. A later implementation of LFS [Seltzer93] used a simple transaction like interface to make segments temporary, until all the meta data necessary to ensure the recoverability of the file system was on disk. 3 Logging Implementations In this section, we describe two different implementations of logging applied to the fast file system. ....
[Article contains additional citation context not shown here]
Seltzer, M., Bostic, K., McKusick, M.K., Staelin, C. "An Implementation of a Log-Structured File System for UNIX," Proceedings of the 1993 USENIX Winter Technical Conference, pp. 307--326. San Diego, CA, Jan. 1993.
Virtual Log Based File Systems for a Programmable Disk - Wang, Anderson, Patterson (1999) (30 citations) (Correct)
No context found.
Seltzer, M., Bostic, K., McKusick, M., and Staelin, C. An Implementation of a Log-Structured File System for UNIX. In Proc. of the 1993.
Storage Systems Support for Multimedia Applications - Halvorsen, Griwodz, al. (2003) (1 citation) (Correct)
No context found.
Margo Seltzer, Keith Bostic, Marshall K. McKusick, and Carl Staelin. An implementation of a logstructured file system for unix. In USENIX Technical Conf., pages 307--326, San Diego, CA, USA, January 1993.
USENIX Association - Usenix (2004) (2 citations) (Correct)
No context found.
Margo I. Seltzer, Keith Bostic, et al. An Implementation of a Log-Structured File System for UNIX. Proceedings of the 1993.
Coordinating Access to Computation and Data in Distributed Systems - Thain (2004) (Correct)
No context found.
Margo Seltzer, Keith Bostic, Marshall Kirk McKusick, and Carl Staelin. An Implementation of a Log-Structured File System for UNIX. In Proceedings of the USENIX Winter Technical Conference (USENIX Winter '93), pages 307--326, San Diego, CA, Jan 1993.
Coordinating Access to Computation and Data in Distributed Systems - Thain (2004) (Correct)
No context found.
Margo Seltzer, Keith Bostic, Marshall Kirk McKusick, and Carl Staelin. An Implementation of a Log-Structured File System for UNIX. In Proceedings of the USENIX Winter Technical Conference (USENIX Winter '93), pages 307--326, San Diego, CA, Jan 1993.
Cryptographic Access Control in a Distributed File System - Christian (2003) (Correct)
No context found.
M. I. Seltzer, K. Bostic, M. K. McKusick, and C. Staelin. An implementation of a log-structured file system for UNIX. In USENIX Winter, pages 307--326, 1993.
Data Placement For Copy-On-Write Using Virtual Contiguity - Peterson (2002) (Correct)
No context found.
M. Seltzer, K. Bostic, M. K. McKusick, and Carl Staelin. An implementation of a logstructured file system for UNIX. In Proceedings of the Winter 1993.
Performance Analysis and Optimization of the Hurricane File System.. - Tam (2003) (Correct)
No context found.
Margo Seltzer, Keith Bostic, Marshall Kirk McKusick, and Carl Staelin. An implementation of a logstructured file system for UNIX. In Winter USENIX. USENIX Association, January 1993.
WOLF–A Novel Reordering Write Buffer to Boost the Performance of .. - Wang, Hu (2002) (1 citation) (Correct)
No context found.
SELTZER, M., BOSTIC, K., MCKUSICK, M. K., AND STAELIN, C. An implementation of a log-structured file system for UNIX. In Proceedings of Winter 1993.
Soft Updates: A Solution to the Metadata Update Problem in.. - Gregory Ganger Yale (1995) (12 citations) (Correct)
No context found.
M. Seltzer, K. Bostic, M. McKusick, C. Staelin, "An Implementation of a Log-Structured File System for UNIX", Winter USENIX Conference,January 1993, pp. 201--220.
DualFS: a New Journaling File System without Meta-Data.. - Piernas, Cortes, Garcia (2002) (Correct)
No context found.
M. Seltzer, K. Bostic, M. K. McKusick, and C. Staelin. An implementation of a log-structured file system for UNIX. In Proc. of the Winter 1993.
Meta-data Snapshotting: A Simple Mechanism for File.. - Soares, Krieger, Silva (2003) (Correct)
No context found.
M. I. Seltzer, K. Bostic, M. K. McKusick, and C. Staelin, "An Implementation of a Log-Structured File System for UNIX," in USENIX Winter, 1993, pp. 307--326. [Online]. Available: http://citeseer.nj.nec.com/seltzer93implementation.html
Ext3cow: The Design, Implementation, and Analysis of Metadata .. - Peterson, Burns (2003) (Correct)
No context found.
M. Seltzer, K. Bostic, M. K. McKusick, and Carl Staelin. An implementation of a log-structured file system for UNIX. In Proceedings of the Winter 1993.
Meta-data Snapshotting: A Simple Mechanism for - File System Consistency (Correct)
No context found.
M. I. Seltzer, K. Bostic, M. K. McKusick, and C. Staelin, "An Implementation of a Log-Structured File System for UNIX," in USENIX Winter, 1993, pp. 307--326. [Online]. Available: http://citeseer.nj.nec.com/seltzer93implementation.html
Design Evolution of the EROS Single-Level Store - Shapiro, Adams (2002) (1 citation) (Correct)
No context found.
M. Seltzer, K. Bostic, M. K. McKusick, and C. Staelin. An implementation of a logstructured file system for UNIX. In Proceedings of the USENIX Winter 1993 Technical Conference, pages 307--326, San Diego, CA, USA, 25--29 1993.
SOFTWARE---PRACTICE AND EXPERIENCE, VOL. 24(11), 981--999.. - Techniques For File (Correct)
No context found.
Margo Seltzer, Keith Bostic, Marshall Kirk McKusick, and Carl Staelin, `An implementation of a logstructured file system for UNIX', Proceedings of Winter 1993 USENIX, January 1993, pp. 307--326.
First 50 documents Next 50
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC