D. D. E. Long, B. R. Montague, and L.-F. Cabrera, "Swift/RAID: A distributed RAID system," Computing Systems, vol. 7, no. 3, 1994.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....data distribution schemes, reads require fragments from multiple servers. Moreover, the set of fragments must correspond to the same write operation or else the reconstituted data will be incoherent. Examples of distributed storage systems that use erasure coding include Zebra [21] SwiftRAID [35], Intermemory [10] Cheops [4] Myriad [9] and PASIS [59, 60] A challenge that must be confronted in the design of decentralized storage systems is that of partially completed write operations. Write operations in progress and incomplete write operations by clients that crash are both instances ....

Darrell D. E. Long, Bruce R. Montague, and Luis-Felipe Cabrera. Swift/RAID: a distributed RAID system. Computing Systems, 7(3):333--359. Usenix, Summer 1994.

....portions of the namespace must be redistributed amongst the servers, and all client mountpoints must be modified. Conventional wisdom says that such balancing is rarely performed. Many cluster file system designs also maintain data redundancy, either via replication [6, 24] or parity [2, 10, 15], to provide fault tolerance. With traditional servers, a single failure makes a portion of the namespace unavailable. Many cluster file system designs support both incremental growth and graceful degradation. As servers are added, files are migrated or replicated to them and their resources ....

Darrell D. E. Long, Bruce R. Montague, and Luis-Felipe Cabrera. Swift/RAID: a distributed RAID system. Computing Systems, 7(3):333-- 359. Usenix, Summer 1994.

....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 when some have failed, the ....

....parity computed across the stripe units. With these data distribution schemes, reads require fragments from multiple servers. Moreover, the set of fragments must correspond to the same write operation. Examples of decentralized storage systems that use erasure coding include Zebra [12] SwiftRAID [21], Intermemory [8] and Myriad [7] 2.1 Related work There has been much work on protocols for maintaining consistency across distributed servers. This section highlights some relevant related work. A common approach to dealing with the partial writer problem is two phase commit [11] This works ....

Darrell D. E. Long, Bruce R. Montague, and Luis-Felipe Cabrera. Swift/RAID: a distributed RAID system. Computing Systems, 7(3):333--359. Usenix, Summer 1994.

....meant to be more general purpose. Virtual disks [6] and logical disks [24] provide a device abstraction that hide the physical connectivity of the disk. These systems make a disk connected to a remote host to appear as if it is logically virtually connected to a local host. Striping related work [13, 27, 19] study the impact of device level and serverlevel parallelism on I O performance. These systems still treat disks as merely as storage. Our work will extend the function of any disk on the network to encompass local processing at the disk. Active Networks [39, 30] envisions code migration to the ....

D. D. E. Long, B. R. Montague, and L. Cabrera. Swift/RAID: A distributed RAID system. Computing Systems, 7(3):333--359, 1994.

....secure disks (NASD) at CMU [11] Virtual disks [4] and logical disks [8] provide a device abstraction that hide the physical connectivity of the disk. These systems make a disk connected to a remote host to appear as if it is logically virtually connected to a local host. Striping related work [10, 22, 15] study the impact of device level and server level parallelism on I O performance. These systems still treat disks as merely as storage. Our work will extend the function of any disk on the network to encompass local processing at the disk. Putting intelligent logic into a single chip has also ....

D. E. Long, B. R. Montague, and L. Cabrera. Swift/RAID: A distributed RAID system. Computing Systems, 7(3):333-359, 1994.

....UltraNet and to the client without passing through the server s memory. In this case, the server manages the data and initiates transfers, but need not be in the data path, a canonical example of third party transfers and the uses of networkattached peripherals. The Swift distributed RAID array[12, 44] was the first project to propose striping of data across multiple network connections as an alternative to striping on local disks. Their approach involves creating transfer plans to support the striping. The TickerTAIP distributed RAID array[14] is composed of network attached disks. It ....

D. D. E. Long, B. R. Montague, and L.-F. Cabrera. Swift/RAID: A distributed RAID system. Computing Systems, 7(3):333--359, 1994.

No context found.

D. D. E. Long, B. R. Montague, and L.-F. Cabrera, "Swift/RAID: A distributed RAID system," Computing Systems, vol. 7, no. 3, 1994.

No context found.

D. D. E. Long, B. R. Montague, and L.-F. Cabrera. Swift/RAID: A distributed RAID system. Computing Systems, 7(3):333--359, 1994.

No context found.

D. D. E. Long, B. R. Montague, and L.-F. Cabrera. Swift/RAID: A distributed RAID system. Computing Systems, 7(3):333--359, 1994.

....meant to be more general purpose. Virtual disks [5] and logical disks [24] provide a device abstraction that hide the physical connectivity of the disk. These systems make a disk connected to a remote host to appear as if it is logically virtually connected to a local host. Striping related work [12, 27, 18] study the impact of device level and server level parallelism on I O performance. These systems still treat disks as merely as storage. Our work will extend the function of any disk on the network to encompass local processing at the disk. Active Networks [38, 30] envisions code migration to the ....

D. D. E. Long, B. R. Montague, and L. Cabrera. Swift/RAID: A distributed RAID system. Computing Systems, 7(3):333--359, 1994. 25

....meant to be more general purpose. Virtual disks [5] and logical disks [24] provide a device abstraction that hide the physical connectivity of the disk. These systems make a disk connected to a remote host to appear as if it is logically virtually connected to a local host. Striping related work [12, 27, 18] study the impact of device level and server level parallelism on I O performance. These systems still treat disks as merely as storage. Our work will extend the function of any disk on the network to encompass local processing at the disk. Active Networks [38, 30] envisions code migration to the ....

D. D. E. Long, B. R. Montague, and L. Cabrera. Swift/RAID: A distributed RAID system. Computing Systems, 7(3):333--359, 1994.

....on variants of RAID schemes [PGK88] These schemes were much more storage efficient than mirroring, at the expense of access performance. Neither the scalability, nor the multicomputer environment was the target for that work. There were nevertheless attempts towards the latter goal, e.g. SS90] [MLC93]. An on going research project, the DIY RAID system, addresses the scalability of RAID schemes more specifically [A al97] The goal is a thousand node RAID configuration. Nevertheless, the system aims to provide 1 availability, or 2 availability at best, using the orthogonal RAID approach ....

Montague, B., Long, D., Cabrera, L. SWIFT/RAID A Distributed Raid System. IBM Res. Rep. RJ 9501, 1993, 25.

No context found.

Darrell D. E. Long, Bruce R. Montague, and Luis-Felipe Cabrera. Swift/RAID: A distributed RAID system. Computing Systems, 7(3):333-359, 1994.

No context found.

D. D. E. Long, B. R. Montague, and L.-F. Cabrera. Swift/RAID: a distributed RAID system. Technical Report UCSC-CRL-94-06, University of California, Santa Cruz, 1994.

No context found.

D. D. E. Long, B. R. Montague, and L.-F. Cabrera. Swift/RAID: a distributed RAID system. Technical Report UCSC-CRL-94-06, University of California, Santa Cruz, 1994.

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