Indian Journal of Science and Technology

Abstract

Background/Objectives: One of the most important factors in the design of enterprise computer networks is data storage method. Increase the volume, recovery and security of stored information is the biggest challenge. Using backup systems is a perfect solution to solve these challenges. Traditional methods can be used to perform backups, but new methods and new backup systems are more efficient. Methods/Statistical Analysis: To analyze and compare overall backup session time, backup session processing time, overhead, fragmentation of large data and efficiency of the IP-based schedule and the heuristic-based job scheduling algorithm. Findings: My analysis show that the simple heuristic-based job scheduling algorithm with adaptive number of active DAs is close to optimal while having no additional computing overhead compared to the compute expensive IP-based solution. The simulator-based on heuristic-based job scheduling algorithm enables the analysis to minimize the number of actively used backup servers while meeting backup window constraints and data restore satisfying rates. Applications/Improvements: This algorithm is close to the results that reduce backup time by 60 percent without additional computing overhead and fragmentation of large data processing compared to IPbased methods. As well as, the algorithm can simulate the analysis of a backup system will help. The proposed approach significantly increases the performance, reliability and quality of implemented solutions. 
Keywords: Algorithm, Backup, Performance, Scalability, Schedulin