dcsimg

RAID For the 21st Century

When RAID was first invented the capacity of disk drives was measured in hundreds of megabytes; today, terabyte-sized drives are beginning to appear. Large drives take much longer to rebuild than their smaller predecessors. Application performance degradation that occurs during a RAID rebuild is intolerable for any important operational system. Since the number of disk media failures expected during each read over the surface of a disk grows proportionately with the massive increase in density, media defects in these large drives increase the likelihood of a catastrophic RAID failure and loss of all data. Commercial sites running thousands of concurrent users or processes can avoid the penalties these large disks impose, while taking advantage of their performance, scalability, and availability in an easily managed and cost-efficient fashion.

When RAID was first invented the capacity of disk drives was measured in hundreds of megabytes; today, terabyte-sized drives are beginning to appear. Large drives take much longer to rebuild than their smaller predecessors. Application performance degradation that occurs during a RAID rebuild is intolerable for any important operational system. Since the number of disk media failures expected during each read over the surface of a disk grows proportionately with the massive increase in density, media defects in these large drives increase the likelihood of a catastrophic RAID failure and loss of all data. Commercial sites running thousands of concurrent users or processes can avoid the penalties these large disks impose, while taking advantage of their performance, scalability, and availability in an easily managed and cost-efficient fashion.

Comments are closed.