Nagendra Srivastava
- 1. RAID Technology By Nagendra Srivastava
- 2. CONTENT Introduction Techniques/methods Mirroring Striping Parity Advantages and Disadvantages Uses Conclusion
- 3. What is RAID History > It first defined by David A. Patterson, Garth A. Gibson, and Randy Katz at the University of California, Berkeley in 1987 to describe a redundant array of inexpensive disks. > What is RAID - It is a technology that allowed computer users to achieve high levels of storage reliability from low-cost and RAID allows you to store the same data redundantly (in multiple paces) in a balanced any to improve overall performance.
- 4. Benefits of RAID > Data loss can be very dangerous for an organization > RAID technology prevents data loss due to disk failure > RAID technology can be implemented in hardware or software > Servers make use of RAID technology
- 5. Hardware vs. Software RAID > Software RAID • Software RAID: run on the server’s CPU • Directly dependent on server CPU performance and load • Occupies host system memory and CPU operation, degrading server performance > Hardware RAID • Hardware RAID: run on the RAID controller’s CPU • Does not occupy any host system memory. Is not operating system dependent • Host CPU can execute applications while the array adapter's processor simultaneously executes array functions: true hardware multi-tasking
- 6. RAID • Redundant Arrays of Independent Drives • Benefits – Improved data availability – Improved I/O performance – Increased scalability • Levels supported by Array controllers – RAID 0 – Data striping – RAID 1 – Drive mirroring – RAID 4 – Data guarding – RAID 5 – Distributed data guarding – RAID 6/ADG - Advanced data guarding – RAID 0+ 1 – Mirroring and striping
- 7. Single Drive and Drive Arrays Single Drive Drive Arrays
- 8. Drive Array Features • Data striping across multiple drives • Multiple channels • Request processing Single Fixed Disk Drive Drive 1 Array 3 2 4 1 2 3 4 1234 Drive Array Organization
- 9. Techniques/Methods Mirroring Parity 10101010 XOR 11111111 = 01010101 11111111 XOR 01010101 = 10101010 10101010 XOR 01010101 = 11111111
- 11. RAID Level 0 — Data Striping File divided into chunks (or segments) and then written (striped) across multiple drives
- 12. RAID 0 Uses striping I/O performance gain No Data redundancy Not fault tolerant Not considered “true” RAID
- 13. Striping Factor 64KB Host Data Fixed Striping Factor Based on Drive Sizes
- 14. RAID Level 1 — Drive Mirroring Data Written to Two or More Separate Mirrored Drives
- 15. RAID 1 Uses mirroring Also known as duplexing Fault tolerant High Disk overhead Mirroring typically handled system software Simplest RAID design
- 16. RAID Level 0+1 Disk 0 is mirrored to disk 2 and disk 1 is mirrored to disk 3. Then disk 0 is striped with disk 1 and disk 2 is striped to disk 3.
- 17. RAID 0+1
- 18. RAID 0+1 RAID 1+0 requires an array with four or more physical disks. The disks are mirrored in pairs and data blocks are striped across the mirrored pairs Advantages Highest read and write performance No loss of data as long as no failed disks are mirrored to any other failed disk Disadvantages Expensive and Low disk capacity
- 19. RAID Level 4 — Data Guarding Data striped across multiple drives and then its parity sum is calculated and written to a dedicated parity drive
- 20. RAID 4 Advantages: Disadvantages: Very high read rates Very slow write rates Even small writes fill up Multiple files read at parity write queue once Inefficient data recovery Uses: Even more Complex Web Servers, and Controller Design than other high read, low RAID 3 write situations
- 21. RAID Level 5 — Distributed Data Guarding Data is striped across multiple drives and then its parity sum calculated and striped across multiple drives. Example of 64KB striped across five drives using 4KB chunks.
- 22. RAID 5 RAID 5 uses a parity data formula to create fault tolerance. In RAID 5 each block of data stripe contains parity data that is calculated for the other data blocks in that strip. The blocks of parity data are distributed over the physical disks that make up the logical drive with each physical disk containing only one block of parity data It is referred to as data guarding. Advantages High read performance No loss of data if one physical disk fails More usable disk capacity Disadvantages Relatively low write performance Data loss occurs if a second disk fails before data from the first failed disk is rebuilt
- 23. RAID ADG (Advanced Data Guarding) > RAID ADG is similar to RAID 5 except this RAID level writes 2 sets of parity stripped across all drives. > Protects against failure of ANY 2 drives in the array A B C P Q D E P Q F G P Q H I P Q J K L RAID ADG
- 24. RAID ADG (Advanced Data Guarding) > P = f1(A, B, C) = RAID 5 Parity > Q = f2(A, B, C) = new ADG Parity A B C P Q D E P Q F G P Q H I P Q J K L RAID ADG
- 25. RAID ADG (Advanced Data Guarding) > If 2 parity drives are selected, the system can sustain failure of ANY 2 drives. X X A B C P Q D E P Q F G P Q H I P Q J K L RAID ADG
- 26. RAID ADG (Advanced Data Guarding) or RAID 6 RAID advanced data guarding sometimes referred to as RAID 6, is similar to RAID 5 in that parity data is generated and stored to protect against data loss caused by physical disk failure Advantages High read performance High data availability More usable disk capacity
- 27. RAID 10 Combining RAID 0 and RAID 1 is often referred to as RAID 10 which offers higher performance than RAID 1 but at much higher cost Uses multiple (mirrored) RAID 1 in a single array Data striped across all mirrored sets Very high fault tolerance High performance rate
- 28. RAID 10 Characterized by: Each drive duplicated High implementation cost
- 29. Comparing RAID Levels RAID 0 RAID 1 RAID 5 RAID 10 Read High 2X High High Write High 1X Medium High Fault No Yes Yes Yes tolerance Disk High Low High Low utilization Key Data lost Use double the Lower throughput Very expensive, not problems when any disk disk space with disk failure scalable fails Key High I/O Very high I/O A good overall High reliability with advantages performance performance balance good performance
- 30. On-Line Spare • Replacement for failed drive • Requires hardware fault tolerance • Background rebuild process • Four On-Line Spares maximum (Smart Array controller) Mirrored Mirrored Pair Pair Mirrored Pair On-Line On-Line Spare Spare Before During After Failure Failure Replacement
- 31. Conclusion So what have we learned here? Well we have learned that RAID is not just a bug spray. RAID is a good solution for companies or individuals carving more transfer performance, redundancy and storage capacity in their data storage systems.
- 32. Thanks