Optimal MDS codes for cooperative repair.
Two widely studied models of multiple-node repair in distributed storage systems are centralized repair and cooperative repair. The centralized model assumes that all the failed nodes are recreated in one location, while the cooperative one stipulates that the failed nodes may communicate but are distinct, and the amount of data exchanged between them is included in the repair bandwidth. We present two families of $(n,k)$ MDS codes with optimal cooperative repair. Codes in the first family support optimal repair of any two erasures from any $d$ helper nodes for any given $k\le d\le n-2.$ Codes in the second family support optimal repair of any $h$ failed nodes, $2\le h\le n-k-1,$ from any $k+1$ helper nodes.
Publisher URL: http://arxiv.org/abs/1801.09665