| M. Beck, T. Moore, and J. S. Plank. An end-to-end approach to globally scalable network storage. In Proc. ACM SIGCOMM, Pittsburgh, PA, Aug. 2002. |
....systems. We plan to use the External Data Representation (XDR) or Porch Universal Checkpointing Format (UCF) 35, 32] for representing the checkpoints. We also plan to separate the storage nodes for checkpoints from the computa tional nodes for application execution by employing the eXNode [6] architecture. This will provide robust fault tolerant mechanism for withstanding the processor failures in SRS. We also intend to collaborate with the CUMULVS project [20] to provide a generic visualization architecture that will be used to monitor the execution of malleable applications. There ....
M. Beck, T. Moore, and J. Plank. An End-to-End Approach to Globally Scalable Network Storage. In ACM SIGCOMM 00 Conference, Pittsburgh, PA, USA, August 2002.
....File Transfer Protocol (FTP) which emphasizes security and high performance access to large files. NeST [9] is a single storage device encompassing multiple protocols, space management, replication, security, and quality of service. All of these devices export a plain file abstraction, The ExNode [8] is higher level structure built on top of a distributed set of IBP servers. The ExNode is analogous to a standard filesystem inode and provides what is described by its designers as something like a file. It logically merges extents spread across a distributed system into a single logical file, ....
M. Beck, T. Moore, and J. Plank. An end-to-end approach to globally scalable network storage. In ACM SIGCOMM, Pittsburgh, Pennsylvania, August 2002.
....[13] and L [9] are modern extensions of Landmark routing, whereas PeerNet to our knowledge is the first protocol for dynamic networks with similarity to Area Routing. For a survey of ad hoc routing, see [10] and for a survey of distributed hash tables, see [14] Recently, several efforts such as [15, 8] address some of the same issues as PeerNet but do so by adding functionality to the existing IP infrastructure. 2 Operations Overview The key idea in PeerNet is the separation of the identity and the address of a node. For now, we can assume that each PeerNet node has one unique identifier and ....
M. Beck, T. Moore, and J. S. Plank. An end-toend approach to globally scalable network storage. In SIGCOMM, August 2002.
No context found.
M. Beck, T. Moore, and J. S. Plank. An end-to-end approach to globally scalable network storage. In ACM SIGCOMM '02, Pittsburgh, August 2002.
.... Data Mover Depot NFU Applications Physical (Hardware) Access (OS) Figure 1: The Logistical Computing Stack The Network Storage Stack The IBP Depot: Again, the network storage stack has been previously explained and implemented, and its effectiveness has been demonstrated elsewhere [BMP02, PBB 01, ASP 02] However, an explanation will help in the explanation of the computation stack. The lowest non trivial layer of the storage stack is the Internet Backplane Protocol (IBP) storage depot. This is server daemon software and a client library that allows storage owners to insert ....
M. Beck, T. Moore, and J. S. Plank. An end-to-end approach to globally scalable network storage. In SIGCOMM 2002, Pittsburgh, August 2002.
....storage resources to be part of the wide area network in an efficient, flexible, sharable and scalable way. Its model, which achieves all these goals for data transmission, is the IP stack, and its guiding principle has been to follow the tenets laid out by End to End arguments [SRC84, RSC98, BMP02] Two fundamental principles of this layering are that each layer should (a) abstract the layers beneath it in a meaningful way, but (b) expose an appropriate amount of its own resources so that higher layers may abstract them meaningfully (see [BMP01, BMP02] for more detail on this approach) ....
.... by End to End arguments [SRC84, RSC98, BMP02] Two fundamental principles of this layering are that each layer should (a) abstract the layers beneath it in a meaningful way, but (b) expose an appropriate amount of its own resources so that higher layers may abstract them meaningfully (see [BMP01, BMP02] for more detail on this approach) IBP Logistical File System Local Access Physical Logistical Tools L Bone exNode Applications Figure 1: The Network Storage Stack 2.1 IBP The lowest layer of the storage stack that is globally accessible from the network is the Internet Backplane Protocol (IBP) ....
