Seagate and Seagate Technology are registered tradem arks of Seagate Technology LLC.
Cheetah, SeaTools, SeaF ONE, SeaBOARD, Sea TDD, and the Wave logo are eith er regis
tered trademarks or trademarks of Seagate T echnology LLC. Other product names are registered trademarks or trademarks of their owners.
Seagate reserves the right to ch ange, witho ut notice, product offerings or spec ifications . No
part of this publication may be reproduc ed in any form w ithout wr itte n per mi ssio n of Seagate
Technology LLC.
-
Revision status summary sheet
RevisionDateSheets Affected
Rev. A11/22/04All.
Rev. B05/04/05Pages 11, 37, 58, and 59.
This manual describes Seagate Technology® LLC, Cheetah® SAS (Serial Attached SCSI) disc drives.
Cheetah drives support the SAS Pr oto co l spec ifi cati ons to the exte nt de sc ribe d in this man ual . The SAS Inter-
face Manual (part number 100293071) describes t he general SAS characteri stics of this and other Se agate
SAS drives.
Figure 1. Cheetah 15K.4 SAS disc drive
Cheetah 15K.4 SAS Product Manual, Rev. B1
2Cheetah 15K.4 SAS Product Manual, Rev. B
2.0Standards, compliance and reference documents
The drive has been developed as a system peripheral to the highest standards of design and construction. The
drive depends on its host equip ment to provide adequate power and e nvironment for optimum performa nce
and compliance with appl icable i ndustry and g overnme ntal regulations. Special attention mu st be given in the
areas of safety, power distribution, s hielding, audible n oise control, and temperature regulat ion. In particular,
the drive must be securely mo unted to guar antee the spe cified pe rformance char acterist ics. Mountin g by bot
tom holes must meet the requirements of Section 8.3.
2.1Standards
The Cheetah SAS family com pl ies wi th Se aga te s tandar ds a s no ted in the app ropr iat e se cti on s of this ma nual
and the Seagate SAS Interface Manual, part number 100293071.
The Cheetah disc drive is a UL recognized component per UL1950, CSA certified to CAN/CSA C22.2 No. 950-
95, and VDE certified to VDE 0805 and EN60950.
2.1.1Electromagnetic compatibility
The drive, as delivered , is designed for system integrati on and installation into a suitable encl osure prior to
use. The drive is supplied as a suba ssembly and is not subject to Sub part B of Part 15 of the FCC Rules and
Regulations nor the Radio Interference Regulations of the Canadian Department of Communications.
The design characteristics of the drive serve to minimize radiation when installed in an enclosure that provides
reasonable shielding. T he drive is ca pable of meeting the Cl ass B limits of the FCC Rules and Regulati ons of
the Canadian Department of C omm uni ca tio ns whe n p ro per ly pack ag ed; howe ve r, it is the user’s responsibil ity
to assure that the dri ve meets the approp riate EMI r equirements in thei r system. Sh ielded I/O c ables may be
required if the enclosure does not prov ide adequate shie lding. If the I/O cables are external to the enc losure,
shielded cables should be used, with the shields grounded to the enclosure and to the host controller.
-
2.1.1.1Electromagnetic susceptibility
As a component a ssembly, the drive is not requ ired to meet any susc eptibi lity perfor mance requ irements. It is
the responsibilit y of those integrating the dr iv e wit hin the ir sy ste ms to pe rfo rm th os e te sts r eq ui red a nd des ig n
their system to ensure th at equipment operating in the same sy stem as the drive or external to the syste m
does not adversely affect the performance of the drive. See Tables
Cheetah 15K.4 SAS Product Manual, Rev. B3
2 and 3, DC power requirements.
2.2Compliance
2.2.1Electromagnetic compliance
Seagate uses an inde pendent laborato ry to confirm comp liance with the directives/standa rds for CE Markin g
and C-Tick Marking. The drive was tested in a representative system for typical applications. The selected sys
tem represents the most popular characteristics for test platforms. The system configurations include:
• Typical current use microprocessor
• 3.5-inch floppy disc drive
• Keyboard
• Monitor/display
• Printer
• External modem
•Mouse
Although the test system with this Seagate model complies with the directives/standards, we cannot guarantee
that all systems will comply. The computer manufacturer or system integrator shall confirm EMC compliance
and provide the appropriate marking for their product.
