An Efficient Data Deduplication based on Tar-format Awareness in Backup Applications

* student authors Disk-based backup storage system is utilized widely, and data deduplication is becoming an essential tech-nique in the system because of the advantage of a space-efficiency. Usually, user’s several files are aggregated into a single Tar file at primary storage, and the Tar file is transferred and stored to the backup storage system periodically (e.g., a weekly full backup) [1]. In this pa-per, we present a boundary-aware chunking (BAC) technique in a data deduplication backup storage system to improve space-efficiency by using a hint of the stor-ing Tar file. Chunking technique has the role of making chunks. Because chunks are basic granularity for detecting an