- Sector slipping
Sector slipping is a technique used to deal with defective sectors in hard disk drives. Due to the volatility of
hard disk s from their moving parts and low tolerances some sectors become defective. Defective sectors can even come on hard disks from the factory so most disks incorporate a bad-block recovery system to cope with these issues.Description
During a low-level format, defect lists are populated which store which sectors are bad. The bad sectors are then mapped and a sector slipping algorithm is utilized. When using sector slipping for bad sectors disk access is not largely affected. The drive will just skip over the sectors and takes the same time as it would have to read the sector. Spare sectors are located on the disk to aid in having sectors to “slip” other sectors down to, allowing for the preservation of sequential ordering of the data. Accuracy of programs reliant on static knowledge of cylinders and block positions will be comprised though.
Bad sectors that are found during normal usage of the disk are not capable of having the sector slipping algorithm applied. Instead a linear reallocation, or sector forwarding, is used where a bad sector is replaced with a sector from a spare area. Doing this does affect access times as the disk will need to seek to the spare sector since all further lookups of the bad sector will redirect to the new sector.
Example
Logical Sectors Physical SectorsPre Low-Level Format0 0 1 1 2 2 3 3 4 4 5 5 6 7
Logical Sectors Physical SectorsPost Low-Level Format
0 -------> 0 1 -------> 1 2 -------> 2 3 (Bad) 3 -------> 4 5 (Bad) 4 -------> 6 5 -------> 7
In this example, physical sectors 3 and 5 were found to be bad. The sectors were then slipped down to allow for the logical sectors to be placed in sequential order on good sectors. Sector 3 was slipped down to 4 and sector 4 was slipped down to 6. The rest of the sectors were slipped down to the remaining spare sectors: Sector 4 to 6 and sector 5 to 7.
References
Operating System Concepts, 7th Ed. , Silberschatz, Galvin and Gagne [http://www.ece.cmu.edu/~ganger/papers/CSE-TR-194-94.pdf Scheduling for Modern Disk Drives and Non-Random Workloads] by Bruce L Worthington, Gregory R. Ganger and Yale N. Patt
Wikimedia Foundation. 2010.
