Functional-Repair-by-Transfer.PDFVIP

Functional-Repair-by-Transfer Regenerating Codes Kenneth W. Shum and Yuchong Hu Abstract— In a distributed storage system, a data ?le is 1st packet 2nd packet distributed to several storage nodes, such that the original ?le Node 1 A C + D can be decoded from any subset of the storage nodes of size Node 2 B D + A larger than or equal to a certain threshold. Upon the failure of a storage node, we would like to regenerate it with minimal amount Node 3 C A + B of data transmissions from the surviving nodes to the new node. Node 4 D B + C This performance metric is called the repair-bandwidth. Another performance metric is the disk input/output (I/O) cost, which Fig. 1. An example of exact regenerating codes with repair-by-transfer. measures the number of bits a storage node needs to read out from its memory in order to repair the failed node. In this paper, we give examples of linear regenerating codes with minimal disk I/O cost and repair-bandwidth, without any linear mixing in the packet in each storage node is an uncoded packet, the second F helping storage nodes. one is the sum of two packets, with addition performed in 2, Index Terms—Cloud storage, distributed storage system, re- the ?nite ?eld of size 2. Let x ⊕ y denote the unique integer generating codes, network coding. z ∈ {1, 2, 3, 4} such that x + y ≡ i mod 4. The parity-check packets in nodes i ⊕ 2 is the sum of the uncoded packets in I. INTRODUCTION node i and node i⊕1. One can verify that the four information Regenerating codes, as introduced by Dimakis et al. in [2], packets can be decoded from any two nodes. For example, is a class of er