Storage system capable of managing a plurality of snapshot families and method of snapshot family based read

摘要

A method for a snapshot family based reading of data units from a storage system, the method comprises: receiving a read request for reading a requested data entity, searching in a cache memory of the storage system for a matching cached data entity, if not finding the matching cached data entity then: searching for one or more relevant data entity candidates stored in the storage system; selecting, out of the one or more relevant data entity candidates, a selected relevant data entity that has a content that has a highest probability, out of contents of the one or more relevant data entity candidates, to be equal to the content of the requested data entity; and responding to the read request by sending the selected relevant data entity.

主权项

1. A method for a snapshot family based reading of data units from a storage system, the method comprises:
receiving a read request for reading a requested data entity, the requested data entity is associated with a certain snapshot of a certain snapshot family and with a certain logical address; searching in a cache memory of the storage system for a matching cached data entity, wherein a matching cached data entity is associated with the certain snapshot of the certain snapshot family and with the certain logical address; responding to the read request by sending the matching cached data entity if finding the matching cached data entity; if not finding the matching cached data entity then:
searching for one or more relevant data entity candidates stored in the storage system;wherein each relevant data entity candidate of the one or more relevant data entity candidates is (a) associated with the certain logical address, (b) associated with the certain snapshot family, and (c) has a content that has a non-zero probability to equal a content of the requested data entity;selecting, out of the one or more relevant data entity candidates, a selected relevant data entity that has a content that has a highest probability, out of contents of the one or more relevant data entity candidates, to be equal to the content of the requested data entity; andresponding to the read request by sending the selected relevant data entity.