Method of compensating for repeatable runout error
AssigneeHGST Netherlands B.V.
InventorsJonathan D. Coker, Jeffrey L. Furlong, David R. Hall, David J. Stanek
Life of this patent
Estimating the term…
Abstract
A method of correcting repeatable run out (RRO) errors for a HDD in which RRO data is instead stored in consolidated form within the reserved area of the disk, instead of storing RRO data in the servo patterns for each HDD track or sector. RRO data is preferably stored in the reserved area of a hard disk drive in compressed form. The compressed RRO data is subsequently read into DRAM in compressed form and then decompressed for use. Predictive techniques determine what compressed RRO data is needed for upcoming read/write operations.
Claims25
A method of correcting repeatable run out (RRO) errors in a rotating data-storage device having a magnetic read/write head and a magnetic disk with a plurality of data tracks and configured for shingled magnetic recording, the method comprising:
- obtaining a value for a position error signal indicating the relative difference in position between the ideal position of the magnetic read/write head relative to a data track and the actual position of the magnetic head relative to the data track;
- calculating an RRO offset value to correct positional error of the magnetic read/write head relative to a plurality of data tracks;
- storing the RRO offset values for said plurality of data tracks as RRO data in a reserved area on the magnetic disk, said reserved area being a data storage area on the magnetic disk that is unavailable to software other than host BIOS code;
- predicting the next track to be written based on the sequential nature of the shingled magnetic recording process;
- reading into dynamic random access memory (DRAM) from the reserved area on the magnetic disk the RRO data for the next track to be written;
- accessing the RRO data from DRAM to correct the magnetic head position relative to the data track, the RRO data containing the RRO data for the next track to be written;
- and
- writing the next track using the RRO data loaded into DRAM.
A method of correcting repeatable run out (RRO) errors in a rotating data-storage device having a magnetic read/write head and a magnetic disk with a plurality of data tracks, the method comprising:
- obtaining a value for a position error signal indicating the relative difference in position between the ideal position of the magnetic read/write head relative to a data track and the actual position of the magnetic head relative to the data track;
- calculating an RRO offset value to correct positional error of the magnetic read/write head relative to a plurality of data tracks;
- and
- storing the RRO offset values for said plurality of data tracks as RRO data in a reserved area on the magnetic disk in a consolidated and compressed form, said reserved area being a data storage area on the magnetic disk that is unavailable to software other than host BIOS code;
- reading into dynamic random access memory (DRAM) from the reserved area on the magnetic disk compressed RRO data which has been determined based on predictive techniques of what RRO data is needed for upcoming read/write operations;
- decompressing the RRO data in the dynamic random access memory (DRAM);
- accessing the RRO data from DRAM to correct the magnetic head position relative to the data track;
- and
- performing the predicted read/write operations using the RRO data.
Dependencies are inferred from the claim text.