## Network file storage with graceful performance degradation (2005)

### Cached

### Download Links

- [www.paradise.caltech.edu]
- [paradise.caltech.edu]
- [www.paradise.caltech.edu]
- [faculty.cs.tamu.edu]
- DBLP

### Other Repositories/Bibliography

Venue: | ACM Transactions on Storage |

Citations: | 9 - 3 self |

### BibTeX

@ARTICLE{Jiang05networkfile,

author = {Anxiao (andrew Jiang and Jehoshua Bruck},

title = {Network file storage with graceful performance degradation},

journal = {ACM Transactions on Storage},

year = {2005},

volume = {1},

pages = {189}

}

### OpenURL

### Abstract

A file storage scheme is proposed for networks containing heterogeneous clients. In the scheme, the performance measured by file-retrieval delays degrades gracefully under increasingly serious faulty circumstances. The scheme combines coding with storage for better performance. The problem is NP-hard for general networks; and this paper focuses on tree networks with asymmetric edges between adjacent nodes. A polynomial-time memory-allocation algorithm is presented, which determines how much data to store on each node, with the objective of minimizing the total amount of data stored in the network. Then a polynomial-time data-interleaving algorithm is used to determine which data to store on each node for satisfying the quality-of-service requirements in the scheme. By combining the memory-allocation algorithm with the data-interleaving algorithm, an optimal solution to realize the file storage scheme in tree networks is established.

### Citations

10958 |
Computers and Intractability: A Guide to the Theory of NP-completeness
- Garey, Johnson
- 1990
(Show Context)
Citation Context ...are all non-negative integers. nv denotes the number of requirements that v has. ✷ The file storage problem defined above is NP-hard for general graphs, because the NP-complete dominating set problem =-=[3]-=- can be reduced to it. In this paper, we study the case where the graph G = (V, E) is a tree. We assume G has asymmetric edges, which means that for any two adjacent vertices, the two directed edges o... |

648 |
Online computation and competitive analysis
- Borodin, El-Yaniv
- 1998
(Show Context)
Citation Context ... [9], dynamic replication schemes based on estimated temporal data-access statistics (e.g., caching [16]), on-line algorithms that optimize the file-access performance against the worst future events =-=[1]-=-, etc. In those schemes, the file can be seen as encoded with a (1, 0) code, so they are a special case of the more general file-storage model where files are stored in the form of error-correcting co... |

267 |
A case for redundant arrays of inexpensive disks
- Patterson, Gibson, et al.
(Show Context)
Citation Context ...ored distributively, which can be seen as using a (k, 0) code (for some integer k). Error-correcting codes have played a more important role in disk-storage systems and server clusters — such as RAID =-=[14]-=- and DPSS [12] — where files are stored using non-trivial errorcorrecting codes, but there the concept of network is not significant. Works that study the general problem of combining network file sto... |

218 | A Survey of Web Caching Schemes for the Internet
- Wang
- 2002
(Show Context)
Citation Context ...includes median or center type of schemes that minimize the average or maximum file-access cost [4], [9], dynamic replication schemes based on estimated temporal data-access statistics (e.g., caching =-=[16]-=-), on-line algorithms that optimize the file-access performance against the worst future events [1], etc. In those schemes, the file can be seen as encoded with a (1, 0) code, so they are a special ca... |

208 |
An algorithmic approach to network location problems: Part 2. The p-medians
- Kariv, Hakimi
- 1979
(Show Context)
Citation Context ...the value of b to a). We present below a memory-allocation algorithm that uses the technique of searching the tree from its leaves toward its root. Similar techniques have been used in several papers =-=[10]-=-, [15], to solve the domination problem. DEFINITION 2.2. AN OPTIMAL MEMORY BASIS A set {w(v)|v ∈ V } is called an optimal memory basis if there exists an optimal solution to the memory allocation prob... |

189 |
A best possible heuristic for the k-center problem
- Hochbaum, Shmoys
- 1985
(Show Context)
Citation Context ...s the entire file in its original form or none of it, is a topic that has been studied in depth [2]. It includes median or center type of schemes that minimize the average or maximum file-access cost =-=[4]-=-, [9], dynamic replication schemes based on estimated temporal data-access statistics (e.g., caching [16]), on-line algorithms that optimize the file-access performance against the worst future events... |

