Embed Size (px)
Citation preview
Performance factors for HP ProLiant Serial
Attached SCSI
Technology brief, 2nd edition
Introduction......................................................................................................................................... 2
HDD controller technologies .............................................................................................................. 2
HDD interface technologies ............................................................................................................... 3
Key HDD design parameters for enterprise environments ...................................................................... 3
SAS use for enterprise systems............................................................................................................... 4
HDD performance comparisons............................................................................................................. 4
Performance comparison between SATA and Midline SAS drives .......................................................... 5
Performance comparisons between SAS drives .................................................................................... 7
Performance comparisons between 6 Gb SAS and 3 Gb SAS............................................................. 15
Performance comparisons based on platter speed.............................................................................. 17
SAS drive specifications ..................................................................................................................... 19
Conclusion........................................................................................................................................ 20
For more information.......................................................................................................................... 21
Call to action .................................................................................................................................... 21
2
Introduction
Enterprise-class hard disk drives (HDD) must meet the maximum reliability and scalable performance
goals of the 24/7 enterprise environment. While every minute of server downtime can prove costly,
even a server’s slow performance can increase your operating costs. Hard disk controllers, interfaces,
and drives have evolved to address the reliability and performance issues of the enterprise storage
system.
Serial Attached SCSI (SAS) has become the de facto HDD standard for mission-critical applications.
This paper explains the emergence of SAS and the key parameters of SAS drive technology. It also
includes technical data and comparisons of our latest small form factor SAS drives available at the
time of publication.
HDD controller technologies
HDD controllers use logic-based solutions to enhance the operating efficiency of multiple drives,
including:
Data buffering and read/write caching
Queuing control and read/write reordering
Error management and pre-failure warning
Redundant Array of Independent Drives (RAID)
HDD controllers employ on-board data buffering and caching techniques to optimize the use of slow
drives. Queuing techniques such as Tagged Command Queuing (TCQ) and Native Command
Queuing (NCQ) allow controllers and compatible HDDs to take advantage of the read/write head
position for more efficient drive operations. RAID is another logic-based solution that places data in
stripes across multiple drives to enhance reliability, performance, and data integrity. A variety of RAID
strategies are in use:
RAID 0–Striping to two or more disks (no parity) for performance improvement (no redundancy)
RAID 1–Mirroring data on two disks (no parity) for redundancy, slight performance improvement
RAID 0+1–Striping and then mirroring for redundancy and performance improvement
RAID 1+0 (10)–Mirroring and then striping for redundancy and performance improvement
RAID 3–Striping (byte level) with parity for improved performance and fault tolerance
RAID 4–Striping (block level) with parity for improved performance and fault tolerance
RAID 5–Striping with distributed parity for improved performance and fault tolerance
RAID 50–Striping of RAID 0 with the distributed parity of RAID 5 (RAID 0 array striped across RAID
5 elements requires at least six drives)
RAID 6–Striping with dual distributed parities for improved performance and fault tolerance
RAID 60–Striping of RAID 0 with the distributed double parity of RAID 6 (RAID 0 array striped
across RAID 6 elements requires at least eight disks)
The choice of RAID strategy depends on the desired balance of protection and performance, along
with the number of hard drives available.
For more information about disk drive technology and RAID refer to the HP technology brief titled “Server Drive
Technology” available at
http://h20000.www2.hp.com/bc/docs/support/SupportManual/c01071496/c01071496.pdf
3
HDD interface technologies
Since the days of Advanced Technology Attachment (ATA, also known as Integrated Drive Electronics
or IDE), the server industry has transitioned through several HDD interface technologies:
Small computer system interconnect (SCSI)
Serial Attached ATA (SATA)
Serial Attached SCSI (SAS)
Table 1 lists the key characteristics of these HDD interface technologies.
Table 1. Comparison of HDD interface technologies
SCSI [1] SATA SAS
Transfer/connection type Parallel/shared
bus
Serial/point-to-point Serial/point-to-point
Current bandwidth [2] 320 MB/s 3.0 Gb/s (300
MB/s)
6 Gb/s (600 MB/s)
Future bandwidth growth No Yes, to 6 Gb/s Yes, 12 Gb/s
# of devices supported
per interface per
connection
16 15 [3] 16, 256 [4]
HDD type supported SCSI SATA SAS and SATA
Relative reliability Good Adequate Very good
Best suited for Enterprise servers
(replaced by SAS)
Entry level servers Enterprise servers
NOTES:
[1] SCSI data provided for historical reference only.
