IBM SG24-4576-00 - Manual (Page 45)

1.6.6.8 Summary of RAID Performance Characteristics
RAID-0: Block Interleave Data Striping without parity
• Fastest data-rate performance
• Allows seek and drive latency to be performed in parallel
• Significantly outperforms single large disk
RAID-1: Disk Mirroring/Disk Duplexing and Data Strip mirroring (RAID-1,
Enhanced)
• Fast and reliable, but requires 100% disk space overhead
• Data copied to each set of drives
• No performance degradation with a single disk failure
• RAID-1 enhanced provides mirroring with an odd number of drives
RAID-2: Bit Interleave Data Striping with Hamming Code
• Very fast for sequential applications, such as graphics modelling
• Almost never used with PC-based systems
RAID-3: Bit Interleave Data Striping with Parity
• Access to all drives to retrieve one record
• Best for large sequential reads
• Very poor for random transactions
• Poor for any write operations
• Faster than a single drive, but much slower than RAID-0 or RAID-1 in random
environments
RAID-4: Block Interleave Data Striping with one Parity Disk
• Best for large sequential I/O
• Very poor write performance
• Faster than a single drive, but usually much slower than RAID-0 or RAID-1
RAID-5: Block Interleave Data Striping with Skewed Parity
• Best for random transactions
• Poor for large sequential reads if request is larger than block size
• Better write performance than RAID-3 and RAID-4
• Block size is key to performance, must be larger than typical request size
• Performance degrades in recovery mode (when a single drive has failed)
Table 7. Summary of RAID Performance Characteristics
RAID Level Capacity Large Transfers High I/O Rate Data Availability
Single Disk Fixed (100%) Good Good 1
RAID-0 Excellent Very Good Very Good Poor 2
RAID-1 M o d e r a t e ( 5 0 % ) Good Good Good
RAID-2 Very Good Good Poor Good
RAID-3 Very Good Very Good Poor Good
RAID-4 Very Good Very Good Poor Good
RAID-5 Very Good Very Good Good Good
Note:
1 The MTBF (mean time before failure) for single disks can range from 10,000 to 1,000,000 hours.
2 Availability = MTBF of one disk divided by the number of disks in the array.
30
NetWare Integration Guide