A new fault tolerance system for cloud storage

We present a class of array codes which are called Random Extensible Codes (RE Code). Benefit from the novel codes, we can encode an array of n disks using t redundant disks to protect the information against a failure of t/2 disks with n and t arbitrary. We can also add several extra disks to an existed disk array to improve its fault tolerant capability. This technique can be applied in a variety of occasions such as large-Scale fault-tolerant network storage system, highly unstable communication and so on, which differs itself from traditional array codes.