Blocklet pattern identification
View Patent ↗Apparatus, methods, and other embodiments associated with blocklet pattern identification are described. One example method includes accessing a blocklet produced by a computerized data de-duplication parsing process before providing the blocklet to a duplicate blocklet determiner. The example method also includes hashing a portion of the blocklet to produce a pattern indicating hash and then identifying the blocklet as a pattern blocklet if the pattern indicating hash matches a pre-determined pattern indicating hash. To improve efficiency in a data de-duplication process, the blocklet pattern identifying may be performed independently from a data structure and process used by the duplicate blocklet determiner. If the blocklet is a pattern blocklet, then the method includes selectively controlling the duplicate blocklet determiner to not process the pattern blocklet. The duplicate determination is not needed because a pattern determination has already been made.
1. An apparatus, comprising:
a processor;
a memory;
a set of logics; and
an interface to connect the processor, the memory, and the set of logics, the set of logics comprising:
a first logic configured to select a blocklet to be examined for the presence of a pattern capable of being compressed, where the blocklet was produced by a data de-duplication parser;
a second logic configured to determine whether the blocklet is a pattern blocklet, and
a third logic configured to determine an entropy of the blocklet, and where the first logic selects the blocklet to be examined for the presence of a pattern capable of being compressed upon determining that the entropy of the blocklet falls within a pre-determined range.
2. The apparatus of claim 1 , where the first logic selects the blocklet to be examined for the presence of a pattern capable of being compressed upon determining that the blocklet is a maximum size blocklet produced by the data de-duplication parser.
3. The apparatus of claim 1 , where the second logic is configured to determine whether the blocklet is a pattern blocklet using a fingerprint hash associated with a duplicate blocklet determination.
4. The apparatus of claim 1 , where the second logic is configured to determine whether the blocklet is a pattern blocklet using a pattern hash different from a fingerprint hash associated with a duplicate blocklet determination.
5. The apparatus of claim 1 , where the memory is configured to store one or more pre-determined hashes associated with patterns to be recognized, where the one or more pre-determined hashes are configured to identify a contiguous run of single repeating values or a contiguous run of two or more repeating values, and where the second logic is configured to determine whether the blocklet is a pattern blocklet by comparing a hash of the blocklet to a hash stored in the memory.
6. The apparatus of claim 1 , where the third logic is configured to control, as a function of the entropy, a blocklet matching process or an index lookup process.