Electromagnetic compliance for the European Union
If this model has the CE Marking it co mplies with the European Union requirements of the El ectromagnetic
Compatibility Directive 89 /336/EEC of 03 May 1989 as amend ed by Dir ective 92 /31/ EEC of 28 Apri l 1992 an d
Directive 93/68/EEC of 22 July 1993.
-
Australian C-Tick
If this model has th e C-Tick Marking it complies with the Austr alia/New Zeal and Standard AS/NZS3548 199 5
and meets the Electromagnet ic Compatibility (EMC) Framework requi rements of Australia’s Spectrum Man
agement Agency (SMA).
Korean MIC
If this model has the Korean Ministry of Information and Communication (MIC) logo, it complies with paragraph
1 of Article 11 of the Electromagnetic Compatibility (EMC) Control Regulation and meets the Electroma gnetic
Compatibility Framework requirements of the Radio Research Laborator y (RRL) Ministry of Information and
Communication Republic of Korea.
Taiwanese BSMI
If this model has t wo Chine se words meani ng “E MC ce rtifica tion” follo wed by an eig ht digi t identifi catio n n umber, as a Marking, it complies with Chinese National Standard (CNS) 13438 and meets the Electromagnetic
Compatibility (EMC) Fra mework requ irements of the Taiwanese Bureau of Standards, Metrology, and Inspec
tion (BSMI).
-
-
4Cheetah 15K.4 SAS Product Manual, Rev. B
2.3Reference documents
Cheetah 15K.4 SAS Installation Guide
Seagate part number: 100350602
SAS Interface Manual
ANSI SAS Documents
ANSI Small Computer System Interface (SCSI) Documents
Specification for Acoustic Test Requirement and Procedures
Package Test Specification
Package Test Specification
In case of conflict between this document and any referenced document, this document takes precedence.
Seagate part number: 100293071
SFF-83233.5” Drive Form Factor with Serial Connector
SFF-8460HSS Backplane Design Guidelines
SFF-8470Multi Lane Copper Connector
SFF-8482SAS Plug Connector
ANSI INCITS.xxx Serial Attached SCSI (SAS) Standard (T10/1562-D)
ISO/IEC 14776-xxx SCSI Architecure Model-3 (SAM-3) Standard (T10/1561-D)
ISO/IEC 14776-xxx SCSI Primary Commands-3 (SPC-3) Standard (T10/1416-D)
ISO/IEC 14776-xxx SCSI Block Commands-2 (SBC-2) Standard (T10/1417-D)
X3.270-1996(SCSI-3) Architecture Model
Seagate part number: 30553-001
Seagate P/N 30190-001 (under 100 lb.)
Seagate P/N 30191-001 (over 100 lb.)
Cheetah 15K.4 SAS Product Manual, Rev. B5
6Cheetah 15K.4 SAS Product Manual, Rev. B
3.0General description
Cheetah drives combin e giant magnetoresisti ve (GMR) heads, partia l response/maximum likelihood (PRML)
read channel electroni cs, e mbedded servo te chnol ogy, and a Serial Attached SCSI (SA S) inter face t o provid e
high performance, h igh c apaci ty d ata s tor age f or a v ari ety o f sy st ems inc l uding engineering workstations, net
work servers, mainframes, and supercomputers. The Serial Attached SCSI interface is designed to meet next-
generation computing demands for performance, scalability, flexibility and high-density storage requirements.
Cheetah drives are rando m ac c ess s to ra ge d evic es de si gne d to support the Serial Attached SCS I P rotoc ol as
described in the ANSI spec ifications , this documen t, and the SAS Inte rface Manual ( part number 1002 93071)
which describes the g eneral interface characteristics o f this drive. Chee tah drives are classi fied as intelligen t
peripherals and provide level 2 confo rm ance ( hig hes t lev el) with the A NSI SC SI- 1 s tandard. Th e SA S connec
tors, cables and elec trical interface are com patible with Serial ATA (SATA), giving future user s the choice of
populating their systems with either SAS or SATA hard disc drives. This allows you to continue to leverage your
existing investment in SCSI while gaining a 3Gb/s serial data transfer rate.
The head and disc assembly (HD A) is sealed at the factory. Air recircul ates within the HDA through a non-
replaceable filter to maintain a contamination-free HDA environment.
Note.Never disassemb le the HDA and do not attemp t to service items in the sealed enclosure (heads,
media, actuator, etc.) as this requires special facilities. The drive does not contain user-replaceable
parts. Opening the HDA for any reason voids your warranty.