[2] Actual data rates are slightly lower due to protocol overhead.
[3] Through the use with a SATA port multiplier
[4] Through the use with SAS port expanders
As Table 1 suggests, the SAS interface offers the best solution for the enterprise environment and has
emerged as the preferred choice for high input/out applications.
Key HDD design parameters for enterprise environments
Enterprise-class HDDs must provide maximum performance under a 100 percent duty cycle and
continuous Input/Output (I/O) workload in a high-vibration environment. While some have used the
term Mean Time Between Failures (MTBF) to express the length of HDD life in general, a more
meaningful benchmark, the Annual Failure Rate (AFR), better defines the estimated life of an HDD in a
continuous, high workload environment.
The AFR is the relation, expressed as a percentage, between the MTBF and the number of hours that
the device is expected to run per year (100 percent duty cycle = 8760 hours per year). For example,
you can calculate the AFR for an enterprise HDD with an MTBF of 1,200,000 hours as follows:
1,200,000 hours/8760 hours = 136.9863 years, then (1 failure/136.9863 years) x 100% = AFR of 0.73%
An AFR of 0.73 percent means that we can expect 0.73 percent of the population of HDDs to fail in
the average year. In other words, in a system of 100,000 drives, we could expect 730 to fail. (These
AFR calculations are for illustration purposes only. The actual failure rate could vary depending on manufacturing
deviations, material quality, and the actual application environment, among other factors.)
4
The AFR is less applicable for smaller systems but is meaningful for high-density infrastructures with
thousands or hundreds of thousands of drives. SAS drives, particularly small form factor (SFF) SAS
drives, typically have the lowest AFR.
Another important factor to HDD performance is seek time, or the time from the initiation of a read or
write action until the time a data transfer begins. The smaller platters of SFF SAS drives yield lower
seek times, an advantage in file servers with frequent random accesses.
SAS use for enterprise systems
This section describes SAS design features that affect the performance and operating efficiency of
SAS HDDs for enterprise-class systems. For a general discussion of SAS technology, refer to the HP
white paper “Serial Attached SCSI Storage Technology” available at the following HP web page:
http://h20000.www2.hp.com/bc/docs/support/SupportManual/c01613420/c01613420.pdf
Since RAID strategies and AFR numbers favor high-density systems, SFF SAS drives are an attractive
solution for large enterprise systems. Compared to 3.5-inch large form factor (LFF) SAS drives,
2.5-inch SFF SAS drives have these advantages:
Lower AFR
15 to 20 percent higher Input/output Operations Per Second (IOPS)
30 percent smaller physical size
Up to half the power consumption
Increasing HDD density in a RAID-based system invariably enhances the overall performance of the
storage system. Increasing the drive density also expands the choice of RAID strategies as indicated in
Table 2.
Table 2. Comparison of drive support in two 1U HP ProLiant servers
Server DL320 G6 DL380 G6
Drive type LFF SAS [1] or SATA SFF SAS or SATA
Number of drives 4 8
RAID support RAID 0, 1 [1] RAID 0, 1, 10, 5, 6, 50, 60 [2]
NOTES:
[1] Requires optional SAS controller
[2] RAID 6, 60, requires 256MB or 512MB with Battery Backed Write Cache and Smart Array Advanced Pack
(SAAP) License key options (standard on P812)
At the drive level, smaller/faster is usually better. SFF drives offer a spindle-per-U advantage that
yields better performance in most systems.
Traditionally, LFF drives have a cost-per-GB advantage over SFF types. For workstation and small-to-
medium business customers, getting the most storage capacity for the money is a key factor in setting
up a system. For the enterprise IT architect, however, the goal should be to get the most work (that is,
more IOPS) out of the system., High density storage systems achieves this goal.
HDD performance comparisons
This section provides performance results of comparisons between HDD interface types, form factors,
and platter speeds. You should consider other factors that play a part in system performance when
evaluating systems, including the operating system, application environment, controller type, RAID
strategy, and general workload.
