抽象的

Recovery of Disk Failure in RAID-5 Using Disk Replacement Algorithm

M.P.Ramkumar Dr.N.Balaji , G.Rajeswari

In (Redundant Arrays of Independent Disk) RAID, failure of single disk is tolerated up to Level 5. The performance of disk drives may degrade due to I/O requests directed towards failed disk. RAID controller supports data reconstruction in the event of a disk failure, to recover the data from failed disk. In degraded mode, the recovery of data from failed disk would cause additional workflow in all operational disks. Another mode of recovery called Hot-spare, where disks that are in the end of their lifetime are replaced with spare disks without requirement for additional I/O operations and parity calculations. In this work an existing disk replacement algorithm known as M-SSTF (Modified SSTF) is used for early recovery of failed disk. Early recovery of single disk failure might help tolerating another subsequent disk failure. Hence it is important to rebuild a failed disk as early as possible.