External RAID Controller & Subsystem Operation Manual

Infortrend
9-5
9.1.2 S.M.A.R.T. (Self-Monitoring, Analysis
and Reporting Technology )
This section provides a brief introduction to S.M.A.R.T. as one way
to predict drive failure and Infortrend’s implementations with
S.M.A.R.T. for preventing data loss caused by drive failure.
A. Introduction
Self-Monitoring, Analysis and Reporting Technology (S.M.A.R.T.) is
an emerging technology that provides near-term failure prediction
for disk drives. When S.M.A.R.T. is enabled, the drive monitors
predetermined drive attributes that are susceptible to degradation
over time.
If a failure is likely to occur, S.M.A.R.T. makes a status report
available so that the host can prompt the user to back up data on the
failing drive. However, not all failures can be predicted. S.M.A.R.T.
predictability is limited to the attributes the drive can monitor which
are selected by the device manufacturer based on the attribute’s
ability to contribute to the prediction of degrading or fault
conditions.
Although attributes are drive specific, a variety of typical
characteristics can be identified:
• head flying height
• data throughput performance
• spin-up time
• re-allocated sector count
• seek error rate
• seek time performance
• spin try recount
• drive calibration retry count
Drives with reliability prediction capability only communicate a
reliability condition as either good or failing. In a SCSI
environment, the failure decision occurs at the disk drive, and the
host notifies the user for action. The SCSI specification provides a
sense bit to be flagged if the disk drive determines that a reliability
issue exists. The system then alerts the user/system administrator.
B. Infortrend's Implementations with S.M.A.R.T.
Infortrend is using ANSI-SCSI Informational Exception Control
(IEC) document X3T10/94-190 standard.