5
Performance comparison between SATA and Midline SAS drives
Systems supporting SAS can configure a mix of SAS and SATA drives, but you can more easily
achieve your performance goals using SAS drives. You cannot mix SATA and SAS drives within a
logical drive.
Figures 1 through 3 are plots of performance data from benchmark tests of SATA drives and Midline
SAS drives using the following configuration:
System—HP StorageWorks D2600 Modular Smart Array
SAS Controller—HP Smart Array P411, firmware version 3.50, 512 MB cache, RAID 0
Figure 1. Sequential Read operations with HP Smart Array P411 Controller, RAID 0.
Figure 1 above compares read performance between an array of midline SAS drives to an array of
midline SATA drives. Results indicate an array of 6Gb SAS drives can read a single stream of data at
a rate of more than 1.6 GiB/s, as compared to an array of SATA drives reading at a rate of less than
1 GiB/s. The performance difference is due to the SATA interface being less efficient than a SAS
interface. The SATA interface can only run at 3 Gb/s even if the attached drives are rated to higher
speeds, whereas the SAS controller runs at 6 Gb/s with 6Gb SAS drives.
0
200
400
600
800
1000
1200
1400
1600
1800
4KiB 64KiB 256KiB 512KiB 1MiB
Midline SATA
Midline SASMiB
/s
Sequential Read, RAID-0, 12 Drives
6
Figure 2. Sequential Write operations with HP Smart Array P411 Controller, RAID 0
Figure 2 illustrates that for write request sizes beginning with 256KiB, an array of midline SAS drives
can write a single stream of data more than twice as fast as a similar array of SATA drives.
Once again, the performance difference is due to a combination of the SATA interface being less
efficient than a SAS interface and to the SATA interface running at 3 Gb/s, while the SAS controller
runs at 6 Gb/s with 6Gb SAS drives.
Figure 3 below indicates that an array of midline SAS drives can achieve 20% more random IOPS
than a similar array of SATA drives. (For Figures 3 through 11 comparisons, the X-axis values 1
through 256 represent queue depth.)
In theory the drive, RPM for the SATA and SAS drives are the same. This suggests that the SAS
interface is slightly more efficient, resulting in a small performance increase.
The observed difference may result from the drives’ physical characteristics (for example, elevator
sorting.) Another factor may be the limits the SA controller sets on the SATA drive queue depth. The
difference should not influence performance of small-block random workloads.
0
200
400
600
800
1000
1200
1400
1600
1800
4KiB 64KiB 256KiB 512KiB 1MiB
Midline SATA
Midline SAS
Sequential Write, RAID-0, 12 Drives
MiB
/s
7
Figure 3. Random operations of SAS compared to SATA with HP Smart Array Controller P411 Controller, RAID 0
Performance comparisons between SAS drives
This section includes comparisons of SAS drives based on platter speed, along with differences in
RAID strategies.
Figure 4 below shows that a RAID 0 array of ten SAS Small Form Factor (SFF) 15K rpm drives can
read, or write, a single sequential stream of data 30% faster than a similar array of 10K rpm SFF SAS
drives. Thus, a drive spin increase of 5k rpm can result with a 30% performance increase for both
read and write operations.
