E. K. Lee and C. A. Thekkath. Petal: distributed virtual disks. Architectural Support for Programming Languages and Operating Systems. Published as SIGPLAN Notices, 31(9):84--92, 1996.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....The more recent Elephant file system provides a more complete range of versioning options for recovery from user error [37] Its heuristics attempt to keep only those versions of a file that are most important to users. Many modern systems support snapshots to assist recovery from system failure [11, 19, 20, 25, 34]. Most closely related to CVFS are Spiralog [15, 21] and Plan9 [33] which use a log structured file system to do online backup by recording the entire log to tertiary storage. Chervenak, et al. performed an evaluation of several snapshot systems [10] Version control systems are user programs ....

E. K. Lee and C. A. Thekkath. Petal: distributed virtual disks. Architectural Support for Programming Languages and Operating Systems. Published as SIGPLAN Notices, 31(9):84--92, 1996.

....of the data item, the main difficulty is identifying and retaining the most recent instance. It is often necessary for a reader to contact multiple storage nodes to ensure that it sees the most recent instance. Examples of distributed storage systems that use this design include Harp [34] Petal [33], BFS [8] and Farsite [1] Some distributed storage systems spread data among storage nodes more space efficiently, using era2 sure coding or even simple striping. With striping, a data item is divided into fragments, all of which are needed to reconstruct the entire data item. With erasure ....

....codes when partial writes provide enough information to reconstruct the original data. A common approach to dealing with concurrency is to suppress it, either via leases [19] or optimistic concurrency control [31] Alternately, many non Byzantine tolerant systems (e.g. Harp [34] and Petal [33]) serialize their actions through a primary storage node, which becomes responsible for completing the update. An alternate approach to handling both problems is to have the data stored on storage nodes be immutable [50, 51] By definition, this eliminates the difficulties of updates for existing ....

Edward K. Lee and Chandramohan A. Thekkath. Petal: distributed virtual disks. Architectural Support for Programming Languages and Operating Systems (Cambridge, MA, 1--5 October 1996.

....functionality by transparently translating some client requests into redirected server requests. The same role can be played by a collection of small intermediaries at the front ends of the servers. 1 Introduction File services implemented as cooperating clusters of servers would be great [2, 10, 12, 13, 24]. They offer incremental scalability of storage capacity and performance. They can spread data amongst themselves so as to balance the workload. They can keep redundant data for fault tolerance. They can provide high end features with commodity components. For many years, the community has known ....

Edward K. Lee and Chandramohan A. Thekkath. Petal: distributed virtual disks. Architectural Support for Programming Languages and Operating Systems (Cambridge, MA, 1--5 October 1996.

....the official policies or endorsements, either expressed or implied, of the Air Force Research Laboratory or the U.S. Government. Keywords: Decentralized storage, consistency protocol, versioning servers, distributed file systems 1 Introduction Survivable storage systems (e.g. Petal [19], Myriad [7] SwiftRAID [21] PASIS [35] and Cheops [3] preserve and provide access to data even when a subset of storage nodes fail. The common architecture for such systems spreads data redundantly (either via replication or erasure coding) across a set of decentralized storage nodes. Even ....

....of the data item, the main difficulty is identifying and retaining the most recent instance. It is often necessary for a reader to contact multiple storage nodes to ensure that it sees the most recent instance. Examples of decentralized storage systems that use this design include Harp [20] Petal [19], BFS [6] and Farsite [5] Some decentralized storage systems spread data among storage nodes more space efficiently, using erasure coding or even simple striping. With striping, a data item is divided into fragments, all of which are needed to reconstruct the entire data item. With erasure ....

[Article contains additional citation context not shown here]

Edward K. Lee and Chandramohan A. Thekkath. Petal: distributed virtual disks. Architectural Support for Programming Languages and Operating Systems (Cambridge, MA, 1--5 October 1996.

