Fractional repetition (FR) codes are a family of repair-efficient storagecodes that provide exact and uncoded node repair at the minimum bandwidthregenerating point. The advantageous repair properties are achieved by atailor-made two-layer encoding scheme which concatenates an outermaximum-distance-separable (MDS) code and an inner repetition code. In thispaper, we generalize the application of FR codes and propose heterogeneousfractional repetition (HFR) code, which is adaptable to the scenario where therepetition degrees of coded packets are different. We provide explicit codeconstructions by utilizing group divisible designs, which allow the design ofHFR codes over a large range of parameters. The constructed codes achieve thesystem storage capacity under random access repair and have multiple repairalternatives for node failures. Further, we take advantage of the systematicfeature of MDS codes and present a novel design framework of HFR codes, inwhich storage nodes can be wisely partitioned into clusters such that datareconstruction time can be reduced when contacting nodes in the same cluster.
展开▼