Title :
Clustered Outband Deduplication on Primary Data
Author :
Agrawal, Archana Satynarayan ; Malhotra, Jyoti
Author_Institution :
Dept. of Inf. Technol., MIT Coll. of Eng., Pune, India
Abstract :
Data reduplication is a special technique to recognize the duplicate data, which stores only one copy for all redundant data, and creates a link to that copy so when data is access by the user, one will have to refer that link. Out-band reduplication is done, after the backup has been written. In today´s era, data reduplication has become the necessity and critical component for the primary storage. In this paper, we have discussed the technique of Out-band data reduplication on primary storage workloads such as data used in common day-to-day life and user directories. This paper also discusses the implementation of out-band reduplication for primary storage in order to increase the RAM performance, throughput, and efficient lookups in the memory. To improve the lookup performance of the system, we have used cuckoo filter instead of multi-layer traditional filters.
Keywords :
storage management; RAM performance; clustered outband deduplication; cuckoo filter; multilayer traditional filters; out-band data reduplication technique; primary data reduplication; primary storage; Fingerprint recognition; Indexes; Matched filters; Random access memory; Servers; Throughput; Backup; cache; chunking; lookups; metadata; post-process; primary data and storage;
Conference_Titel :
Computing Communication Control and Automation (ICCUBEA), 2015 International Conference on
Conference_Location :
Pune
DOI :
10.1109/ICCUBEA.2015.93