40%
60%
80%
100%
120%
140%
160%
4KiB RANDOMREAD Aligned
4KiB RANDOMWRITE Aligned
4KiB OLTPAligned
64KiB RANDOMREAD Aligned
64KiB RANDOMWRITE Aligned
1MiB RANDOMREAD Aligned
1MiB RANDOMWRITE Aligned
Re
lati
ve
Pe
rfo
rma
nc
e(
%)
12-SAS-7200, P411, fw:3.50, 512MB, R0(256KB), 1-D2600, 1LD
12-SAS-7200, P411, fw:3.50, 512MB, R0(256KB), 1-D2600, 1LD
RAID-0 Random Performance, P41112 - 1TB 7.2K 6Gb Midline SAS vs 12 - 1TB 7.2K 3Gb SATA
100%= 7.2K rpm SATA
8
Figure 4. Sequential Read and Write comparison of ten10K SFF drives RAID-0 with ten 15K SFF drives RAID-0
50%
60%
70%
80%
90%
100%
110%
120%
130%
140%
150%
64KiBSEQ READ
256KiBSEQ READ
512KiBSEQ READ
1MiBSEQ READ
Rela
tive
Perf
orm
ance
-(%
)
1x10-Drv, P411, fw:3.50, 512MB, R0(256KB), 1-D2700, 1LD
1x10-Drv, P411, fw:3.50, 512MB, R0(256KB), 1-D2700, 1LD
100%= 10K rpm
RAID-0 Sequential Read; P41110-146GB 15K, 6Gb SFF SAS vs 10-146GB 10K, 6Gb SFF SAS
50%
60%
70%
80%
90%
100%
110%
120%
130%
140%
150%
64KiBSEQ WRITE
256KiBSEQ WRITE
512KiBSEQ WRITE
1MiBSEQ WRITE
Rela
tive
Perf
orm
an
ce
-(
%)
1x10-Drv, P411, fw:3.50, 512MB, R0(256KB), 1-D2700, 1LD
1x10-Drv, P411, fw:3.50, 512MB, R0(256KB), 1-D2700, 1LD
100%= 10K rpm
Sequential Write; P411, RAID-010 - 146GB 15K, 6Gb SFF SAS vs 10-146GB 10K, 6Gb SFF SAS
9
Figure 5 below illustrates that a RAID-0 array of ten 15K rpm SFF SAS drives achieves between 35%
and 45% more 4KiB Random Read IOPS than a RAID-0 array of ten 10K rpm SFF SAS drives.
The figure also shows that when running a mixture of 33% 4KiB Random Write and 67% 4KiB
Random Read I/O, ten SFF 15K rpm drives achieves between 25% and 45% more IOPS.
Once again, drives spinning only 5K rpm faster can result in a significant performance increase.
Figure 5. Performance comparison of ten 10K SFF drives Raid 0 with ten 15K SFF drives RAID-0.
Figures 6 and 7 below reveal that a RAID-5 array of four 15K rpm SFF SAS drives is more than 25%
faster in executing a single sequential read stream, or a single sequential write stream, than a RAID-5
array of four 10K SFF SAS drives.
Once again, a higher rpm spin rate will increase performance. However, the performance is not as
great as other combinations due to RAID-5’s additional parity management operations.
40%
60%
80%
100%
120%
140%
160%
4KiB RANREAD Aligned
4KiB OLTPAligned
Re
lati
ve
Pe
rfo
rma
nc
e-
(%
)
1x10-Drv, P411, fw:3.50, 512MB, R0(256KB), 1-D2700, 1LD
1x10-Drv, P411, fw:3.50, 512MB, R0(256KB), 1-D2700, 1LD
100%= 10K; fw:3.50
Random Performance; P411, RAID-010-146GB 15K, 6Gb SFF SAS vs 10-146GB 10K, 6Gb SFF SAS
10
Figure 6. Sequential Read comparison of four 10K SFF drives RAID-5 with four 15K SFF drives RAID-5.
40%
60%
80%
100%
120%
140%
160%
64KiBSEQ READ
256KiBSEQ READ
512KiBSEQ READ
1MiBSEQ READ
Re
lati
ve
Pe
rfo
rma
nc
e-
(%
)
1x4-Drv, P411, fw:3.50, 512MB, R5(256KB), 1-D2700, 1LD
1x4-Drv,P411, fw:3.50,512MB,R5(256KB), 1-D2700,1LD
100%= 10K SFF; fw:3.50
RAID-5 Sequential Read Performance; P4114-146GB 15K, 6Gb SFF SAS vs 4-146GB 10K, 6Gb SFF SAS
11
Figure 7. Write performance comparison of four 10K SFF drives RAID-5 with four 15K SFF drives RAID-5.
Figure 8 below displays performance differences between a RAID-5 array of ten 15K rpm SFF SAS
drives, compared to a RAID-5 array of ten 10K rpm SFF SAS drives. The 15K array is approximately
40% faster on 4KiB, 64KiB and 1MiB random read IOPS.
The chart also illustrate how a faster rotation rate yield speedier access times; 4KiB online transaction
processing (OLTP) is between 13% and 45% faster with the 15K drives; the 64KiB Random Write is
around 20% faster; and 1MiB Random write is more than 30% faster.
