Seagate (ST16000NM002G) - Инструкция по эксплуатации - Страница 38

Seagate Exos X16 SAS Product Manual, Rev. K

37 

www.seagate.com

Defect and error management

9.0

Defect and error management 

Seagate continues to use innovative technologies to manage defects and errors. These technologies are designed to increase data integrity, 
perform drive self-maintenance, and validate proper drive operation.

SC SI defec t and er ror management involves drive internal defec t/error management and SAS system er ror considerations (errors in

communications between the initiator and the drive). In addition, Seagate provides the following technologies used to increase data integrity and

drive reliability:
• Deferred Auto-Reallocation (see Section 

Figure 9.4

)

• Idle Read After Write (see Section 

Figure 9.5

)

The read error rates and specified storage capacities are not dependent on host (initiator) defect management routines.

9.1

Drive internal defects/errors 

During the initial drive format operation at the factory, media defects are identified, tagged as being unusable, and their locations recorded on

the drive primar y defects list (referred to as the “P’ list and also as the ETF defect list). At factor y format time, these known defects are also

reallocated, that is, reassigned to a new place on the medium and the location listed in the defects reallocation table. The “P” list is not altered

after factory formatting. Locations of defects found and reallocated during error recovery procedures after drive shipment are listed in the “G” list 
(defects growth list). The “P” and “G” lists may be referenced by the initiator using the Read Defect Data command.

Details of the SCSI commands supported by the drive are described in the 

SAS Inter face Manual

. Also, more information on the drive Error 

Recovery philosophy is presented in the 

SAS Interface Manual

.

9.2

Drive error recovery procedures

When an error occurs during drive operation, the drive, if programmed to do so, performs error recovery procedures to attempt to recover the

data. The error recovery procedures used depend on the options previously set in the Error Recovery Parameters mode page. Error recovery and

defect management may involve using several SCSI commands described in the 

SAS Interface Manual

. The drive implements selectable error 

recovery time limits required in video applications.

The error recovery scheme supported by the drive provides a way to control the total error recovery time for the entire command in addition to

controlling the recovery level for a single LBA. The total amount of time spent in error recovery for a command can be limited using the Recovery

Time Limit bytes in the Error Recovery mode page. The total amount of time spent in error recovery for a single LBA can be limited using the Read

Retry Count or Write Retry Count bytes in the Error Recovery mode page. 

The drive firmware error recovery algorithms consist of 12 levels for read recoveries and five levels for write. Each level may consist of multiple

steps, where a step is defined as a recovery function involving a single re-read or re-write attempt. The maximum level used by the drive in LBA

recovery is determined by the read and write retry counts.

Table 

Table 4

 equates the read and write retry count with the maximum possible recovery time for read and write recovery of individual LBAs. The

times given do not include time taken to perform reallocations. Reallocations are performed when the ARRE bit (for reads) or AWRE bit (for writes) 
is one, the RC bit is zero, and the recovery time limit for the command has not yet been met. Time needed to perform reallocation is not counted

against the recovery time limit.

When the RC bit is one, reallocations are disabled even if the ARRE or AWRE bits are one. The drive will still perform data recovery actions within

the limits defined by the Read Retr y Count, Write Retry Count, and Recovery Time Limit parameters. However, the drive does not report any

unrecovered errors.