A Fragment Rewriting Method Used in Data Deduplication System

Abstract

The invention discloses a fragment rewriting method for a data deduplication system, which adds a cache as a rewrite-aware cache in the data deduplication, and a data item in the rewrite-aware cache is a container identifier referenced by a data block (Container ID), for a data block determined to be a data fragment, if the container ID (Container ID) referenced by it has been stored in the rewrite-aware cache, then the data does not need to be rewritten; otherwise, the data fragment is rewritten. The method of the invention adds a cache with the same size and the same cache strategy as the data read in the process of data deduplication, avoids unnecessary rewriting of repeated data blocks, and overcomes the defects of existing algorithms. Compared with the existing de-fragmentation algorithm, one is to improve the impact of data fragmentation on read performance and improve read performance by rewriting data fragmentation; the other is to ensure less deduplication rate while improving read performance loss.

Description

technical field [0001] The invention belongs to the field of computer information storage, and in particular relates to a fragment rewriting method based on a data deduplication technology system. Background technique [0002] Data deduplication technology (data deduplication technology), as a reduction technology that can identify and eliminate redundant data and store only a single copy of data, is widely used in backup storage systems, archive storage systems, and even file systems. For example, using data deduplication technology can eliminate 80% to 90% of redundant data in backup storage systems and archive storage systems, up to 80% of redundant data can be eliminated in virtual machine backup, and 3 / 4 file space overhead and 87% backup image overhead. [0003] However, in the system based on data deduplication, the data blocks of the subsequently stored files share the data blocks of the previously stored files, so that the data blocks are scattered rather than con...

AI Technical Summary

Problems solved by technology

[0006] The object of the present invention is to propose an optimized fragment rewriting method for the above defects or improvement needs of the prior art, which selectively rewrites data blocks determined to be fragments, thereby reducing unnecessary fragment data. Rewriting, to solve the technical problems of deduplication rate reduction and reading performance reduction caused by a large number of fragment rewriting existing in the current data deduplication system, compared with the existing fragment rewriting algorithm, it has higher read performance and Deduplication rate

Images