Method and device for caching data

Abstract

The invention provides a data caching method and device. The method comprises the steps that S1, an IO data block issued from an upper layer is received and split into a plurality of small data blocks; S2, the small data blocks obtained after splitting are made to sequentially access a cache; S3, after access is completed, the small data blocks obtained after splitting are encapsulated into the IOdata block, and the IO data block is submitted to an upper-layer interface through a callback function. The device comprises a data splitting module, a data access module and a data encapsulation module, wherein the data splitting module is used for splitting the IO data block into the small data blocks; the data access module is used for realizing reading or writing access to the cache by the small data blocks; and the data encapsulation module is used for encapsulating the small data blocks into the complete IO data block. By means of splitting the IO data block issued from the upper layerinto smaller data blocks first and then making the small data blocks access the cache or a disk, the cache hit rate of an IO request is increased, and cache access speed is increased.

Description

technical field [0001] The present invention relates to the field of computer technology, in particular to a method and device for caching data. Background technique [0002] In computer technology, cache is widely used, between CPU and memory, and in storage software. Data cache is an important determinant of the device's operating capability. It not only directly reflects the performance of the device, but also determines the range of hardware options. [0003] At present, several common data scheduling methods for caching software in the Linux system basically do not do too much processing on the IO data blocks delivered by the upper layer, and directly use the received IO data blocks to access the cache disk or the backend. The end mechanical disk is only optimized in terms of data access priority and access channels. For larger IO data blocks, the method of accessing the IO data block as a whole has a higher impact on the cache hit rate and cache access speed. Can not...

AI Technical Summary

Problems solved by technology

[0003] At present, several common data scheduling methods for caching software in the Linux system basically do not do too much processing on the IO data blocks delivered by the upper layer, and directly use the received IO data blocks to access the cache disk or the backend. The end mechanical disk is only optimized in terms of data access priority and access channels. For larger IO data blocks, the method of accessing the IO data block as a whole has a higher impact on the cache hit rate and cache access speed. Can not meet customer needs

Images