Modern storage systems such as distributed file systems and key-value stores in many cases exhibit data redundancy. The issue is addressed by deduplication, a process of identifying and eliminating duplicate data. While deduplication is typically applied to data stored on disks, the emergence of RAM-based storage systems opens new problems on one hand while being insensitive to some inherent deficiencies of deduplication such as fragmentation. In this paper we present a review of disk- and memory-based deduplication.
LA - eng M1 - LUSY-2014/01 M3 - Report N2 -
Modern storage systems such as distributed file systems and key-value stores in many cases exhibit data redundancy. The issue is addressed by deduplication, a process of identifying and eliminating duplicate data. While deduplication is typically applied to data stored on disks, the emergence of RAM-based storage systems opens new problems on one hand while being insensitive to some inherent deficiencies of deduplication such as fragmentation. In this paper we present a review of disk- and memory-based deduplication.
PB - University of Ljubljana, Faculty of Computer and Information Science PY - 2014 EP - 10 TI - Efficient Deduplication in Disk- and RAM-based Data Storage Systems UR - http://lusy.fri.uni-lj.si/sites/lusy.fri.uni-lj.si/files/publications/tolic2014-tr01.pdf SN - LUSY-2014/01 ER -