HP 166207-B21 Compaq RAID Advanced Data Guarding: A Cost-effective, Fault-tole - Page 5

Rovides, Ptimum, Olerance

Page 5 highlights

TECHNOLOGY BRIEF Compaq RAID Advanced Data Guarding Note: RAID ADG is standard with the fourchannel Compaq Smart Array 5300 (SA 5304/256) Controllers and is sold as an option for the twochannel Smart Array models (SA 5302/128). RAID ADG requires a minimum of 64 MB read/write cache. ... RAID ADG PROVIDES OPTIMUM FAULT TOLERANCE Often, the terms "reliability" and "fault tolerance" are used interchangeably in describing RAID schemes; however, there is a distinction between them. Reliability refers to the likelihood that an individual drive or drive array will continue to function without experiencing a failure. Reliability is typically measured over some period of time. Although reliability is a very important issue for customers, it is a function of the reliability of the individual array components rather than the RAID technology implemented. Fault tolerance, on the other hand, is the ability to withstand and recover from a failure despite disk drive reliability issues. Fault tolerance is provided by some sort of redundancy-mirroring, parity, or a combination of both-and it is typically measured by the number of drives that can fail without bringing down the entire array. The fault tolerance of various RAID levels is as follows: • RAID 0 has no fault tolerance because it uses no type of redundancy. The array will fail if one physical drive fails. • With RAID 1 or 1+0, the maximum number of hard drives that can fail without failure of the array is n/2, assuming none of the failed drives are mirrored to each other. In practice, logical drive failure will usually occur before this maximum number is reached, since the probability of a newly failed drive not being mirrored to a previously failed drive becomes increasingly small as the number of failed drives increases. A RAID 1+0 array will fail if a drive and its mirrored drive fail, but the probability of this decreases as the number of mirrored pairs increase. • RAID 5 can withstand the failure of one physical drive. The array will fail if two drives fail simultaneously. • RAID ADG can withstand the failure of two physical drives. Three hard drives must fail before the entire array will fail. RAID ADG also protects against the loss of data if a drive fails and a defect occurs in a single sector of another drive. This is important if data is being rebuilt after a drive failure and a media defect occurs in one of the good drives. Although RAID 1 and RAID 1+0 provide a higher level of fault tolerance than RAID 5, this protection comes at a very high price because 50 percent of the drives are dedicated to fault protection. For RAID 5 configurations, we recommend using no more than 14 physical drives be used per array due to the increased likelihood of drive array failure with more hard drives. For arrays with more than 14 drives, we recommend RAID ADG due to its fault tolerance and storage efficiency. RAID ADG supports the use of up to 56 physical drives per array because that is the maximum number of hard drives that can be physically connected to a four-channel Compaq Smart Array Controller. Figure 1 shows the relative probability of logical drive failure for different RAID settings and different logical drive sizes, assuming no online spares are present. RAID ADG is less likely to experience a logical drive failure than RAID 0, 5, and 1+0. An online spare (hot spare) can be added to any of the fault-tolerant RAID levels to further decrease the probability of logical drive failure. As soon as drive failure occurs, missing data can be automatically rebuilt from parity data onto the online spare. Without an online spare there is a chance that more hard drives will fail before the failed drive is replaced, leading to failure of the complete drive array and consequent loss of data. This is less likely with RAID ADG than with RAID 5, since RAID ADG can sustain failure of two drives. RAID ADG supports online spare drives and Online RAID Level Migration from RAID 1 or RAID 5.1 1 For more information about online spare drives and Online RAID Level Migration from RAID 1 or RAID 5, please refer to the Advanced Data Guarding for the Smart Array 5300 Controller Reference Guide, part number 188270-001 at www.compaq.com. TC020604TB 5

  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
  • 7
  • 8
  • 9

T
ECHNOLOGY
B
RIEF
Compaq RAID Advanced Data Guarding
5
TC020604TB
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
RAID ADG P
ROVIDES
O
PTIMUM
F
AULT
T
OLERANCE
Often, the terms “reliability” and “fault tolerance” are used interchangeably in describing RAID
schemes; however, there is a distinction between them. Reliability refers to the likelihood that an
individual drive or drive array will continue to function without experiencing a failure. Reliability
is typically measured over some period of time. Although reliability is a very important issue for
customers, it is a function of the reliability of the individual array components rather than the
RAID technology implemented.
Fault tolerance, on the other hand, is the ability to withstand and recover from a failure despite
disk drive reliability issues. Fault tolerance is provided by some sort of redundancy—mirroring,
parity, or a combination of both—and it is typically measured by the number of drives that can fail
without bringing down the entire array. The fault tolerance of various RAID levels is as follows:
RAID 0
has no fault tolerance because it uses no type of redundancy. The array will fail if one
physical drive fails.
With
RAID 1 or 1+0
, the maximum number of hard drives that can fail without failure of the
array is n/2, assuming none of the failed drives are mirrored to each other. In practice, logical
drive failure will usually occur before this maximum number is reached, since the probability
of a newly failed drive not being mirrored to a previously failed drive becomes increasingly
small as the number of failed drives increases. A RAID 1+0 array will fail if a drive and its
mirrored drive fail, but the probability of this decreases as the number of mirrored pairs
increase.
RAID 5
can withstand the failure of one physical drive. The array will fail if two drives fail
simultaneously.
RAID ADG
can withstand the failure of two physical drives. Three hard drives must fail
before the entire array will fail. RAID ADG also protects against the loss of data if a drive
fails and a defect occurs in a single sector of another drive. This is important if data is being
rebuilt after a drive failure and a media defect occurs in one of the good drives.
Although RAID 1 and RAID 1+0 provide a higher level of fault tolerance than RAID 5, this
protection comes at a very high price because 50 percent of the drives are dedicated to fault
protection. For RAID 5 configurations, we recommend using no more than 14 physical drives be
used per array due to the increased likelihood of drive array failure with more hard drives. For
arrays with more than 14 drives, we recommend RAID ADG due to its fault tolerance and storage
efficiency. RAID ADG supports the use of up to 56 physical drives per array because that is the
maximum number of hard drives that can be physically connected to a four-channel Compaq
Smart Array Controller.
Note: RAID ADG is
standard with the four-
channel Compaq Smart
Array 5300 (SA 5304/256)
Controllers and is sold as
an option for the two-
channel Smart Array models
(SA 5302/128). RAID ADG
requires a minimum of
64 MB read/write cache.
Figure 1 shows the relative probability of logical drive failure for different RAID settings and
different logical drive sizes, assuming no online spares are present. RAID ADG is less likely to
experience a logical drive failure than RAID 0, 5, and 1+0. An online spare (hot spare) can be
added to any of the fault-tolerant RAID levels to further decrease the probability of logical drive
failure. As soon as drive failure occurs, missing data can be automatically rebuilt from parity data
onto the online spare. Without an online spare there is a chance that more hard drives will fail
before the failed drive is replaced, leading to failure of the complete drive array and consequent
loss of data. This is less likely with RAID ADG than with RAID 5, since RAID ADG can sustain
failure of two drives. RAID ADG supports online spare drives and Online RAID Level Migration
from RAID 1 or RAID 5
.
1
www.compaq.com
1
For more information about online spare drives and Online RAID Level Migration from RAID 1
or RAID 5, please refer to the
Advanced Data Guarding for the Smart Array 5300 Controller
Reference Guide
, part number 188270-001 at
.