Cheetah drives use a dedicated landing zone at the innermost radius of the media to eliminate the possibility of
destroying or degrading data by landing in the data zone. The heads automatically go to the landing zone when
power is removed from the drive.
-
-
An automatic shipping lock prevents potential damage to the heads and discs that results from movement dur-
ing shipping and h andling. The shipping lock disengages and th e head load process be gins when power is
applied to the drive.
Cheetah drives decode track 0 location data from the servo data embedded on each surface to eliminate
mechanical transducer adjustments and related reliability concerns.
The drives also use a high-performance actuator assembly with a low-inertia, balanced, patented, straight arm
design that provides excellent performance with minimal power dissipation.
Cheetah 15K.4 SAS Product Manual, Rev. B7
3.1Standard features
Cheetah drives have the following standard featur es:
• 1.5 / 3Gbit Serial Attached SCSI (SAS) interface
• Integrated dual port SAS controller supporting the SCSI protocol
• Support for SAS expanders and fanout adapters
• Firmware downloadable using the SAS interface
• 64 - deep task set (queue)
• Supports up to 32 initiators
• Jumperless configuration.
• User-selectable logical block size (512 to 528 bytes per logical block) in any multiple of 4 bytes.
• Industry standard SFF 2.5-inch dimensions
• Programmable logical block reallocation scheme
• Flawed logical block reallocation at format time
• Programmable auto write and read reallocation
• Reallocation of defects on command (Post Format)
• ECC maximum burst correction length of 320 bits
• No preventive maintenance or adjustments required
• Dedicated head landing zone
• Embedded servo design
• Automatic shipping lock
• Embedded servo design
• Self diagnostics performed when power is applied to the drive
• Zone bit recording (ZBR )
• Vertical, horizontal, or top down mounting
• Dynamic spindle brake
• 8,192 Kbyte data buffer (see Section 4.5).
• Drive Self Test (DST)
• Background Media Scan (BGMS)
•Power Save
3.2Media description
The media used on the drive has an alumi num sub st rat e co ated with a thin fi lm magn eti c mater ia l, over coate d
with a proprietary protective layer for improved durability and environmental protection.
3.3Performance
• Programmable multi-segmentable cache buffer
• 300 Mbytes/sec maximum instantaneous data transfers.
• 15k RPM spindle. Average latency = 2.0 msec
• Background processing of queue
• Supports start and stop commands (spindle stops spinning)
• Incorporates industry-standard Self-Monitoring Analysis and Reporting Technology (S.M.A.R.T.)
• 5-year warranty
3.5Formatted capacities
Standard OEM models are formatted to 512 bytes per block. The block size is sele ctable at format time an d
must be a multiple of 4 by tes. Users having the necessary equi pment may modify the data block size before
issuing a format command and obtain different formatted capacities than those listed.
To provide a stable target capacity environment and at the same time provide users with flexibility if they
choose, Seagate recommends product planning in one of two modes:
1. Seagate designs specify capacit y p oi nts at cer tain b lock siz es tha t S eag ate guarantees current and fut ur e
products will meet. We recommend custo mers use this capacity in their project pl anning, as it ensures a
stable operating point with backw ard and forward c ompatibi lity from gener ation to generati on. The curren t
guaranteed operating points for this product are:
2. Se aga te dri ve s a ls o m ay be u se d at the max im um av ail ab le c apacit y at a given block size, but the excess
capacity above the gua rantee d l ev el wi ll v ar y between other drive fami li es an d from generation to genera
tion, depending on how eac h block size actually formats out for zo ne frequencies and splits over servo
bursts. This added capaci ty potent ial may range from 0.1 to 1.3 percent above th e guarant eed capacities
listed above. Us ing th e dr ives in th is mann er gi ves the ab so lute m aximu m capacity poten tial, but the user
must determine if the extra ca pacity potential is useful, or whether their assurance of backward and for
ward compatibility takes precedenc e.
3.6Programmable drive capacity
Using the Mode Select command, the drive can change its capacity to something less than maximum. See the
Mode Select (6) parameter list table in the SAS Interface Ma nual, part numb er 100293071. A value of z ero in
the Number of Blocks field indicates that the drive will not change the capacity it is currently formatted to have.