....The more recent Elephant file system provides a more complete range of versioning options for recovery from user error [37] Its heuristics attempt to keep only those versions of a file that are most important to users. Many modern systems support snapshots to assist recovery from system failure [11, 18, 19, 24]. Most closely related to CVFS is Spiralog, which uses a log structured file system to do online backup by recording the entire log to tertiary storage [14, 20] Chervenak, et. al, performed an evaluation of several snapshot systems [10] Version control systems are user programs that implement a ....

Edward K. Lee and Chandramohan A. Thekkath. Petal: distributed virtual disks. Architectural Support for Programming Languages and Operating Systems (Cambridge, MA, 1--5 October 1996.

....a new standard for object based storage devices. The S4 interface is similar to these. The standard method of intrustion recovery is to keep a periodic backup of les on trusted storage. Several le systems simplify and extend this process by allowing a snapshot 27 to be taken of a le system [11, 12, 18]. This snapshot can then be accessed through the standard le system tools for le retrieval. Spiralog [13] uses a log structured le system to allow for backups to be made during system operation by simply recording the entire log to tertiary storage. While these systems are e ective in ....

Edward K. Lee and Chandramohan A. Thekkath. Petal: distributed virtual disks. Architectural Support for Programming Languages and Operating Systems (Cambridge, MA). Published as SIGPLAN Notices, 31(9):84-92, 1-5 October 1996.

....standard for object based storage 18 devices. The S4 interface is similar to this proposal. The standard method of intrusion recovery is to keep a periodic backup of les on trusted storage. Several le systems simplify and extend this process by allowing a snapshot to be taken of a le system [8, 9, 14]. This snapshot can then be accessed through the standard le system tools for le retrieval. Spiralog [10] uses a log structured le system to allow for backups to be made during system operation by simply recording the entire log to tertiary storage. While these systems are e ective in ....

Edward K. Lee and Chandramohan A. Thekkath. Petal: distributed virtual disks. Architectural Support for Programming Languages and Operating Systems (Cambridge, MA). Published as SIGPLAN Notices, 31(9):84-92, 1-5 October 1996.

....versioning of temporary les and other non critical data, it would also create a location where an intruder could temporarily store exploit tools without fear that they will be recovered. Versioning vs. snapshots: Self securing storage can be implemented with frequent copy on write snapshots [11, 12, 17] instead of versioning, so long as snapshots are kept for the full detection window. Although the audit log can still provide a record of what blocks are changed, snapshots often will not allow administrators to recover short lived les (e.g. exploit tools) or intermediate versions (e.g. system ....

....looking to create a new standard for object based storage devices. The S4 interface is similar to these. The standard method of intrusion recovery is to keep a periodic backup of les on trusted storage. Several le systems simplify this process by allowing a snapshot to be taken of a le system [11, 12, 17]. This snapshot can then be backed up with standard le system tools. Spiralog [13] uses a log structured le system to allow for backups to be made during system operation by simply recording the entire log to tertiary storage. While these systems are e ective in preventing the loss of ....

Edward K. Lee and Chandramohan A. Thekkath. Petal: distributed virtual disks. Architectural Support for Programming Languages and Operating Systems (Cambridge, MA). Published as SIGPLAN Notices, 31(9):84{ 92, 1-5 October 1996.

No context found.

E. K. Lee and C. A. Thekkath. Petal: distributed virtual disks. Architectural Support for Programming Languages and Operating Systems. Published as SIGPLAN Notices, 31(9):84--92, 1996.

No context found.

E. K. Lee and C. A. Thekkath. Petal: distributed virtual disks. Architectural Support for Programming Languages and Operating Systems. Published as SIGPLAN Notices, 31(9):84--92, 1996.

No context found.

Edward K. Lee and Chandramohan A. Thekkath. Petal: distributed virtual disks. Architectural Support for Programming Languages and Operating Systems (Cambridge, MA, 1--5 October 1996.

No context found.

Edward K. Lee and Chandramohan A. Thekkath. Petal: distributed virtual disks. Architectural Support for Programming Languages and Operating Systems (Cambridge, MA, 1--5 October 1996.

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