RAID (Redundant Array of Independent Disks) is a technology that combines multiple disk drive components into a logical unit to increase performance, improve reliability, or both. Common RAID configurations include RAID 0 (striping for performance), RAID 1 (mirroring for redundancy), RAID 5 (striping with parity for redundancy and performance). Hardware and software implementations are used depending on needs. Ongoing developments include faster rebuild times, extended striping, and improved failure prediction.

1. RAIDRAID
Redundant Array ofRedundant Array of
Independent DisksIndependent Disks

2. Outline
Outline
What is RAID?
What is RAID?RAID configurations used.
RAID configurations used.
Performance of each
Performance of each
configuration.
configuration.Implementations.
Implementations.Way forward with RAID.
Way forward with RAID.

3. WhatisRAID?
WhatisRAID?
Stands for Redundant Array of
Stands for Redundant Array of
Independent Disks.
Independent Disks.It’s a technology that enables
It’s a technology that enables
greater levels of performance,
greater levels of performance,
reliability and/or large volumes
reliability and/or large volumes
when dealing with data.
when dealing with data.How??
How?? By concurrent use of two
By concurrent use of two
or more ‘
or more ‘hard disk drives’.
hard disk drives’.
How Exactly??
How Exactly?? Mirroring,
Mirroring,
Stripping (of data) and Error
Stripping (of data) and Error
correction techniques combined
correction techniques combined
with multiple disk arrays give you
with multiple disk arrays give you
the reliability and performance.
the reliability and performance.

4. Mean Time Between FailuresMean Time Between Failures
(MTBF)(MTBF)

5. RAIDflavors
RAIDflavors
Commonly used ones:
Commonly used ones:
1.1.RAID 0
RAID 02.2.RAID 1
RAID 13.3.RAID 5
RAID 54.4.RAID 10
RAID 10Other types used…but
Other types used…but
rarely: RAID 2,3,4,6,50……
rarely: RAID 2,3,4,6,50……

6. RAID 0RAID 0
a.a. It splits data among two or more disks.It splits data among two or more disks.
b.b.Provides good performance.Provides good performance.
c.c. Lack of data redundancy means there isLack of data redundancy means there is
no fail over support with thisno fail over support with this
configuration.configuration.
d.d.In the diagram to the right, the oddIn the diagram to the right, the odd
blocks are written to disk 0 and the evenblocks are written to disk 0 and the even
blocks to disk 1 such that A1, A2, A3,blocks to disk 1 such that A1, A2, A3,
A4, … would be the order of blocks readA4, … would be the order of blocks read
if read sequentially from the beginning.if read sequentially from the beginning.
e.e. Used in read only NFS systems andUsed in read only NFS systems and
gaming systems.gaming systems.

7. RAID 0 analysisRAID 0 analysis
Failure Rate:Failure Rate:
 MTBF of RAID0 is roughly proportional to the numberMTBF of RAID0 is roughly proportional to the number
of disks in the array.of disks in the array.
 Pr(disk fail) = 5%, thenPr(disk fail) = 5%, then
Pr(atleastonefails) = 1 – Pr(nonefails) = 1 – [1-0.05]Pr(atleastonefails) = 1 – Pr(nonefails) = 1 – [1-0.05]22
==
9.75%9.75%
Performance:Performance:
 The fragments are written to their respective disksThe fragments are written to their respective disks
simultaneously on the same sector.simultaneously on the same sector.
 This allows smaller sections of the entire chunk of data toThis allows smaller sections of the entire chunk of data to
be read off the drive in parallel, hence good performance.be read off the drive in parallel, hence good performance.

8. RAID 1RAID 1
•RAID1 is ‘data mirroring’.RAID1 is ‘data mirroring’.
•Two copies of the data are held onTwo copies of the data are held on
two physical disks, and the data istwo physical disks, and the data is
always identical.always identical.
• Twice as many disks are requiredTwice as many disks are required
to store the same data whento store the same data when
compared to RAID 0.compared to RAID 0.
•Array continues to operate so longArray continues to operate so long
as at least one drive is functioning.as at least one drive is functioning.

9. RAID1analysis
RAID1analysis
Failure Rate:
Failure Rate:If Pr(disk fail) = 5%, then
If Pr(disk fail) = 5%, then
the probability of both the
the probability of both the
drives failing in a 2 disk
drives failing in a 2 disk
array is P(both fail) = (0.05)
array is P(both fail) = (0.05)22
= 0.25%.
= 0.25%.
Performance:
Performance:If we use independent disk
If we use independent disk
controllers for each disk,
controllers for each disk,
then we can increase the
then we can increase the
read or write speeds by
read or write speeds by
doing operations in parallel.
doing operations in parallel.