A number other than z er o a nd less tha n t he maximum number of L BA s in the Num ber of B lock s field c han ges
the tota l d rive capacity t o t he v al u e in t h e N umb er o f Bl o cks f ie l d. A v al u e gr ea t er t han t h e m ax im u m nu m be r of
LBAs is rounded down to the maximum capacity.
-
-
3.7Factory-install ed accessories
OEM standard drives are shipped with the Cheetah 15K.4 SAS Installation Guide (part number 100350602).
Cheetah 15K.4 SAS Product Manual, Rev. B9
3.8Factory-install ed options
You may order the following items which ar e incorporated at the manufac turing facility during produ ction or
packaged before shipping. Some of the options available are (not an exhaustive list of possible options):
• Other capacities can be ordered depending on sparing scheme and sector size requested.
• Single-unit shipping pack. The drive is normally shipped in bulk packaging to provide maximum protec tion
against transit damage. Units shipped individually require additional protection as provided by the single unit
shipping pack. Users planning single unit distribution should specify this option.
• The Cheetah 15K.4 SAS Installatio n Guid e, part numb er 1003 50602, is u suall y included with each standar d
OEM drive shipped, but extra copies may be ordered.
• The Safety and Regulatory Agency Specifications, part number 75789512, is usually included with each
standard OEM drive shipped, but extra copies may be ordered.
3.9User-installed accessories
The following accessories are available. All kits may be installed in the field.
• Single-unit shipping pack.
10Cheetah 15K.4 SAS Product Manual, Rev. B
4.0Performance characteristics
This section provides detailed information concerning performance-related characteristics and features of
Cheetah drives.
4.1Internal drive characteristics
ST3146854SS ST373454SSST336754SS
Drive capaci ty146.873.436.7Gbytes (formatted, rounded off value)
Read/write data heads842
Bytes per track471,916471,916471,916Bytes (average, rounded off values)
Bytes per surface18,35118,35118,351Mbytes (unformatted, rounded off value)
Tracks per surface (total)50,86450,86450,864Tracks (user accessible)
Tracks per inch85,00085,00085,000TPI
Peak bits per inch628628628KBPI
Internal data rate685-1,142685-1,142685-1,142Mbits/sec (variable with zone)
Disc rotation speed15k15k15krpm
Avg rotational latency2.02.02.0msec
4.2Seek performance characteristics
See Section 9.4.1, "SAS physical interface" on page 57 and the SAS Interface Manual (part number
100293071) for additional timing details.
4.2.1Access time
Not Including controller overhead
(msec)
ReadWrite
AverageTypical3.54.0
Single trackTypical0.20.4
Full strokeTypical7.47.9
1.Typical access times are measured under nominal conditions of temperature, voltage, and horizontal orientation as measured on a representative sample of drives.
2.Access to data = access time + latency time.
1, 2
Cheetah 15K.4 SAS Product Manual, Rev. B11
4.2.2Format command execution time (minutes)
ST3146854SSST373454SSST336754SS
Maximum (with verify)
Maximum (without veri fy )
906030
453015
4.2.3General performance characteristics
Minimum sect or interleave1 to 1
Data buffer to/from disc media (one 512-byte logical block)*
Minimum685 Mbits/sec
Maximum1,142 Mbits/sec
SAS Interface maximum instantaneous transfer rate300 Mbytes/sec* per port
Logical block sizes
Default is 512-byte data blocks
Variable 512 to 704 byt es per sector in even number of bytes per sector. If n (number
of bytes per sector) is odd, then n-1 will be used.
Read/write consecutive sectors on a trackYes
Flaw reallocation performance impact (for flaws reallocated at format time using
the spare sectors per sparing zone reallocation scheme.)
Average rotation al lat enc y2.0 msec
*Assumes no errors and no relocated logical blocks. Rate measured from the start of the first logical block transfer to or
from the host.
(dual port = 600 Mbytes/sec*)
Negligible
4.3Start/stop time
The drive accepts the comma nds list ed in the SAS Inte rface Man ual less than 3 seconds after DC power has
been applied.
If the drive receives a NOTIFY (ENABLE SPINUP) primitive through either port and has not received a START
STOP UNIT command with the START bit equal to 0, the drive becomes ready for normal operations within 20
seconds (excluding the error recovery procedure).
If the drive receives a START STOP UNIT command with the START bit equal to 0 before receiving a NOTIFY
(ENABLE SPINUP) primitive, the drive waits for a START STOP UNIT command with the START bit equal to 1.