40%
60%
80%
100%
120%
140%
160%
64KiBSEQ WRITE
256KiBSEQ WRITE
512KiBSEQ WRITE
1MiBSEQ WRITE
Re
lati
ve
Pe
rfo
rma
nc
e-
(%
)
1x4-Drv, P411, fw:3.50, 512MB, R5(256KB), 1-D2700, 1LD
1x4-Drv,P411, fw:3.50,512MB,R5(256KB),1-D2700,1LD
100%= 10K SFF; fw:3.50
RAID-5 Sequential Write Performance; P4114-146GB 15K, 6Gb SFF SAS vs 4-146GB 10K, 6Gb SFF SAS
12
Figure 8. Random performance comparison of ten10K SFF RAID-5drives and ten15K SFF RAID-5drives
Figure 9 below provides evidence for reading a single sequential stream of data, at higher queue
depths, a RAID-6 array of four SFF drives can approach the same performance achieved by a RAID-5
array of four SFF drives.
Figure 10 indicates RAID-6 single stream sequential write rate is limited to about 67% the rate of the
RAID-5 array. A significant performance variance occurs because in an array configuration consisting
of four drives, two drives in the RAID-6 array are dedicated to parity, while only one of the four drives
in the RAID-5 array is dedicated to parity; (2/3 = 67%).
40%
60%
80%
100%
120%
140%
160%
4KiB RANREAD Aligned
4KiB RANWRITE Aligned
4KiB OLTPAligned
64KiB RANREAD Aligned
64KiB RANWRITE Aligned
1MiB RANREAD Aligned
1MiB RANWRITE Aligned
Re
lati
ve
Pe
rfo
rma
nc
e-
(%
)1x10-Drv, P411, fw:3.50, 512MB, R5(256KB), 1-D2700, 1LD
1x10-Drv, P411, fw:3.50,512MB,R5(256KB),1-D2700, 1LD
100%= 10K rpm SFF SAS
RAID-5 Random Performance; P41110-146GB 15K 6Gb SAS SFF vs 10-146GB 10K 6Gb SAS SFF
13
Figure 9. Sequential Read performance comparison between RAID-6 and RAID-5SFF drives.
40%
60%
80%
100%
120%
140%
160%
4KiBSEQ READ
64KiBSEQ READ
256KiBSEQ READ
512KiBSEQ READ
Re
lati
ve
Pe
rfo
rma
nc
e-
(%
)
1x4-Drv, P411, fw:3.50, 512MB, R6(256KB), 1-D2700, 1LD
1x4-Drv,P411, fw:3.50,512MB,R6(256KB), 1-D2700,1LD
RAID-6 vs RAID-5, Sequential Read Performance4-146GB 15K 6Gb SFF SAS, P411
100%= RAID-5
14
Figure 10. Sequential Write performance comparison between RAID-6 and RAID-5 SFF drives
Figure 11 below shows the difference in random performance between a RAID-6 array of ten 15K
SFF SAS drives and a RAID-5 array with the same ten drives. Note random read performance is
almost identical between the two RAID levels.
The two RAID configurations’ read equivalent performance illustrates that for random read operations,
all drives in the array are available for each random I/O operation.
However, it is different for write operations. For RAID-5, each random write I/O requires reading two
drives (data drive and parity drive) and writing on two drives (data drive and parity drive). For RAID-
6 ,each random write I/O to the array must read three drives (data drive and two parity drives) with
writing operations performed on three drives (data drive and two parity drives). Therefore, with RAID-
6, random write operations are limited to less than 70% of the compared RAID-5 random write rate.
The decision for choosing RAID-5 or RAID-6 is a tradeoff between performance and higher reliability.
40%
60%
80%
100%
120%
140%
160%
64KiBSEQ WRITE
256KiBSEQ WRITE
512KiBSEQ WRITE
1MiBSEQ WRITE
Re
lati
ve
Pe
rfo
rma
nc
e-
(%
)
1x4-Drv, P411, fw:3.50, 512MB, R6(256KB), 1-D2700, 1LD
1x4-Drv,P411, fw:3.50, 512MB, R6(256KB),1-D2700,1LD
RAID-6 vs RAID-5, Sequential Write Performance4-146GB 15K 6Gb SFF SAS, P411
100%= RAID-5
15
Figure 11. Random performance comparison between RAID-6 and RAID-5 SFF drives.