157 |
Comparative Models of the File Assignment Problem
- Dowdy, Foster
- 1982
(Show Context)
Citation Context ...working at the California Institute of Technology, and by NSF grant CCR-TC-0208975.s2 · ... either stores the entire file in its original form or none of it, is a topic that has been studied in depth =-=[2]-=-. It includes median or center type of schemes that minimize the average or maximum file-access cost [4], [9], dynamic replication schemes based on estimated temporal data-access statistics (e.g., cac... |

36 |
Optimal Placement of Replicas in Trees with Read
- Kalpakis, Dasgupta, et al.
- 2001
(Show Context)
Citation Context ... entire file in its original form or none of it, is a topic that has been studied in depth [2]. It includes median or center type of schemes that minimize the average or maximum file-access cost [4], =-=[9]-=-, dynamic replication schemes based on estimated temporal data-access statistics (e.g., caching [16]), on-line algorithms that optimize the file-access performance against the worst future events [1],... |

35 |
Optimal allocation of resources in distributed information networks
- Mahmoud, Riordan
- 1976
(Show Context)
Citation Context ...ed with a (1, 0) code, so they are a special case of the more general file-storage model where files are stored in the form of error-correcting codes. There also exist schemes using file segmentation =-=[11]-=-, where a file is split into chunks and the chunks are stored distributively, which can be seen as using a (k, 0) code (for some integer k). Error-correcting codes have played a more important role in... |

26 | Optimal file sharing in distributed networks
- Naor, Roth
- 1995
(Show Context)
Citation Context ...orcorrecting codes, but there the concept of network is not significant. Works that study the general problem of combining network file storage with error-correcting codes include the important paper =-=[13]-=- by Naor and Roth — which studies how to store a file using error-correcting codes in a network such that every node can recover the file by accessing only the codeword symbols on itself and its neigh... |

21 |
On k-graceful graphs
- Slater
- 1982
(Show Context)
Citation Context ...lue of b to a). We present below a memory-allocation algorithm that uses the technique of searching the tree from its leaves toward its root. Similar techniques have been used in several papers [10], =-=[15]-=-, to solve the domination problem. DEFINITION 2.2. AN OPTIMAL MEMORY BASIS A set {w(v)|v ∈ V } is called an optimal memory basis if there exists an optimal solution to the memory allocation problem wh... |

11 |
Coding for high availability of a distributed-parallel storage system
- Malluhi, Johnston
- 1998
(Show Context)
Citation Context ...ively, which can be seen as using a (k, 0) code (for some integer k). Error-correcting codes have played a more important role in disk-storage systems and server clusters — such as RAID [14] and DPSS =-=[12]-=- — where files are stored using non-trivial errorcorrecting codes, but there the concept of network is not significant. Works that study the general problem of combining network file storage with erro... |

5 |
Optimal t-Interleaving on Tori
- Jiang, Cook, et al.
- 2006
(Show Context)
Citation Context ...very node can recover the file by accessing only the codeword symbols on itself and its neighbors, with the objective of minimizing the total amount of data stored — and a few other results [5], [7], =-=[8]-=-; however, other than those, research in this field has been very limited. Error-correcting code is a more general way to express a file than the file itself. Therefore, it brings us the flexibility t... |

4 |
Memory Allocation in Information Storage Networks
- Jiang, Bruck
- 2003
(Show Context)
Citation Context ...uch that every node can recover the file by accessing only the codeword symbols on itself and its neighbors, with the objective of minimizing the total amount of data stored — and a few other results =-=[5]-=-, [7], [8]; however, other than those, research in this field has been very limited. Error-correcting code is a more general way to express a file than the file itself. Therefore, it brings us the fle... |

4 | Diversity Coloring for Distributed Data Storage in Networks”, submitted to
- Jiang, Bruck
(Show Context)
Citation Context ... In the data interleaving problem, we use N different colors to represent the N symbols in the codeword, for a more abstract understanding of the problem. 3.2 Data-Interleaving Algorithm In the paper =-=[6]-=-, a solution is presented for coloring the vertices of an undirected tree using N colors, in such a way that for every point of the tree (which can be either a vertex or a point on an edge), there exi... |

3 | Multi-cluster Interleaving on Linear Arrays and Rings
- Jiang, Bruck
(Show Context)
Citation Context ...hat every node can recover the file by accessing only the codeword symbols on itself and its neighbors, with the objective of minimizing the total amount of data stored — and a few other results [5], =-=[7]-=-, [8]; however, other than those, research in this field has been very limited. Error-correcting code is a more general way to express a file than the file itself. Therefore, it brings us the flexibil... |

1 | Multicluster interleaving on linear paths and cycles - JIANG, J |