After receiving a START STOP UNIT command with the START bit equal to 1, the drive waits for a N OTIFY
(ENABLE SPINUP) primit ive. After receiving a NOTIFY (ENA BLE SPINUP) primitive throu gh either port, the
drive becomes ready for normal operations within 20 seconds (excluding the error recovery procedure).
12Cheetah 15K.4 SAS Product Manual, Rev. B
If the drive receives a START STOP UNIT command with the START bit and IMMED bit equal to 1 and d oes
not receive a NOTIFY (ENABL E SPINUP) primitive within 5 seco nds, the drive fails the START STOP UNIT
command.
The START STOP UNIT command may be used to command the drive to stop the spindle. Stop time is 30 seconds (maximum) from removal of DC power. There is no power control switch on the drive.
4.4Prefetch/multi-segmented cache control
The drive provides a prefetch ( read look-ahead) and multi-se gmented cache control algori thms that in many
cases can enhan ce system performance. Cac he refers to the drive buffer stor age space when it is used in
cache operations . To select this feature, th e host sends the Mode Select command with the p roper values in
the applicable bytes in page 08h. Prefetch and cache operations are independent features from the standpoint
that each is enabled and dis abled indepe ndently using the Mode Select c ommand; however, in actual opera
tion, the prefetch feature overlaps cache operation somewhat as described in sections 4.5.1 and 4.5.2.
All default cache and prefetch mode paramete r values (Mode Page 08 h) for standard OEM versions of this
drive family are given in Table
9.3.2.1.
4.5Cache operation
Note.Refer to the SAS Interface Manual for more detail concerning the cache bits.
Of the 8,192 kbytes p hysical buffer space in the dri ve, approximately 7,077 kbytes can be u sed as a cache.
The buffer can be divided into logical segments (using Mode Select Page 08h, byte 13) from which data is read
and to which data is written.
-
The drive keeps track of the logical block addresses of the data stor ed in each segment of the buffer. If the
cache is enabled (see RCD bit in the SAS Interface Manual
is retrieved from the buffer, if possible, before any disc access is initiated. If cache operation is not enabled, the
buffer (still segmented with the required number of segments) is still used, but only as circular buffer segments
during disc medium read oper ations (di sregardin g Prefetch oper ation for th e moment). That i s, the drive does
not check in the buffer segments for the requested read data, but goes directly to the medium to retrieve it. The
retrieved data merely passes throug h some buffer seg ment on the wa y to the hos t. On a c ache miss, all data
transfers to the host are in accorda nce with bu ffer-full ratio rules. On a c ache hit , the drive ig nores the bufferfull ratio rules. See the explanation provided with the information about Mode Page 02h (disconnect/reconnect
control) in the SAS Interface Manual.
The following is a simplified description of the prefetch/cache operation:
Case A—read command is received and the first logical block is already in the cache:
1. Drive transfers to the initiator the first logical block requested plus all subsequent contiguous logical blocks
that are already in the cache. This data may be in multiple segments.
2. Wh en a requested logical block is reached that is not in any cache se gment, the d rive fetches it and any
remaining requested lo gical b lock a ddres ses from th e disc and puts them in a seg ment o f the cache . The
drive transfers the r emaining requested logical blocks from t he cache to the host in a ccordance with the
Mode Select Disconnect/Reconnect parameters, page 02h.
3. If the prefetch feature is enabled, refer to section 4.5.2 for operation from this point.
Case B—A Read command requests data, and the first logical block is not in any segment of the cache:
1. The drive fetc hes the req uested lo gical bloc ks from t he disc and transfers them int o a segme nt, and the n
from there to the host in accordance with the Mode Select Disconnect/Reconnect parameters, page 02h.
), data requested by the host with a read command
2. If the prefetch feature is enabled, refer to section 4.5.2 for operation from this point.
Cheetah 15K.4 SAS Product Manual, Rev. B13
During a prefetch, the drive cross es a cylinder boundary to fetc h data only if the Disco ntinui ty (DISC) bit is se t
to 1 in bit 4 of byte 2 of the Mode Select parameters page 08h. Default is zero for bit 4.
Each cache se gment is actually a s elf-contained circula r buffer whose length is an integer number of logical
blocks. The wrap-ar ound capability of the individual segm ents greatly enhances the cache’s overa ll perfor
mance, allowing a wide range of user-sele ctable configurations. The d rive supports operation of any integer
number of segments from 1 to 16. D ivide the 7 ,077 Kbyt es in the b uffer by the numbe r of seg ments to get the
segment size. Default is 3 segments.