Performance comparisons between 6 Gb SAS and 3 Gb SAS
Today’s SAS technology supports a link rate of 6Gb, allowing twice the bandwidth of previous
generation 3Gb technology. Maximum bandwidth of a 6Gb link is 600 MiB/s while a 3Gb link was
limited to a maximum bandwidth of 300 MiB/s. After taking out communication overhead, this
translates to around 500 MiB/s, and 250 MiB/s, actual data transfer rates per link.
Figures 12 and 13 below display sequential read and sequential write performance for RAID-0
arrays. Both arrays reside in an external enclosure, connected to a Smart Array P411 or a Smart
Array P800 controller, by way of a four-lane SAS port.
The P800 controller interface is limited to a 3Gb SAS link rate, while the more recently introduced
P411controller supports a 6Gb SAS link rate.
Benchmark tests performed on RAID-0 array configurations began with just a single 15K SFF drive
and preceded up to an array of twenty-four SFF15K drives (each drive in the tests was capable of
transfer rates of approximately160 MiB/s).
With small arrays of six or less drives, the performance of the P800 and P411 is comparable.
The performance of the P800 controller is limited to a maximum 3Gb link transfer rate, resulting with
array performance only scaling from one to eight drives. For arrays greater than eight drives the
50%
60%
70%
80%
90%
100%
110%
120%
130%
140%
150%
4KiB RANREAD Aligned
4KiB RANWRITE Aligned
4KiB OLTPAligned
64KiB RANREAD Aligned
64KiB RANWRITE Aligned
1MiB RANREAD Aligned
1MiB RANWRITE Aligned
Re
lati
ve
Pe
rfo
rma
nc
e-
(%
)1x10-Drv, P411, fw:3.50, 512MB, R6(256KB), 1-D2700, 1LD
1x10-Drv, P411, fw:3.50, 512MB, R6(256KB), 1-D2700, 1LD
RAID-6 vs RAID-5, Random Performance10-146GB 15K 6Gb SFF SAS, P411
100%= RAID-5
16
P800 controller’s 3Gb, four-lane SAS port is saturated. Performance no longer increases with
additional drives.
Utilizing 6Gb technology, the P411 controller continues to scale beyond arrays of eight drives until
reaching a saturation point with an array of sixteen or more drives. Maximum performance is
approximately 2000 MiB/s.
Figure 12. RAID-0 256 KiB Sequential Read
0
500
1000
1500
2000
2500
1 2 4 6 8 10 12 16 20 24
P800, 3Gb
P411, 6Gb
RAID-0, 256KiB Sequential Read
Tra
nsf
ter
Rat
e(
MiB
/s)
Drives
17
Figure 13. RAID- 256 KiB Sequential Write
Performance comparisons based on platter speed
The charts in Figure 14 below compare sequential read performance between 10K (rpm) and 15K
SFF SAS drives operating as a 6-drive array with an HP Smart Array P800 controller using RAID 5.
Charts 7A through 7E show the drive array performance for sequential reads per block size at a
specific queue (Q) depth. Queue depth is the number of commands that the storage system is waiting
to execute. In workstations and small server systems, queue depth rarely exceeds 1, but in an
enterprise server environment a large number of users generating requests can result in a queue depth
of tens or hundreds. Chart 7F shows random read IOPS performance.
0
500
1000
1500
2000
2500
1 2 4 6 8 10 12 16 20 24
P800, 3Gb
P411, 6Gb
RAID-0, 256KiB Sequential WriteT
ran
sfte
rR
ate
(M
iB/s
)
Drives
18
Figure 14. Performance comparison: 10K versus 15K RPM, RAID 5 controller and 6 SFF SAS drives
As indicated by the charts, 15K drives show a significant advantage where the requested data blocks
are stripe size (64 KB) or larger. Although the IOPS rating can increase with greater queue depths,
the system is best when the queue depth remains below 32.
Figure 15 below compares the random read performance of 10K and 15K SFF SAS drives for arrays
of three to eight drives, using RAID 5 with the stripe size set to 128 KB.
