SalvageData DiskLife S.M.A.R.T Monitoring Tools

S.M.A.R.T. – Self Monitoring Analysis and Reporting Technology

S.M.A.R.T. is designed to predict failure of a hard disk drive with adequate notification time, allowing a user to back up data preceding the drive failure. S.M.A.R.T. technology is functional for both IDE and SCSI disk drives.

Drive architectures differ from one drive to another. S.M.A.R.T. contains a different set of attributes and thresholds for each drive model. Each drive must be monitored precisely to ensure stability is maintained. The exact attributes monitored depend particularly on the drive manufacturer and model. The acceptable values are programmed into the drive. If the readings for the different attributes fall out of acceptable range, or the trend of an attribute reveals declining performance, a message is sent to the drive’s SMART status register to warn the drive a problem may be occurring.

Common predictable attributes causing failure

Head Flying Height - A descending inclination in flying height will often foreshadow a head crash.

Number of Remapped Sectors - If the drive is remapping numerous sectors due to internally-detected errors, this is a sign the drive may fail shortly.

ECC Use and Error Counts - The number of errors experienced by the drive. Even if this is corrected internally it is a signal that problems are developing. In some cases the trend is more significant than the actual count.

Spin-Up Time - Spin-up time changes can reflect problems with the spindle motor.

Temperature - Increases in drive temperature often reveal spindle motor problems.

Data Throughput - Decreases in the transfer rate can foreshadow a number of internal problems.

Spin Retry Count – Decrease in the value of this attribute often reflects a problem with the drive’s mechanical subsystem.

Seek Time – A decline in seek time is a sign the drive’s heads are beginning to malfunction.

Each drive failure is not 100% preventable. However, any gradual decline in performance or activity from the disk is, in most cases, prone to detection. Frequently these failures are associated with error rates and seek/settle times. It is common for these changes to be so subtle over a period of time a user will not even know it is occurring. Mechanical failures are mainly gradual changes and predictable, accounting for 60% of drive failures.

System Requirements

• Processor: Pentium or better or compatible processor
• Operating System: Windows 95 OSR2, 98, ME, NT, 2000, XP and 2003
• Memory/RAM: 64 MB (minimum)
• 40 MB of hard disk space (minimum)
• 64 MB of disk swap space (minimum)