Note.The size of eac h segment is not reported by Mode Sense command page 08h, bytes 14 and 15.
The value 0XFFFF is always reported regardless of the actual size of the segm ent . Se ndi ng a siz e
specification using the Mode Select command (byt es 14 and 15) does not set up a new segment
size. If the STRICT bit in Mode page 00h (byte 2, bit 1) is set to one , the driv e respon ds as it does
for any attempt to change an unchangeable parameter.
4.5.1Caching write data
Write caching is a write operation by the drive th at mak es u se of a drive buffer storage ar ea wher e the da ta to
be written to the medium is stored while the drive performs the Write command.
If read caching is enabled (RCD=0), then data written to the medium is retained in the cache to be made available for future read cache hits. The s ame buffer space an d segme ntation is use d as set up for read func tions.
The buffer segmentation s cheme is set up or changed independently, having nothing to do with the state of
RCD. When a write command is issued, i f RCD=0, the cache is first checke d to see if any logical blocks tha t
are to be written are a lready stored in the cache f rom a previous read or write comman d. If there are, the
respective cache segments are cleared. The new data is cached for subsequent Read commands.
-
If the number of write data logical blocks exceed the size of the segment being written into, when the end of the
segment is reached, the data is written into the beginning of the same cache segment, overwriting the data that
was written there at the beginning of the operation; however, the drive does not overwrite data that has not yet
been written to the medium.
If write caching is enabled (WCE=1), then the drive may return Good status on a write command after the data
has been transferred into the cache, but befor e the data has been written to the medium. If an error oc curs
while writing the data to the medium, and Good status has already been returned, a deferred error will be gen
erated.
The Synchronize Cache command may be used to force the drive to write all cached write data to the medium.
Upon completion of a Synchronize Cache command, all data received from previous write commands will have
been written to the medium.
Table 9.3.2.1 shows the mode default settings for the drive.
4.5.2Prefetch operation
If the Prefetch fea ture i s ena bl ed, data in co nti guo us lo gic al b lock s on t he d is c immediately beyond tha t whi c h
was requested by a Read c ommand are retrieved and stored in the buffer for immediate transfer from the
buffer to the host on subseque nt Read com mands that request t hose logica l block s (this is tr ue even if cache
operation is disabled). Though the prefetch operation uses the buffer as a cache, finding the requested data in
the buffer is a prefetch hit, not a cache operation hit.
To enable Prefetch, use Mod e Select page 08h, byte 12, bit 5 (Di sable Read Ahead - DRA bit). DRA bit = 0
enables prefetch.
-
Since data that is pr efetch ed r eplaces data alr eady in som e b uffer segmen ts, the ho st can limit the a mount o f
prefetch data to optimize system per formance. The Max Prefetch field (bytes 8 and 9) limits the amount of
prefetch. The drive does not use the Prefetch Ceiling field (bytes 10 and 11).
14Cheetah 15K.4 SAS Product Manual, Rev. B
During a prefetch operation, the drive cros ses a cylinder boundary to fetch more data only if Mode paramete rs
page 08h, byte 2, bit 4 is set to 1 (Discontinuity--DISC bit).
When prefetch (read look -ahead) is enabled (enabled by DRA = 0), it opera tes under the control of ARLA
(Adaptive Read Look-Ahead). If the host uses software interleave, ARLA enables prefetch of contiguous
blocks from the disc when it senses that a prefetch hit will likely occur, even if two consecutive read operations
were not for physically co ntiguous blocks of data (e.g. “software interleave ”). ARLA disables pr efetch when it
decides that a prefetch hit will not likely occur. If the host is not using software interleave, and if two sequential
read operations are no t for c ontigu ous bloc ks of data, ARLA d isables prefetc h, but as long as se quentia l read
operations request contiguous blocks of data, ARLA keeps prefetch enabled.
4.5.3Optimizing cache performance for desktop and server applications
Desktop and server applications require different drive caching operations for optimal performance. This
means it is difficult to provide a single configuration that meets both of these needs. In a desktop environment,
you want to con figure the cache to respond q uickly to repetitive ac cesses of multiple s mall segmen ts of data
without taking the time to “look ahea d” to the next conti guous segmen ts of data. In a server environ ment, you
want to configure the cac he to provide large volumes of seque ntial data in a non-repetitive manner. In this
case, the ability of the cache to “look ahead” to the next contiguous segmen ts of sequential data is a good
thing.
