Two-Level Metadata Management for Data Deduplication System
Data deduplication is an essential solution to reduce storage space requirement. Especially chunking based data deduplication is very effective for backup workloads which tend to be files that evolve slowly, mainly through small changes and additions. In this paper, the authors introduce a novel data deduplication scheme which can be efficiently used with low bandwidth network in a rapid time. The key points are using tree map searching and classifying data as global and metadata. These are the main aspects to influencing fast performance of the data deduplication.