Transparent, fine-grained and adaptive data compression
AssigneeEMC IP Holding Company LLC
InventorsJunping Zhao, Kenneth J. Taylor, Lin Peng, Kun Wang
Life of this patent
Estimating the term…
Abstract
Transparent, fine-grained, and adaptive data compression is described. A system determines a first data chunk and a second data chunk in a persistent storage. The system determines a first data read count and/or a first data write count for the first data chunk, and a second data read count and/or a second data write count for the second data chunk. The system determines then a first data compression status for the first data chunk and a second data compression status for the second data chunk. Based on the first data compression status and second data compression status, the system stores data in the first data chunk and data in the second data chunk to the persistent storage.
Claims20
A system comprising:
- a processor-based application stored on a non-transitory computer-readable medium, which when executed on a computer, will cause one or more processors to:
- determine a first data chunk and a second data chunk in a persistent storage;
- determine at least one of a first data read count for the first data chunk and a first data write count for the first data chunk, and at least one of a second data read count for the second data chunk and a second data write count for the second data chunk;
- determine a first data compression status for the first data chunk based on at least one of the first data read count and the first data write count;
- determine a second data compression status for the second data chunk based on at least one of the second data read count and the second data write count, the second data compression status differing from the first data compression status;
- store, based on the first data compression status, data in the first data chunk to the persistent storage;
- and
- store, based on the second data compression status, data in the second data chunk to the persistent storage.
A computer-implemented method for transparent, fine-grained, and adaptive data compression, the method comprising:
- determining a first data chunk and a second data chunk in a persistent storage;
- determining at least one of a first data read count for the first data chunk and a first data write count for the first data chunk, and at least one of a second data read count for the second data chunk and a second data write count for the second data chunk;
- determining a first data compression status for the first data chunk based on at least one of the first data read count and the first data write count;
- determining a second data compression status for the second data chunk based on at least one of the second data read count and the second data write count, the second data compression status differing from the first data compression status;
- storing, based on the first data compression status, data in the first data chunk to the persistent storage;
- and
- storing, based on the second data compression status, data in the second data chunk to the persistent storage.
A computer program product, comprising a non-transitory computer-readable medium having a computer-readable program code embodied therein to be executed by one or more processors, the program code including instructions to:
- determine a first data chunk and a second data chunk in a persistent storage;
- determine at least one of a first data read count for the first data chunk and a first data write count for the first data chunk, and at least one of a second data read count for the second data chunk and a second data write count for the second data chunk;
- determine a first data compression status for the first data chunk based on at least one of the first data read count and the first data write count;
- determine a second data compression status for the second data chunk based on at least one of the second data read count and the second data write count, the second data compression status differing from the first data compression status;
- store, based on the first data compression status, data in the first data chunk to the persistent storage;
- and
- store, based on the second data compression status, data in the second data chunk to the persistent storage.
Dependencies are inferred from the claim text.