Active@ Hard Disk Monitor

LSoft Technologies LSoft Technologies products. Twitter page LSoft Technologies products LSoft Technologies products LSoft Technologies products
Active@ Hard Disk Monitor

S.M.A.R.T. Technology

S.M.A.R.T. monitors the most important device parameters. These parameters are called attributes. The list of parameters required to monitor is determined by a manufacturer. Every such attribute has some value. Normally values are in the range from 0 to 100 (however it might be up to 255). The value shows a reliability of some attribute related to it's etalon value which determined by a manufacturer.

The high value tells about stability of the particular parameter. The low value tells about fast degrading of probability of soon failure, thus the higher the value the better. For each attribute a manufacturer defines a threshold - a minimum value at which it guarantees the problem free work of the device. If the value falls below the threshold, the failure is very likely.

Attributes may be critical and non critical. When a critical attribute's value go below its threshold it leads to device's failure. When a non critical attributes's value become lower than a threshold, it means that a problem exists but disk continue to work, though with degradation of some characteristics (for example, performance).

The most important critical attributes are:

Raw Read Error Rate - the frequency of read errors which are caused by hardware.

Spin Up Time - the time it takes for a disk to come from the idle state to the full spin speed. When a normalized value is calculated the actual value is compared to the etalon set at the factory. Even if spin up time is not maximum, it's OK as long as it keeps the value. The reason of having a spin up time different from etalon might be a power supply unit.

Spin Up Retry Count - the count of retries to spin the disks to their working speed in case when the first attempt failed. Non zero raw value (or non maximum value) tells about the problems in disk mechanics.

Seek Error Rate - the frequency of heads positioning error. High raw value tells about existing problems which might be a servo marks damage, extensive thermal disk expansion, mechanical problems in the heads block etc. The constant high value means that everything is fine.

Reallocated Sector Count - the number of sector remapping operations. The modern disks can analyze sectors stability on the fly and remap them in case of failure.

The most important non critical (informational) attributes are:

Start/Stop Count - full number of spindle's starts and stops. Every motor is guaranteed to make only a certain number of starts and stops. This value is defined as a threshold. First 7200 rpm disks had very unreliable motors and could survive only small number of switches.

Power On Hours - the number of hours spent in ON state. As a threshold a passport MBTF value is used.

Drive Power Cycle Count - the number of full turn on and off cycles. This parameter can help to determine the disk's usage before the purchase.

Temperature - the value of built-in temperature sensor. The temperature have a great impact on a disks's heath and life span even if it is in the normal range.

Current Pending Sector Count -the number of sector candidates for remapping. Those sectors are not considered bad yet but the reading them is different from the reading the stable sectors.

Uncorrectable Sector Count - the number of errors accessing to a sector which were not corrected. The possible reasons might be mechanical problems and disk surface damage.

UDMA CRC Error Rate - the number of errors happened during data transmittance over the external interface. They can be caused by low quality cables and non standard working conditions.

Write Error Rate - shows the frequency of errors during the writing to the disk. It can serve as an indicator of disk surface quality and mechanical parts of the drive.

All errors and parameter changes are stored in the S.M.A.R.T. log.

See also: