Mahadev Satyanarayanan, John H. Howard, David A. Nichols, Robert N. Sidebotham, Alfred Z. Spector, and Michael J. West. The ITC distributed file system: Principles and design. In Proceedings of the Tenth Symposium on Operating Systems Principles, pages 35--50. ACM, December 1985.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....this optimism will generally pay off. Although it is possible to implement parallel kernel operations, we have not done so in the current implementation. Coordinating Accesses to View Entries Currently, users specify the entity type in their user profile. Entities can be defined per Ficus volume [13]. The entity type is encoded in the entity parameter in Figure 1. This parameter is only used by the readViewEntry and writeViewEntry routines. Therefore, the cost and the complexity of providing a consistent view to a specific entity depends on the cost of reading and writing view entries. For ....

Mahadev Satyanarayanan, John H. Howard, David A. Nichols, Robert N. Sidebotham, Alfred Z. Spector, and Michael J. West. The ITC distributed file system: Principles and design. In Proceedings of the Tenth Symposium on Operating Systems Principles, pages 35--50. ACM, December 1985.

....blocks in the local memory of a file server and the client machine and to reuse them when it can be ascertained that they are still valid. Client memory based caching can greatly improve performance, as discussed in [Howard et al. 1988] Kazar 1988] Nelson et al. 1988] Ousterhout et al. 1988] [Satyanarayan et al. 1985] and [Schroeder et al. 1985] In RHODOS, the caches on client computers will be in their main memory. In the next subsection, we addresses the following design issues related to cache management: ffl granularity of cached data, ffl the manner of propagating modifications made to cached copies, ....

Satyanarayan, M., Howard, J.H., Nichols, D.N., Sidebotham, R.N., Spector, A.Z., and West, M.J. The ITC Distributed File System: Principles and Design. Proceedings of the 10th ACM Symposium on Operating Systems Principles. December, pages 35-50.

.... interface used in Ficus is referred to [5, 4] Third, the optimistic consistency policy and reconciliation mechanism used for incorporating uncoordinated updates to replicated directories is exploited to manage the super tree of replicated volumes (analogous to the AFS volume location database [17]) Thus no new replication and consistency policy or mechanism is needed for this fairly general name service. Not only does this design reduce implementation and simplify the architecture, it provides the appropriate semantics and availability for naming data. 1.1 Relation to Other Work Ficus is ....

....prevent the Locus approach from scaling beyond a relatively small number of sites. Further, Ficus is a modular extension to the Unix file system, not a full distributed operating system. Ficus derives its notion of a volume as a granularity of sub tree management from the Andrew File System [17]. It shares many of the same goals as AFS for scale, and Coda [18] for reliability and availability via optimistic replica management. Coda makes similar use of version vectors for update update conflict detection. Ficus differs from AFS and Coda in its basic model of a distributed file system; ....

Mahadev Satyanarayanan, John H. Howard, David A. Nichols, Robert N. Sidebotham, Alfred Z. Spector, and Michael J. West. The ITC distributed file system: Principles and design. In Proceedings of the Tenth Symposium on Operating Systems Principles, pages 35--50. ACM, December 1985.

....servers to improve system scalability The clue tables mechanism implements dynamic binding between clients and servers upon file open to achieve load balancing among servers. This is the main distinction of the clue tables mechanism when compared with other distributed file systems such as AFS[1, 5], Coda[6] Locus[7] V kernel[8] Amoeba[9] Ficus[10] and Deceit[11] that also implement dynamic binding. With clue tables, a client, upon opening a file, multicasts access requests to servers that have a replica of the file. If more than one server acknowledges the request, the client always ....

Mahadev Satyanarayanan, John H. Howard, David A Nichols, Robert N. Sidebotham, Alfred Z. Spector, and Michael J. West. The ITC Distributed File System: Principles and Design. In Proc. of the 10th Symposium on Operating Systems Principles, pages 35--50. ACM, Dec. 1985.

No context found.

M. Satyanarayan, J. Howard, D. Nichols, R. Sidebotham, A. Spector, and M. West. The ITC Distributed File System: Principles and Design. In Tenth Symposium on Operating Systems, pages 35--50, 1985.

No context found.

Mahadev Satyanarayanan, John H. Howard, David A. Nichols, Robert N. Sidebotham, Alfred Z. Spector, and Michael J. West, `The ITC distributed file system: Principles and design', Proceedings of the Tenth Symposium on Operating Systems Principles. ACM, December 1985, pp. 35--50.

No context found.

Mahadev Satyanarayanan, John H. Howard, David A. Nichols, Robert N. Sidebotham, Alfred Z. Spector, and Michael J. West, `The ITC distributed file system: Principles and design', Proceedings of the Tenth Symposium on Operating Systems Principles. ACM, December 1985, pp. 35--50.

No context found.

Mahadev Satyanarayanan, John H. Howard, David A. Nichols, Robert N. Sidebotham, Alfred Z. Spector, and Michael J. West. "The ITC Distributed File System: Principles and Design." In Proceedings of the Tenth Symposium on Operating Systems Principles, pp. 35--50. ACM, December 1985.

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