Array Erasure Codes with Preset Fault Tolerance Capability

The array erasure code, an ideal method for fault tolerance in storage systems, however, is obstructed by its impossibility to set the fault tolerant ability according to dynamic application environment for practical purpose. In view of this, this paper presents a new class of array erasure codes, with the greatest contribution to the array codes which can be obtained according to the preset number of fault tolerance and storage efficiency in dynamic structure, and fault tolerance capability can be presented is not limited in theory. In addition, the new array code has the advantages of simple structure, easy to realize, with no strong constraints to satisfy in structure. Only binary XOR operations are required for coding and decoding for a high operational efficiency, and fixed update penalty and repair cost which will not increase with the expansion of system size or increase of fault tolerance capability.