Section 1 : Storage System

Data Management:
RAID
Why RAID

• Performance limitation of disk drive

• An individual drive has a certain life expectancy
• Measured in MTBF
• Example
• If the MTBF of a drive is 750,000 hours, and there are 100 drives in the array, then the
MTBF of the array becomes 750,000 / 100, or 7,500 hours
• RAID was introduced to mitigate this problem
• RAID provides:
• Increase capacity
• Higher availability
• Increased performance
RAID Array Components
Physical
Array

Logical
Array

RAID
Controller
Hard Disks

Host

RAID Array
RAID Implementations

• Hardware (usually a specialized disk controller card)

• Controls all drives attached to it
• Array(s) appear to host operating system as a regular disk drive
• Provided with administrative software
• Software
• Runs as part of the operating system
• Performance is dependent on CPU workload
• Does not support all RAID levels
RAID Levels

• 0 Striped array with no fault tolerance

• 1 Disk mirroring
• Nested RAID (i.e., 1 + 0, 0 + 1, etc.)
• 3 Parallel access array with dedicated parity disk
• 4 Striped array with independent disks and a dedicated parity disk
• 5 Striped array with independent disks and distributed parity
• 6 Striped array with independent disks and dual distributed parity
 Data Organization: Striping Stripe

Strip

Stripe

Strip 1 Strip 2 Strip 3

Stripe 1
Stripe 2

Strips
RAID 0
0

1
5
9

RAID 2
Controller 6
10

3
Host 7
11
RAID 1

RAID
Block 0
1 Block 0
1
Controller

Host
Nested RAID – 0+1 (Striping and
Mirroring)
RAID 1

Block 0

Block 2

RAID RAID 0
Block 0
3
2
1
Controller

Block 1

Host Block 3
Nested RAID – 0+1 (Striping and
Mirroring)
RAID 1

Block 0 Block 0

Block 2 Block 2

RAID RAID 0
Controller

Block 1 Block 1

Host Block 3 Block 3

Nested RAID – 1+0 (Mirroring and
Striping)
RAID 0

Block 1

Block 3

RAID RAID 1
Block 2
0
Controller

Block 1

Host Block 3
Nested RAID – 1+0 (Mirroring and
Striping)
RAID 0

Block 0 Block 1

Block 2 Block 3

RAID RAID 1
Controller

Block 0 Block 1

Host Block 2 Block 3

RAID Redundancy: Parity
0 4

1
6 5
9

RAID 1
?
Controller

3
Host 7 7
11
The middle drive fails:
Parity calculation 4 + 6 + 1 + 7 = 18 0123
4 + 6 + ? + 7 = 18 4 518
67

? = 18 – 4 – 6 – 7
?=1 Parity Disk
RAID 3

Block 0
3
2
1 RAID0
Block
Controller
Block
Parity1
Generated
Block 2
Host
Block 3
P0123
RAID 5
Block 0
Block 4

Block 1
Block 5

Parity
RAID4 Block 2
Block 0
4 Block 0
Generated
Controller Block 6
P405
1627 3
Block 3
Host
P4567

P0123
Block 7
RAID 6 – Dual Parity RAID

• Two disk failures in a RAID set leads to data unavailability and data
loss in single-parity schemes, such as RAID-3, 4, and 5
• Increasing number of drives in an array and increasing drive capacity
leads to a higher probability of two disks failing in a RAID set
• RAID-6 protects against two disk failures by maintaining two parities
• Horizontal parity which is the same as RAID-5 parity
• Diagonal parity is calculated by taking diagonal sets of data blocks from the
RAID set members
• Even-Odd, and Reed-Solomon are two commonly used algorithms for
calculating parity in RAID-6
RAID Comparison
Min Storage
RAID Cost Read Performance Write Performance
Disks Efficiency %

Very good for both

random and sequential Very good
0 2 100 Low
read

Good
Slower than a single
Good disk, as every write must
1 2 50 High Better than a single disk be committed to two
disks

(n-1)*100/n Poor to fair for small

Good for random reads
where n= random writes
and very good for
3 3 number of Moderate Good for large,
sequential reads
disks sequential writes

Fair for random write

(n-1)*100/n Very good for random
Slower due to parity
where n= reads
overhead
5 3 number of Moderate Good for sequential
Fair to good for
disks reads
sequential writes

(n-2)*100/n Very good for random

Moderate Good for small, random
where n= reads
6 4 but more writes
number of Good for sequential
than RAID 5 (has write penalty)
disks reads

1+0
and 4 50 High Very good Good
0+1
 RAID Impacts on Performance
RAID Controller

Ep new = Ep old - E4 old + E4 new

2 XOR
Ep new Ep old E4 old E4 new

P0 D1 D2 D3 D4

• Small (less than element size) write on RAID 3 & 5

• Ep = E1 + E2 + E3 + E4 (XOR operations)
• If parity is valid, then: Ep new = Ep old – E4 old + E4 new (XOR operations)
• 2 disk reads and 2 disk writes
• Parity Vs Mirroring
• Reading, calculating and writing parity segment introduces penalty to every write operation
• Parity RAID penalty manifests due to slower cache flushes
• Increased load in writes can cause contention and can cause slower read response times
RAID Penalty Exercise

• Total IOPS at peak workload is 1200

• Read/Write ratio 2:1
• Calculate IOPS requirement at peak activity for
• RAID 1/0
• RAID 5

Additional Task
Discuss impact of sequential &
Random I/O in different RAID
Configuration
Hot Spares

RAID
Controller
 What do you Think ?
Match the following descriptions to the standard RAID configuration levels.

• Parallel access array with • RAID-0

dedicated parity disk • RAID-1
• Striped array with no fault • RAID-2
tolerance
• RAID-3
• Striped array with independent
disks and distributed parity • RAID-4
• Disk mirroring • RAID-5
• Striped array with independent
disks and dedicated parity disk
Chapter Summary

Key points covered in this chapter:

• What RAID is and the needs it addresses
• The concepts upon which RAID is built
• Some commonly implemented RAID levels
 Case Study

Adobe Acrobat
Document

Exercise 1
For Instructor Use Only

Adobe Acrobat
Solution - Case
Document
Study 1

Exercise 2

Solution - Case
Study 2

** Click to Open the Documents

Summarizing
 Exercise & Quiz

z
ui
Q
er
pt
ha
C

Take the Online

Assessment

** Click to Open the Documents

 Check Your Knowledge

• What is a RAID array?

• What benefits do RAID arrays provide?
• What is the primary difference between RAID 3 and RAID 5?
• What is advantage of using RAID 6?
• What is a hot spare?

Additional Task
Read the Original RAID paper “ A
case for redundant arrays of
inexpensive disc