Heat indices for file systems and block storage
AssigneeDell Products L.P.
InventorsAbhijit Dinkar, Vinod Jayaraman, Murali Bashyam, Goutham Rao
Life of this patent
Estimating the term…
Abstract
Techniques and mechanisms are provided to allow for selective optimization, including deduplication and/or compression, of portions of files and data blocks. Data access is monitored to generate a heat index for identifying sections of files and volumes that are frequently and infrequently accessed. These frequently used portions may be left non-optimized to reduce or eliminate optimization I/O overhead. Infrequently accessed portions can be more aggressively optimized.
Claims20
A method, comprising:
- determining whether a first portion of a file is less frequently accessed than a second portion of the file;
- based on a determination that the first portion of the file is less frequently accessed than the second portion of the file, performing optimization using a higher reduction ratio on the first portion of the file than on the second portion of the file.
An apparatus, comprising:
- a processor configured determine whether a first portion of a file is less frequently accessed than a second portion of the file;
- wherein the processor is further configured to optimize the file;
- wherein optimization is performed using a higher reduction ratio on the first portion of the file than on the second portion of the file based on a determination that the first portion of the file is less frequently accessed than the second portion of the file.
A non-transitory computer readable medium, comprising:
- computer code for determining whether a first portion of a file is less frequently accessed than a second portion of the file;
- and
- computer code for optimizing the file;
- wherein optimization is performed using a higher reduction ratio on the first portion of the file than on the second portion of the file based on a determination that the first portion of the file is less frequently accessed than the second portion of the file.
Dependencies are inferred from the claim text.