l21-raid.pdf

W4118: RAID Instructor: Junfeng Yang References: Modern Operating Systems (3rd edition), Operating Systems Concepts (8th edition), previous W4118, and OS at MIT, Stanford, and UWisc RAID motivation ? Performance ? Disks are slow compared to CPU ? Disk speed improves slowly compared to CPU ? Reliability ? In single disk systems, one disk failure ? data loss ? Cost ? A single fast, reliable disk is expensive 1 RAID idea ? RAID idea: use redundancy to improve performance and reliability ? Redundant array of cheap disks as one storage unit ? Fast: simultaneous read and write disks in the array ? Reliable: use parity to detect and correct errors ? RAID can have different redundancy levels, achieving different performance and reliability ? Seven different RAID levels (0-6) 2 Evaluating RAID ? Cost ? Storage utilization: data capacity / total capacity ? Reliability ? Tolerance of disk failures ? Performance ? (Large) sequential read, write, read-modify-write ? (Small) random read, write, read-modify-write ? Speedup over a single disk 3 Computing cost ? G = number of data disks in a RAID group ? C = number of check disks in a RAID group ? Cost = C/(G+C) 4 Computing reliability ? N = total number of disks ? G = number of data disks in a RAID group ? C = number of check/parity disks in a RAID group ? MTTFdisk = mean time to failure for a disk ? MTTR = mean time to repair for a failed disk ? MTTFraid = ? 5 RAID 0: non-redundant striping ? Structure ? Data striped across all disks in an array ? No parity ? Advantages: ? Good performance: with N disks, roughly N times speedup ? Disadvantages: ? Poor reliability: one disk failure ? data loss 6 RAID 0 performance 7 RAID 1: mirroring ? Structure ? Keep a mirrored (shadow) copy of data ? Advantages ? Good reliability: one disk failure OK ? Good read performance ? Disadvantage ? High cost: one data disk requires one parity disk 8 RAID 1 performance 9 RAID 2: error-correction parity ? Structure ? A data sector striped across data disks ? Compute error-correcting parity and store in parity disks ? Advantages ? Good reliability with higher storage utilization than mirroring ? Disadvantages ? Unnecessary cost: disk can already detect failure ? Poor random performance 10 parity disks RAID 3: bit-interleaved parity ? Structure ? Single parity disk (XOR of each stripe of a data sector) ? Advantages ? Same reliability with one disk failure as RAID2 since disk controller can determine what disk fails ? Higher storage utilization ? Disadvantages ? Poor random performance 11 RAID 4: block-interleaved parity ? Structure ? A set of data sectors (parity group) striped across data disks ? Advantages ? Same reliability as RAID3 ? Good random read performance ? Disadvantages ? Poor random write and read-modify-write performance 12 P RAID 4 performance 13 RAID 5: block-interleaved distributed parity ? Structure ? Parity sectors distributed across all disks ? Advantages ? Good performance 14 P P P P RAID 5 performance 15 RAID6: P+Q redundancy ? Structure ? Same as RAID 5 except using two parity sectors per parity group ? Advantages ? Can tolerate two disk failures 16 P Q P Q P Q Q P RAID levels PowerPoint Presentation