10. RAID 5RAID 5
• RAID 5 is an ideal combinationRAID 5 is an ideal combination
of good performance, good faultof good performance, good fault
tolerance and high capacity andtolerance and high capacity and
storage efficiency.storage efficiency.
• An arrangement of parity andAn arrangement of parity and
CRC to help rebuilding drive dataCRC to help rebuilding drive data
in case of disk failures.in case of disk failures.
• ““Distributed Parity” is the keyDistributed Parity” is the key
word here.word here.

11. RAID5analysis
RAID5analysis
because
ecauseof one disk is not
of one disk is not
quite a harm. We need more
quite a harm. We need more
time if 2 or more disks fail.
time if 2 or more disks fail.
Performance is also as good
Performance is also as good
as RAID 0, if not better. We
as RAID 0, if not better. We
can read and write parallel
can read and write parallel
blocks of data.
blocks of data.One of the drawbacks is that
One of the drawbacks is that
the write involves heavy
the write involves heavy
parity calculations by the
parity calculations by the
RAID controller. Write
RAID controller. Write
operations are slower
operations are slower
compared to RAID 0.
compared to RAID 0.
Pretty useful for general
Pretty useful for general
purpose uses where ‘rea
purpose uses where ‘rea
are more fre
are more freq‘‘w

12. RAID 10RAID 10
a.a. Combines RAID 1 and RAIDCombines RAID 1 and RAID
0.0.
b.b. Which means having theWhich means having the
pleasure of both - goodpleasure of both - good
performance and good failoverperformance and good failover
handling.handling.
c.c. Also called ‘Nested RAID’.Also called ‘Nested RAID’.

13. Implementations
Implementations
tware based RAID:
tware based RAID:
Software implementations
Software implementations
are provided by many
are provided by many
Operating Systems.
Operating Systems.
A software layer sits above
A software layer sits above
the disk device drivers and
the disk device drivers and
provides an abstraction layer
provides an abstraction layer
between the logical
between the logical
drives(RAIDs) and physical
drives(RAIDs) and physical
drives.
drives.
Server's processor is used to
Server's processor is used to
run the RAID software.
run the RAID software.
Used for simpler
Used for simplerconfigurations like RAID
configurations like RAID
and RAID1.
and RAID1.

14. Implementations (contd…)Implementations (contd…)

of RAID req
of RAID re
special-purpose RAID
special-purpose RAID
controller.
controller.
On a desktop system this
On a desktop system this
may be built into the
may be built into the
motherboard.
motherboard.
Processor is not used for
Processor is not used for
RAID calculations as a
RAID calculations as a
separate controller present.
separate controller present.
A PCI-bus-based, IDE/ATA hard disk
RAID
controller, supporting levels 0, 1, and 01.

15. What’s happening present day?What’s happening present day?
RAID 6:RAID 6:
a.a.It is seen as the best way toIt is seen as the best way to
guarantee data integrity as it usesguarantee data integrity as it uses
double parity.double parity.
b.b.Lesser MTBF compared toLesser MTBF compared to
RAID5.RAID5.
c.c.It has a drawback though ofIt has a drawback though of
longer write time.longer write time.

16. Roadahead…
Roadahead…
a virtual
virtual
certainty.
ertainty.
RAID vendors to support
RAID vendors to support
"fast rebuild" features that
"fast rebuild" features that
can restore hundreds of
can restore hundreds of
gigabytes in just an hour or
gigabytes in just an hour or
so!!so!!
Striping(of data) would
Striping(of data) would
extend across RAID groups
extend across RAID groups
-- not just across drives
-- not just across drives
within a group.
within a group.Improved disk diagnostic
Improved disk diagnostic
features should offer more
features should offer more
reliable predictions of
reliable predictions of
impending drive failures,
impending drive failures,
allowing the rebui
allowing the rebui
to begi
to begi

17. References
References
hstorage.tec
hstorage.tec
arget.com/generic/0,2
rget.com/generic/0,2
95582,sid5_gci1196310,0
95582,sid5_gci1196310,0
0.html
0.html
http://www.lascon.co.uk
http://www.lascon.co.uk
/d008005.htm
/d008005.htmhttp://en.wikipedia.org/
http://en.wikipedia.org/
wiki/RAID_0#RAID_0
wiki/RAID_0#RAID_0
http://en.wikipedia.org/
http://en.wikipedia.org/
wiki/RAID
wiki/RAIDPaper - A Case for
Paper - A Case for
Redundant Arrays of
Redundant Arrays of
Inexpensive Disks
Inexpensive Disks
(RAID) - David A
(RAID) - David A
Patterson, Garth Gibson
Patterson, Garth Gibson
and Randy H K
and Randy H K