Optimizing RAID for long term data archives

We present new methods to extend data reliability of disks in RAID systems for applications like long term data archival. The proposed solutions extend existing algorithms to detect and correct errors in RAID systems by preventing accumulation of undetected errors in rarely accessed disk segments. Furthermore we show how to change the parity layout of a RAID system in order to improve the performance and reliability in case of partially defect disks. All methods benefit of a hierarchical monitoring scheme that stores reliability related information. Our proposal focuses on methods that do not need significant hardware changes.