Paper Info
Title
Constructions of Fractional Repetition Codes with Flexible Per-Node Storage and Repetition Degree.
Abstract
This paper considers the construction of fractional repetition (FR) codes with flexible storage capacity and repair for distributed storage systems (DSSs). FR codes are the key to constructing a class of distributed storage codes with exact repair by transfer, where, upon failure, a failed storage node is exactly regenerated by simply downloading symbols from the surviving nodes. A major drawback of existing FR codes is that their parameters are not flexible enough to adapt to system changes in DSSs. To address this issue, this paper proposes two constructions of FR codes, called adaptive-and-resolvable FR codes, based on circulant permutation matrices and affine permutation matrices. In the proposed FR codes, the storage capacity per node and repetition degree of the symbols can be varied simultaneously in a simple manner. Some results on the exact file size that can be supported by the proposed FR codes are provided, based on the girth of the corresponding Tanner graph. Furthermore, the proposed FR codes are also shown to meet a Singleton-like bound on the minimum distance for certain parameter ranges.
Year
Venue
Field
2017
IEEE Global Communications Conference
Affine transformation,Computer science,Distributed data store,Permutation matrix,Computer network,Arithmetic,File size,Circulant matrix,Tanner graph,Spread spectrum,Encoding (memory)
DocType
ISSN
Citations 
Conference
2334-0983
0
PageRank 
References 
Authors
0.34
0
1
Name
Order
Citations
PageRank
Yi-Sheng Su1488.42