Error detection/correction and fault detection/recovery – Data processing system error or fault handling – Reliability and availability
Patent
1997-12-02
2000-01-11
Palys, Joseph E.
Error detection/correction and fault detection/recovery
Data processing system error or fault handling
Reliability and availability
714 15, 714 48, G06F 1100
Patent
active
060147547
DESCRIPTION:
BRIEF SUMMARY
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to the field of produced error processing in a magnetic disk drive. More particularly, the present invention relates to a method for realizing the optimum execution form of an error recovery procedure (ERP) and its system.
2. Background Information
A magnetic disk drive is an apparatus for recording data in accordance with magnetic changes of the surface of a disk and reading the data. A transducer head serving as a converter is positioned at a predetermined position of a track with data recorded and reads information recorded in tracks of a disk rotating at a high speed or writes information in the tracks. If any error occurs when data is read by the head, it is hitherto known that there are various error recovery procedures (ERPs) for recovering the error.
A read error occurs in a disk due to scratches, unevenness of a magnetic body, or change of the magnetic body with time which are not found while the disk is manufactured. A general error recovery code ECC is used for an error normally caused in a data part and error recovery processing. Moreover, when a read gain or an off-track is changed or an MR (magnetic-resistance) element is used as a read head, various recovery processings are executed including the change of a bias value of the MR element. When these recovery processings are executed and data is reread, the data is continuously used. However, when an error cannot be recovered by a plurality of error recovery processings, the error is classified as a hard error. Or, when it is possible to rerecord (reassign) data in other area, the data is rerecorded in other area by setting the area concerned on the disk as an unusable area.
A magnetic-resistance (MR) conversion head is used as one of the existing conversion heads. In the case of the magnetic-resistance conversion head, the output resistance of the head changes in accordance with the change of a magnetic field. The resistance change is converted to a DC voltage signal by flowing a predetermined current to an MR element and data read is executed.
However, the method for reading a resistance change includes a thermal asperity as one of the read errors. The thermal asperity is defined as a phenomenon in which a protrusion formed on a disk collides with a read head to cause a resistance change in an MR element due to a temperature change and thereby, an abnormal signal is generated.
The thermal asperity is not found in manufacturing a disk but it frequently occurs while a system is operated and measures have not completely been taken for the error in most cases.
Not only the thermal asperity but also various error factors can always be present. Therefore, it is requested to execute error recovery processing steps for solving these various errors.
In the case of error recovery processing, it is general to change or adjust standard read conditions specified between a disk, magnetic head, and HDC (hard disk controller) one by one and reread data. Read conditions include an off-track value which is a deviation of the center of a magnetic head from that of a track, a bias current value supplied to an MR element when the magnetic head uses the MR element, and the speed of a PLL circuit for stabilizing a signal amplification factor and a sampling frequency by means of auto gain control (AGC) for making the amplitude of a regenerative signal constant (tracking speed is set to a predetermined value).
When an error occurs, these conditions are changed, that is, the tracking characteristic is intentionally deteriorated, the amplification factor of the auto gain control (AGC) for signal amplification is adjusted, and data is sequentially reread. When data read is successfully completed, the error recovery processing terminates.
The execution sequence of a lot of processings included in the error recovery processing routine is fixed. That is, the execution sequence such as firstly, adjustment of AGC, secondly, adjustment of PLL, and thirdly, change of bias value of MR head is
REFERENCES:
patent: 4516165 (1985-05-01), Cunningham et al.
patent: 4701915 (1987-10-01), Kitamura et al.
patent: 4800563 (1989-01-01), Itagaki et al.
patent: 4821125 (1989-04-01), Christensen et al.
patent: 4866712 (1989-09-01), Chao
patent: 5269017 (1993-12-01), Hayden et al.
patent: 5293613 (1994-03-01), Hayden et al.
patent: 5325364 (1994-06-01), Ouchi et al.
patent: 5331476 (1994-07-01), Fry et al.
patent: 5379162 (1995-01-01), Cunningham et al.
patent: 5386324 (1995-01-01), Fry et al.
patent: 5421003 (1995-05-01), Escola et al.
"Invoking Error Recovery Procedures", IBM Technical Disclosure Bulletin, vol. 24, No. 8, Jan. 1982, pp. 4051-4052.
Bussan Matthew J.
International Business Machines - Corporation
Mai Rijue
Palys Joseph E.
LandOfFree
Signal recording and regenerating apparatus and error recovery p does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Signal recording and regenerating apparatus and error recovery p, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Signal recording and regenerating apparatus and error recovery p will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-1470809