Run-time Performance Optimization and Job Management in a Data Protection Solution
Cherkasova, Ludmila; Lau, Roger; Burose, Harald; Kalambur, Subramaniam Venkata; Kappler, Bernhard; Veeranan, Kuttiraja
Keyword(s): data Protector, backup management. job scheduling, performance evaluation, automated parameter tuning
Abstract: The amount of stored data in enterprise Data Centers quadruples every 18 months. This trend presents a serious challenge for backup management: one either needs to continuously scale the backup infrastructure or to significantly improve the performance and efficiency of existing backup tools. In this work, we discuss potential performance shortcomings of the traditional backup solutions. We analyze historic data on backup processing from eight backup servers in HP Labs, and introduce two additional metrics associated with each backup job, called job duration and job throughput. Our goal is to design a backup schedule that minimizes the overall completion time for a given set of backup jobs. This problem can be formulated as a resource constrained scheduling problem which is known to be NP-complete. As an efficient heuristic for the classic optimization problem, we propose a novel job scheduling algorithm, called FlexLBF. The scheduler utilizes extracted information from historic data and provides a significant reduction in the backup time (up to 50%), improved quality of service, and reduced resource usage (up to 2-3 times). Moreover, the proposed framework automates parameter tuning to avoid manual configuration by system administrators while helping them to achieve nearly optimal performance.
External Posting Date: May 4, 2011 [Fulltext]. Approved for External Publication
Internal Posting Date: July 6, 2010 [Fulltext]