A reliability optimization method for RAID-structured storage systems based on active data migration

Highlights? We allocate reserved space in every disk. ? We arrange to redirect the data in the non-repairable sectors to the reserved space. ? We copy data of the unreliable disk to a new one when disk is in poor reliability. ? Data migration can avoid lengthy data reconstruction. ? It adjust migration speed to reduce the impact on the front end performance. The reliability of RAID system can utilize reconstruction operations to recover data when the disk fails; however, it will result in longer recovery time and negative impact on system performances. Moreover, the possibility of secondary failure will be increased during the recovery period. This paper proposed a new reliability optimization method for RAID system with Active data Migration called RAM, which allocates reserved space in every disk and arranges to redirect the data in the non-repairable sectors to the reserved space called Internal-disk Data Migration. When the number of the bad sectors in the disk exceeds threshold or the disk is in poor reliability, the system will copy data of the unreliable disk to a new one called External-disk Data Migration, this process can avoid lengthy data reconstruction. It can dynamically adjust migration speed to reduce the impact on the front end performance. When the I/O load of system is bursting, the I/O requests from user have high-priority. The overall results indicate that the RAM can improve reliability of the system with little influence.