1/14
Looks like no tags are added yet.
Name | Mastery | Learn | Test | Matching | Spaced |
---|
No study sessions yet.
What does RAID stand for and what was its original purpose?
Redundant Arrays of Inexpensive Disks - originally showed inexpensive drives could achieve reliability of expensive drives through clever configurations.
What three main techniques are used in RAID systems?
Mirroring (data duplication), striping (data distribution), and error correction (parity/redundancy).
What is RAID 0 and what are its characteristics?
Striped disk array without parity:
Improves performance via parallel reads/writes
Decreases reliability (failure of any drive loses all data)
No fault tolerance
How does RAID 1 (mirroring) work and what are its tradeoffs?
Duplicate sets of drives where all writes go to both:
Doubles read performance
No write performance improvement
Most expensive RAID configuration
Highest reliability among basic RAID levels
Why are RAID 2 and RAID 3 no longer used?
They proved prohibitively expensive, inferior to other techniques, and required synchronized drive spinning.
What is the key limitation of RAID 4?
The dedicated parity drive becomes an I/O bottleneck since every write requires parity recalculation.
How does RAID 5 improve upon RAID 4?
By distributing parity blocks across all drives in round-robin fashion, eliminating the parity drive bottleneck.
What additional protection does RAID 6 provide?
Double parity allowing survival of two simultaneous drive failures (vs. RAID 5's single failure tolerance).
What's the difference between RAID 0+1 and RAID 1+0 (10)?
RAID 0+1: Mirror of stripes (two striped sets mirrored)
RAID 1+0: Stripe of mirrors (multiple mirrored pairs striped)
RAID 10 has better performance during failures.
What happens when a drive fails in RAID 1 vs RAID 5?
RAID 1: Continues with reduced read performance
RAID 5: Severe write performance degradation until rebuilt
What is "hot swapping" and why is it important for RAID?
Ability to replace failed drives without powering down the system, crucial for maintaining continuous operation.
What are "warm standby" drives in RAID systems?
Spare drives already connected but not active, ready to power on and rebuild when a failure occurs.
Why might RAID 10 be preferred over RAID 5 despite higher cost?
Better performance during failures and faster rebuild times (only need to rebuild failed mirror pair).
What special hardware considerations do RAID systems require?
Independent I/O channels/controllers for each drive to ensure true failure independence.
How does RAID improve upon single drive MTBF?
Through redundancy that allows continued operation and data recovery even when individual drives fail.