Self-Monitoring and Reporting Technology
- technology designed to monitor the disk's activities and
performance, and to warn against possible data loss in case some changes
of parameters has been made.
be able to use this technology in its completeness, there has to exist the
support of the motherboard (especially in the BIOS system), the disk(EIDE
and SCSI) and the OS on your computer.
- Enabling and disabling of the SMART functions you can find and use in
the BIOS system (Advanced BIOS Features - HDD S.M.A.R.T. Capability - Enabled/Disabled).
- Most newer disks already have the SMART functions implemented in the disk's
firmware. The SMART technology differs in the amount of functions
according to particular producers. Before the SMART technology was
accepted as standard the particular SMART functions had been presented
under different names, e. g. Predictive Failure Analysis (PFA) by IBM or Drive Failure Predication (DFP)
System - Without the OS support the possible problems will not be
reported to the user. More recent Microsoft Windows systems contain this
support. For advanced monitoring there exist a great number of SW: an
interesting program called ActiveSmart you can find on www.ariolic.com/activesmart
or EZSmart by Storagesoft,
Norton System Doctor's SMART sensor by Symantec.
When an error has been indicated, it is advisable
to back up the data properly, use a proper disk diagnostics and if needed to
change the disk.
The features are set by the producer. Each
feature can have a maximal or a minimal value that may not be exceeded.
If this happens, a warning against possible problems will appear. The amount
and the type of the features depend on the producer - there are
monitored up to several dozens of features in some disks.
What can be monitored? E. g. the disk's performance, bad sectors, calibration, CRC errors, disk spin up time, distance between the head and the disk, temperature,
features of medium, heads, motor or servomechanism.
main task of the monitoring is to monitor different features of the disk
and in case of some larger deflection to predict potential problems. These
problems are presupposed to appears as a result of wear, material defect
or some other faults that can be detected partially in advance. So,
when the disk does not work for 100%, it is possible that there is
something wrong with it. The SMART technology cannot predict the errors
occured in consequence of bad handling, an error caused by the user,
HW equipment or some other errors that cannot be predicted. Practically,
the percentage of unpredictable errors is very low.
Further information you can find on the webs
of particular producers -
e. g. IBM