Frequency and migration based re-parsing
View Patent ↗Example apparatus and methods associated with frequency and migration based re-parsing are provided. One example data de-duplication apparatus includes a migration logic and a parsing logic. The migration logic may be configured to perform a data transfer according to an access frequency to the data. The parsing logic may be configured to re-parse the data based on the access frequency to the data. In different examples, parsing the data may be performed in response to migrating the data. In one example, parsing the data may be performed during or after the migration. Additional examples illustrate parsing the data to balance performance against reduction in light of access frequency to the data block.
1. A method, comprising:
accessing frequency data for a data block in a de-duplication system, where the frequency data indicates at least an access frequency to the data block; and
selectively performing, as a function of the frequency data, one or more of, migrating the data block from a first storage location to a second storage location, and parsing the data block,
where the data block was previously de-duplicated and the parsing is a re-parsing of the data block, or
where parsing the data block includes selecting a parsing algorithm and applying the selected parsing algorithm to the data block based, at least in part, on the frequency data.
2. The method of claim 1 , where migrating the data block is performed when the access frequency falls below a predetermined threshold.
3. The method of claim 2 , where the first storage location is a hot storage, and the predetermined threshold is a minimum access frequency to maintain the data block in the first storage location.
4. The method of claim 1 , where migrating the data block is performed when the access frequency exceeds a predetermined threshold for the first storage location.
5. The method of claim 4 , where the first storage location is a cold storage and the predetermined threshold is a maximum access frequency to maintain the data block in the first storage location.
6. The method of claim 1 , where parsing the data block includes one of, parsing the data block while the data block is being migrated from the first storage location to the second storage location, and parsing the data block after the data block has been migrated to the second storage location.
7. The method of claim 6 , where the parsing algorithm is selected, at least in part, to reconcile the parsing algorithm with a second algorithm used to parse data blocks associated with the second storage location.
8. The method of claim 6 , where the parsing algorithm is selected based, at least in part, on a type of the second storage location.
9. The method of claim 8 , where parsing the data block includes using a parsing algorithm to improve reduction in data when the second storage location is a cold storage and using a parsing algorithm to enhance performance when the second storage location is a hot storage.
10. The method of claim 1 , where the parsing algorithm is configured to enhance performance when the frequency data indicates a high access frequency and the parsing algorithm is configured to enhance reduction in data when the frequency data indicates a low access frequency.
11. A non-transitory computer-readable medium storing computer executable instructions that when executed by a computer cause the computer to perform a method, the method comprising:
accessing frequency data for a data block in a de-duplication system, where the frequency data indicates at least an access frequency to the data block; and
selectively performing, as a function of the frequency data, one or more of, migrating the data block from a first storage location to a second storage location, and parsing the data block,
where migrating the data block is performed in response to determining that the frequency data indicates a frequent source of access to the data block is from a remote source associated with the second storage location, and
where migrating the data block includes copying the data block to the second storage location at the remote source, and where the frequent source of access is a remote peer group associated with the remote source that accesses the data block more frequently than another peer group.
12. A non-transitory computer-readable medium storing computer executable instructions that when executed by a computer cause the computer to perform a method, the method comprising:
parsing a data block based, at least in part, on an access frequency to the data block, where parsing the data block based, at least in part, on the access frequency includes selecting one of, a parsing algorithm to enhance reduction when the second data storage location is a cold storage, and a parsing algorithm to enhance performance when the second data storage location is a hot storage, where the hot storage stores data that is frequently accessed and the cold storage stores data that is infrequently accessed.
13. The non-transitory computer-readable medium of claim 12 , where the parsing is triggered by a migrating.
14. A data de-duplication apparatus, comprising:
a processor;
a memory;
a set of logics; and
an interface configured to connect the processor, the memory, and the set of logics, the set of logics comprising:
a migration logic configured to transfer a data block from a first data store to a second data store based, at least in part, on an access frequency to the data block, and where the migration logic is configured to select the second data store as a function of the access frequency, and where the second data store is selected from a plurality of data stores with different latency attributes; and
a parsing logic configured to re-parse the data block based, at least in part, on the access frequency to the data block.
15. The data de-duplication apparatus of claim 14 , where the parsing logic is further configured to reparse the data block when the data block is being transferred or after the data block has been transferred based, at least in part, on one or more of, a property of the second data store, and the access frequency.