The Performance Mode (PM) bit contr ols the way the drive switches the cache buffer into different modes o f
segmentation. In “server mode” (PM bit = 0), the drive can dynamically cha nge the number of cache buffer
segments as needed to opt imize the performa nce, based on the command st ream from the hos t. In “desktop
mode” (PM bit = 1), the number of segments is maintained at the value defined in Mode Page 8, Byte 13, at all
times (unless changed by using a Mode Select command). For additional information about the PM bit, refer to
the Unit Attention Pa rameters page (00h) of the Mod e Sense command (1Ah) in the SAS Interface Manual,
part number 100293071.
Cheetah 15K.4 SAS Product Manual, Rev. B15
16Cheetah 15K.4 SAS Product Manual, Rev. B
5.0Reliability specifications
The following reli ability specifica tions assume co rrect host and drive operational interfa ce, including al l interface timings, power supply voltages, environmental requirements and drive mounting constraints.
Seek error rate:Less than 10 errors in 108 seeks
Read Error Rates
Recovered DataLess than 10 errors in 1012 bits transferred (OEM default settings)
Unrecovered DataLess than 1 sector in 1015 bits transferred
Miscorrected DataLess than 1 sector in 1021 bits transferred
Interface error rate:Less than 1 error in 1012 bits transferred
Annualized Failure Rate (AFR):0.62% (see section 5.2.1)
Service Life5 years
Preventive maintenance:None required
1.Error rate specified with automatic retries and data correction with ECC enabled and all flaws reallocated.
5.1Error rates
The error rates stated in this manual assume the following:
• The drive is operated in accordance with this manual using DC p ower as defined in paragraph 6.2, "DC
power requirements."
• Errors caused by host system failures are excluded from error rate computations.
• Assume random data.
• Default OEM error recovery settings are applied. This includes AWRE, ARR E, full read retries, full write
retries and full retry time.
1
5.1.1Recoverable Errors
Recovereable errors are those detected and corrected by the drive, and do not require user intervention.
Recoverable Data errors will use co rrection, although ECC on -the-fly is not c onsidered for purposes of recov-
ered error specifica tio ns.
Recovered Data error rate is determine d using read bits transfer red for recove rable errors occurring dur ing a
read, and using write bits transferred for recoverable errors occurring during a write.
5.1.2Unrecoverable Erro rs
An unrecoverable data erro r is defined as a fail ure of the drive to rec over data from the me dia. These errors
occur due to head/medi a or write problems . Unrecoverable data errors are only detected du ring read opera
tions, but not caused by the r ead. If an unrec overable data erro r is detected, a ME DIUM ERROR (03 h) in the
Sense Key will be reported. Multiple unrecoverable data errors resulting from the same cause are treated are 1
error.
Cheetah 15K.4 SAS Product Manual, Rev. B17
-
5.1.3Seek errors
A seek error is defined as a fail ur e of the dr i ve to pos it ion the heads to the addressed track. After detecting an
initial seek error, the drive automatically performs an error recovery process. If the error recovery process fails,
a seek positioning error (Error code = 15h or 02h) will be reporte d with a Hardware error (04h) in t he Sense
Key. Recoverab le seek er rors are sp ecifi ed at Less than 10 erro rs in 10
8
seeks. Unrecoverable seek errors
(Sense Key = 04h) are classified as drive failures.
5.1.4Interface errors
An interface error is defin ed as a failure of the receiver on a port to recover the d ata as transmitted by the
device port connected to the receiver. The error may be detected as a running disparity error, illegal code, loss
of word sync, or CRC error.
5.2Reliability and service
You can enhance the r eliability of Cheetah disc drives by ensuring that the drive recei ves adequate cooling.
Section 6.0 provides te mperature mea surements and othe r informatio n that may be used to en hance the service life of the drive. Section 8.2 provides recommended air-flow information.
5.2.1Annualized Failure Rate (AFR)
The production disc dr ive shall achieve an AFR of 0.62% when oper ated in an environmen t that ensures the
HDA case temperatur es spec ified in Sectio n 6. 4 are not exce eded. S hort -term ex cursi ons u p to th e spec ification limits of the operating e nvironment will not affect AFR perfor mance. Continual or sustained operation at
case temperatures above the values shown in Section 6.4.1 may degrade product reliability.