A: Sequential read, Q1 B: Sequential read, Q4
C: Sequential read, Q8 D: Sequential read, Q16
E: Sequential read, Q32 F: Random IOPS
10K RPM SAS HDD 15K RPM SAS HDD
MiB
/s
Mib
/s
MiB
/s
MiB
/s
MiB
/s
IOPS
19
Figure 15. Random read performance, RAID 5
As indicated in Figure 15, 15K drive arrays consistently provide higher random read IOPS than 10K
drive arrays. Also, overall performance scales with array size (number of drives). Note that
performance is best when retrieving data blocks that are larger than the RAID 5 stripe size (128 KB in
this case). Configure the stripe size accordingly to maximize performance for a specific application.
SAS drive specifications
Table 3 below compares the seek time and power consumption of SFF (2.5-inch) SAS drives available
from HP at the time of publication.
NOTE:
Drive types and specifications are subject to change without notice.
Refer to the following site for a listing and specifications of
currently available SAS drives:
http://h18004.www1.hp.com/products/servers/proliantstorage/
serial/sas/index.html
2500
2000
1500
1000
500
IOPS
No. of drives
10K HDD & P400 controller 15K HDD & 6402 controller
4 KB Random Read
3 4 5 6 7 8
64 KB Random Read
3 4 5 6 7 8
512 KB Random Read
3 4 5 6 7 8
20
Table 3. Comparison of seek time and power consumption of HP SFF SAS drives
Capacity/RPM 36 GB/15K 72 GB/10K 72 GB/15K 146 GB/10K
Seek time:
Single track
Average
Full stroke
0.20 ms
3.0 ms
7.0 ms
0.20 ms
4.0 ms
8.1 ms
0.20 ms
3.0 ms
7.0 ms
0.60 ms
4.1 ms
8.0 ms
Power
consumption:
Idle
Maximum
5.74 watts
7.25 watts
5.96 watts
8.42 watts
6.11 watts
8.29 watts
6.31 watts
8.73 watts
All drives listed In Table 3 have a cache/buffer size of 16 MB and a transfer time of 3 Gb/s.
Conclusion
Today’s server industry uses a variety of hard drive types. The right HDD type is the one that provides
the best return on investment for a particular server application. Economical service and good dollar-
per-gigabyte value SATA drives offer the best value to small-to-medium businesses or environments
with less-than-demanding applications.
In the mission-critical, performance-oriented, enterprise server environment where hard drive duty
cycle is 100 percent, SFF SAS drives excel in performance and reliability. Since SFF drives require
only 70 percent of the space and half the power of 3.5-inch drives, users can achieve higher drive
densities per U space without a significant increase in power consumption. Higher drive densities
provide better overall performance, reliability, and lower operating costs.
© Copyright 2011 Hewlett-Packard Development Company, L.P. The information
contained herein is subject to change without notice. The only warranties for HP
products and services are set forth in the express warranty statements accompanying
such products and services. Nothing herein should be construed as constituting an
additional warranty. HP shall not be liable for technical or editorial errors or omissions
contained herein.
TC1006478, May2011
For more information
Visit the URLs listed below if you
Resource description
QuickSpecs for HP SAS drives
“Serial ATA technology” technology
brief
“Serial-Attached SCSI
technology” technology brief
Call to action
Send comments about this paper to
Follow us on Twitter:
Packard Development Company, L.P. The information
contained herein is subject to change without notice. The only warranties for HP
products and services are set forth in the express warranty statements accompanying
ing herein should be construed as constituting an
additional warranty. HP shall not be liable for technical or editorial errors or omissions
For more information
ed below if you need additional information.
Resource description Web address
QuickSpecs for HP SAS drives http://h18004.www1.hp.com/products/quickspecs/12244_na/12
244_na.html
“Serial ATA technology” technology http://h20000.www2.hp.com/bc/docs/support/SupportManual/c0
0301688/c00301688.pdf
Attached SCSI storage
technology” technology brief
http://h20000.www2.hp.com/bc/docs/support/SupportManual/c0
1613420/c01613420.pdf
Call to action
Send comments about this paper to [email protected]
Follow us on Twitter: http://twitter.com/ISSGeekatHP.
http://h18004.www1.hp.com/products/quickspecs/12244_na/12
http://h20000.www2.hp.com/bc/docs/support/SupportManual/c0
p.com/bc/docs/support/SupportManual/c0