M. Beck, T. Moore, and J. S. Plank. An end-to-end approach to globally scalable network storage. In ACM SIGCOMM '02, Pittsburgh, August 2002.
....Lab at the University of Tennessee has developed the Network Storage Stack, which is modeled after the TCP IP stack. The purpose of this stack is to provide a scalably sharable storage service that will increase efficiency, performance, and functionality of distributed applications [4]. The paper is organized as follows: In section 2 we provide an overview of the components of the Network Storage Stack including the Logistical Runtime System, which is our testbed. In section 3 we describe applications that have been or are being built using this infrastructure. These ....
....allocate most of a new, empty drive and then as other applications consume disk space, the depot revokes Soft allocations to maintain enough disk space to prevent write to disk failures. Because IBP has a few, simple operations, we believe that the service can scale as easily as the Internet. See [4] for a more detailed look at IBP including types of allocations and other policies. 2.2 L Bone Once a user knows the hostnames and ports for some IBP depots, the user can allocate space and then write and read from it. To help users find IBP depots, part of the next layer up of the stack is the ....
Beck, M., Moore, T., and Plank, J.S. An end-to-end approach to globally scalable network storage. Proc. of ACM SIGCOMM '02, Pittsburgh, August 2002.
No context found.
M. Beck, T. Moore, and J. S. Plank. An end-to-end approach to globally scalable network storage. In Proc. ACM SIGCOMM, Pittsburgh, PA, Aug. 2002.
No context found.
M. Beck, T. Moore, and J. S. Plank. An end-to-end approach to globally scalable network storage. In Proc. ACM SIGCOMM, Pittsburgh, PA, Aug. 2002.
No context found.
M.Beck, T. Moore, J.S. Plank, An end-to-end approach to globally scalable network storage, in: Proceedings of the ACM SIGCOMM Conference, Pittsburgh, PA, 2002.
No context found.
M. Beck, T. Moore, and J. S. Plank. An end-to-end approach to globally scalable network storage. In Proc. ACM SIGCOMM, Pittsburgh, PA, Aug. 2002.
No context found.
M. Beck, T. Moore, and J. S. Plank. An end-to-end approach to globally scalable network storage. In Proc. of ACM SIGCOMM Conf., August 2002.
No context found.
Micah Beck, Terry Moore, and James Plank. An end-to-end approach to globally scalable network storage. In Proceedings of the ACM SIGCOMM '02 Conference on Communications Architectures and Protocols, August 2002.
No context found.
BECK, M., MOORE, T., PLANK, J.S. An End-to-End Approach to Globally Scalable Network Storage. In Proceedings of the ACM SIGCOMM Conference (Pittsburgh, PA, August 2002).
No context found.
Micah Beck, Terry Moore, and James Plank. An end-to-end approach to globally scalable network storage. In Proceedings of the ACM SIGCOMM '02 Conference on Communications Architectures and Protocols, August 2002.
No context found.
BECK, M., MOORE, T., PLANK, J.S. An End-to-End Approach to Globally Scalable Network Storage. In Proceedings of the ACM SIGCOMM Conference (Pittsburgh, PA, 2002).
No context found.
Micah Beck, Terry Moore, and James Plank. An end-to-end approach to globally scalable network storage. In Proceedings of the ACM SIGCOMM '02 Conference on Communications Architectures and Protocols, August 2002.
No context found.
M. Beck, T. Moore, and J. Plank. An end-to-end approach to globally scalable network storage. In ACM SIGCOMM 2002.
No context found.
BECK, M., MOORE, T., AND PLANK, J. S. An End-to-End Approach to Globally Scalable Network Storage. In Proc. of ACM SIGCOMM 2002.
No context found.
M. Beck, T. Moore, and J. S. Plank. An end-to-end approach to globally scalable network storage. In Proc. of ACM SIGCOMM 2002.
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