Estimated power-on operation hours means power-up hours per disc drive times the total number of disc
drives in service. Ea ch dis c drive shall h ave accu mulate d at le ast nine months of operatio n. Data sha ll be ca lculated on a rolling average base for a minimum period of six months.
AFR is based on the following assumption s:
• 8,760 power-on hours per year.
• 250 average on/off cycles per year.
• Operations at nominal voltages.
• Systems will provide adequa te cooling to ensure the case temper atures specified in Section 6.4.1 are not
exceeded.
Drive failure means any stoppage or substandard performance caused by drive malfunction.
A S.M.A.R.T. predictive fail ure indic ates that the dri ve is deter iorating to an immin ent failu re and is co nside red
an AFR hit.
5.2.2Preventive maintenance
No routine scheduled preventive maintenance is required.
5.2.3Hot plugging the drive
When a disc is powered on by switchin g the powe r or hot plugged, the drive run s a self test befor e attemp ting
to communicate o n its’ interfaces. When the self test co mpletes successfull y, the drive initiates a Link R eset
starting with OOB. An attached devic e should respond to the link rese t. If the link reset attempt fails, or any
time the drive looses sync, the drive initiated l ink reset. The drive will i nitiate link reset once p er second but
alternates between por t A and B. Therefor e each port will atte mpt a link reset on ce per 2 seconds assuming
both ports are out of sync..
If the self-test fails, the does not respond to link reset on the failing port.
18Cheetah 15K.4 SAS Product Manual, Rev. B
Note.It is the respon si bi li ty o f the sy s tems in tegrator to assure that no tem per atu re, energy, voltage haz-
ard, or ESD potential haza rd is presented dur ing the hot connect/di sconnect oper ation. Discharge
the static electricity from the drive carrier prior to inserting it into the system.
Caution.The drive motor must come to a complete stop prior to changing the plane of operation. This time is
required to insure data integrity.
5.2.4S.M.A.R.T.
S.M.A.R.T. is an acronym for Se lf-Moni toring Analy sis and Repor ting Technology. This technology is intende d
to recognize conditio ns that indicate imminent drive failure and is desig ned to provide sufficient warn ing of a
failure to allow you to back up the data before an actual failure occurs.
Note.The drive’s firmware monitors specific attributes for degradation over time but can’t predict instanta-
neous drive failures.
Each monitored a ttrib ute has b een sele cted to moni tor a specifi c se t of fai lure c onditi ons in th e oper ating per formance of the drive and the thresholds are optimized to minimize “false” and “failed” predictions.
Controlling S.M.A.R.T.
The operating mode of S.M.A.R.T. is controlled by the DEXCPT and PERF bits on the Informational Exceptions
Control mode page (1Ch). Use th e DEXCPT bit to ena ble or disab le the S.M.A.R. T. feature. Se tting the DE X
CPT bit disables all S.M.A.R.T. functions. When enabled, S.M.A.R.T. collects on-line data as the drive performs
normal read and write operations. When the PERF bit is set, the drive is consi dered to be in “On-line Mod e
Only” and will not perform off-line functions.
-
You can measur e off-line attributes a nd force the dri ve to save the data by using the Rez ero Unit command .
Forcing S.M.A.R.T. resets the timer so that the next scheduled interrupt is in two hours.
You can interrogate the drive through the host to determine the time remaining before the next scheduled measurement and data logging process occurs. To accomplish this, issue a Log Sense command to log page 0x3E.
This allows you to c ontrol when S.M.A.R.T. interruptions occ ur. Forcing S.M.A.R.T. with the RTZ command
resets the timer.
Performance impact
S.M.A.R.T. attribute data is saved to the disc so that the events that caused a predi ctive failure can be recreated. The drive measures and saves paramete rs once every two hour s subject to an idle pe riod on the drive
interfaces. The pro cess o f meas uring off-line attribute data a nd sa ving data to th e disc is u ninterr uptable. Th e
maximum on-line only processing delay is summarized below:
Maximum processing delay
S.M.A.R.T. delay times
On-line only delay
DEXCPT = 0, PERF = 1
50 milliseconds
Fully-enabled delay
DEXCPT = 0, PERF = 0
300 milliseconds
Cheetah 15K.4 SAS Product Manual, Rev. B19
Loading...
+ 65 hidden pages
You need points to download manuals.
1 point = 1 manual.
You can buy points or you can get point for every manual you upload.