Methods and apparatus for compression and network transport of data in support of continuous availability of applications

摘要

Methods and apparatus for compressing data for network transport in support of continuous availability of applications are described. One computer-implemented method of compressing data includes receiving a current instance of data in an input buffer. A candidate chunk of data is selected from the input buffer. A signature hash is computed from a signature length range of data within the candidate chunk. A matching dictionary entry having a matching signature hash from a multi-tiered dictionary is identified. The matching dictionary entry prospectively identifies a location of a prior occurrence of a selected range of consecutive symbols including the signature length range of data within at least one of the current instance of data and a prior instance of data in the input buffer. A dedupe processed representation of the instance of data is formed wherein a dedupe item is substituted for the selected range of consecutive symbols if the selected range is verified as recurring. The dedupe item identifies the location of the prior occurrence of the selected range in accordance with the matching dictionary entry.