计算机组织与体系结构课件07.pdfVIP

c0864 Computer Organization and Architecture Lecture 07 RAID Tongwei Ren Software Institute, Nanjing University Mar. 18, 2014 Review 鈥 Error correction 鈥 Parity checking 鈥 Hamming code 鈥 Cyclic redundancy check NANJING UNIVERSITY 2 3 RAID 鈥 Redundant Array of Independent Disks 鈥 Basic idea 鈥 With multiple disks, separate I/O requests, even a single I/O request, can be handled in parallel, as long as the data required reside on separate disks 鈥 Characteristic 鈥 RAID is a set of physical disk drives viewed as a single logical drive 鈥 Data are distributed across the physical drives 鈥 Redundant disk capacity is used to store parity information, which guarantees data recoverability in case of a disk failure NANJING UNIVERSITY 4 RAID Level NANJING UNIVERSITY 5 RAID Level 0 鈥 The data are striped across the available disks 鈥 Not include redundancy to improve performance (not a true member of the RAID family) 鈥 Usage 鈥 High rate data transfer 鈥 High I/O request rate NANJING UNIVERSITY 6 RAID Level 0 (cont.) 鈥 Compared to a single large disk 鈥 If two different I/O requests are pending for two different blocks of data, then there is a good chance that the requested blocks are on different disks NANJING UNIVERSITY 7 RAID Level 1 鈥 Redundancy is achieved by the simple expedient of duplicating all the data 鈥 Data striping is used NANJING