220 likes | 477 Views
RAID Technology. CS350 Computer Organization Section 2 Larkin Young Rob Deaderick Amos Painter Josh Ellis. Overview. Background Term coined by researchers at UC-Berkeley R edundant A rray of I nexpensive D isks Basics of RAID Disk arrays Logical/physical
E N D
RAID Technology CS350 Computer Organization Section 2 Larkin Young Rob Deaderick Amos Painter Josh Ellis
Overview • Background • Term coined by researchers at UC-Berkeley • Redundant Array of Inexpensive Disks • Basics of RAID • Disk arrays • Logical/physical • Controller (software/hardware)
Techniques/Methods • Mirroring • Parity 10101010 XOR 11111111 = 01010101 11111111 XOR 01010101 = 10101010 10101010 XOR 01010101 = 11111111
Techniques/Methods (cont’d) • Striping
RAID 0 • Uses striping • I/O performance gain • No Data redundancy • Not fault tolerant • Not considered “true” RAID
RAID 1 • Uses mirroring • Also known as duplexing • Fault tolerant • High Disk overhead • Mirroring typically handled system software • Simplest RAID design
RAID 2 • Error Detection and Correction • Parity and Hamming Code • Used to identify and correct errors • Hamming Code • Uses parity bits to verify data integrity • Parity bits signify oddness of data bits • ECC (Error Correction Code)
RAID 2 Characterized by: • “On the fly” data correction • High Ratio of Data disks to ECC disks • Not a commercially viable solution • High Costs Involved
RAID 3: XOR • Exit Hamming Code, Enter XOR (eXclusive OR) • XOR in Action: 01012 XOR 00112 = 01102 If ????2 XOR 11002 = 01102, Then 11002 XOR 01102 = ????2 = 10102 Thus we can use XOR results to recover lost data
Advantages: High Read/Write Transfer Rates Disk failures don’t slow the system Low Ratio of Data Disks to Parity Disks Disadvantages: Transaction rate slowed by Parity Disk Complex Controller Design Software Implementation Resource Intensive RAID 3: Pros and Cons
RAID 3: Uses • Video Production • High-end Video and Image Editing • Other uses that require high throughput of data
RAID 4 • Offshoot of RAID 3 • No Striping, Entire files written to individual disks • Parity Data created on a sector-by-sector basis
Advantages: Very high read rates Multiple files read at once Uses: Web Servers, and other high read, low write situations Disadvantages: Very slow write rates Even small writes fill up parity write queue Inefficient data recovery Even more Complex Controller Design than RAID 3 RAID 4: Pros, Cons, and Uses *Has most of the other Advantages and Disadvantages of RAID 3
RAID 5 • Highest Read data transaction rate • Medium Write data transaction rate • Most complex controller design • Used For Server Applications.
RAID 6 • Fault tolerance • Very complex controller design • Very poor write performance
RAID 7 • Extremely high cost per MB • Fast access times • Improved write performance with increasing number of drives
RAID 10 • Uses multiple (mirrored) RAID 1 in a single array • Data striped across all mirrored sets • Very high fault tolerance • High performance rate
RAID 10 Characterized by: - each drive duplicated - high implementation cost
RAID 53 • Consists of a striped array made up of RAID 3 segments • Very expensive • Achieves high rate of I/O
RAID 0+1 • Acts as a mirrored array whose parts consist of RAID 0 arrays • Very expensive • Works well in handling images • very high data transfer rate
RAID 0+1 - Not great reliability - one disk failure leaves you left with only the characteristics of a striped array (no fault tolerance)
Conclusion • Grown more complex • Large price range • Huge step forward for industry • Not for the masses yet