RESAR: Reliable storage at exabyte scale reconsidered

Abstract

Stored data needs to be protected against device failure and irrecoverable sector read errors, yet doing so at exabyte scale can be challenging because of the large number of failures that must be handled. We have developed and presented RESAR (Robust, Efficient, Scalable, Autonomous, Reliable) storage, an approach to storage system redundancy that only uses XOR-based parity and employs a graph to lay out data and parity [5]. Here we add to our prior work by giving a proof of our analytic results for RESAR reliability. The results presented here are valid for any size RESAR layouts and allow us to make an exact comparison of spiral RESAR robustness with an equivalent RAID 6 disk array.