An optimal disk allocation strategy for partial match queries on non-uniform cartesian product files

The disk allocation problem addresses the issue of how to distribute a file on to several disks to maximise the concurrent disk accesses in response to a partial match query. In the past this problem has been studied for binary as well as for p-ary cartesian product files. In this paper, we propose a disk allocation strategy for non-uniform cartesian product files by a coding theoretic approach. Our strictly optimal disk allocation strategies are based on a large and flexible class of maximum distance separable (MDS) codes, namely the redundant residue codes.