Abstract: Disclosed herein is a mechanism for use in a data processing system for recovering from errors detected when reading data from an array. At least two arrays, each of which may be a distinct portion of a single physical array, contain identical data. When data is written, it is written into both arrays. When data is read, it is read from one of the arrays. If an error is detected on readout, there will be a system retry and the other array will be accessed at the next read request. So long as no errors are detected, each successive read will be from the same array. An error detected on readout will cause the next read operation to access the other array.

Abstract: The disclosure enables concurrent access to a cache by main storage and a processor by means of a cache control which provides two cache access timing cycles during each processor storage request cycle. The cache is accessible to the processor during one of the cache timing cycles and is accessible to main storage during the other cache timing cycle. No alternately accessible modules, buffering, delay, or interruption is provided for main storage line transfers to the cache.