Data is stored in a distributed data storage system comprising a plurality ofdisks. When a disk fails, system reliabilityis restored by executing a set of reconstructions according to a schedule.System reliability is characterized by a dynamicNormalcy Deviation Score. The schedule for executing the set ofreconstructions is determined by a minimum intersection policy.A set of reconstructions is received and divided into a set of queues rank-ordered by redundancy level ranging from a lowestredundancy level to a highest redundancy level. For reconstructions in eachqueue, an intersection matrix is calculated. Diskscoresfor each disk are calculated. The schedule for the set of reconstructions isbased at least in part on the intersection matrices, theNormal Deviation Scores, and the diskscores.
展开▼
