Logical Data Deletion in High-Performance De-Duplication Backup
Tian Ming Yang, Zhen Tang, Ya Ping Wan, Wei Sun · Applied Mechanics and Materials · 2013
Data de-duplication divides backup stream into chunks and eliminates duplicate chunks across the entire system, thus remarkably reduces the storage and bandwidth requirement for backups. However, this technique also introduced many new problems among which the performance problem has been resolved by many of the existing solutions, while the logical data deletion problem is not well studied till now. This paper studied the logical data deletion mechanism in de-duplication backup systems, analyzed the memory overhead of the Bloom filter, which supports both high performance de-duplication and logical data deletion, and proposed a lazy deletion method to minimize the influence of logical data deletion on de-duplication performance.