International Business Machines Corporation

John E Bish

Arturo Avila Mojica

Glen Alan Jaquette

David E Call

Judson A McDowell

Lawrence D Tipton

James Lee Hafner

Brian Gerald Goodman

Bret S Weber

David Alan Burton

Wendy Ann Belluomini

Kenneth F Day, Iii

Daniel W Fok

A data storage method comprises storing first data in at least a first data chunk, wherein the first data chunk is a logical representation of one or more sectors on at least a first disk drive in a storage system; storing first metadata, associated with the first data, in at least a first appendix, wherein the first appendix is a logical representation of a sector region on at least the first disk drive in the storage system, and wherein the first metadata comprises first atomicity metadata (AMD) and first validity metadata (VMD) associated with the first data; and storing a copy of the first VMD for the first data in at least one low latency non-volatile storage (LLNVS) device.

Data integrity validation in storage systems

Provided is a method, system, and program for initializing a storage device comprising multiple storage units through a storage controller. A request is received to initialize a storage space that spans the storage units. A determination is made of a layout of sets of sequential data blocks of the storage space on the storage units, wherein sets of sequential data blocks are written across the storage units. For each set of sequential data blocks, a determination is made of a block address in each storage unit where a first data block in the set of sequential data blocks will be written. A write command is generated for each of the storage units, wherein each write command indicates the determined block address for the storage unit and a block number. The write command is transmitted with one block of initialization data to the storage unit. The storage unit receiving the write command writes the block of initialization data from the indicated block address and writes the block of initialization data to each subsequent consecutive block in the storage unit a number of times equal to the block number.

Method, system, and program for initializing a storage device comprising multiple storage units through a storage controller

The present invention provides a method and an apparatus to accurately and consistently distinguish blank sectors from written sectors on optical media despite drive and media variations. A readback envelope threshold is established each time optical media is mounted into an optical drive. In one embodiment, the gain of readback amplifiers are set to a first level, the optical head seeks to and reads a sector or set of sectors on the optical disk known to have information recorded thereon and a first readback envelope generated. The gain of the readback preamplifier is then set to a second level, the optical head seeks to and reads another sector or set of sectors known to have information recorded thereon and a second readback envelope generated. The readback threshold is calculated from the amplitudes of the first and second envelopes. Once the threshold has been calculated for a disk, the drive can scan for a blank sector by comparing the amplitude of the sector's readback envelope with the threshold: if the amplitude is less than the threshold, and no other written format feature is readily indicated the sector can be classified as being blank and it is safe to write to the sector.

Calibration of readback threshold in an optical storage device to

The present invention provides a method and an apparatus to accurately and consistently distinguish blank sectors from written sectors on optical media despite drive and media variations. A readback envelope threshold is established each time optical media is mounted into an optical drive. In one embodiment, the gain of readback amplifiers are set to a first level, the optical head seeks to and reads a sector or set of sectors on the optical disk known to have information recorded thereon and a first readback envelope generated. The gain off the readback preamplifier is then set to a second level, the optical head seeks to and reads another sector or set of sectors known to have information recorded thereon and a second readback envelope generated. The readback threshold is calculated from the amplitudes of the first and second envelopes. Once the threshold has been calculated for a disk, the drive can scan for a blank sector by comparing the amplitude of the sector's readback envelope with the threshold: if the amplitude is less than the threshold and no other written format feature is readily indicated, the sector can be classified as being blank and it is safe to write to the sector.

Calibration of readback threshold in an optical storage device

The present invention locates spare sectors for replacing secondary defects found during disk use. A sorted defective sector list is maintained for tracking the spare sectors that have been previously used for replacements of other secondary defects. When a secondary defect address is found, a secondary defect list is updated both on the disk and in the drive's memory. A spare sector, for replacing the secondary defect is located by first calculating a group number to determine which group the secondary defect is located in. Next, the boundaries for a plurality of spare sectors allocated for that group is determined. After searching the replacement sector list, the first available spare sector of the plurality of spare sectors is identified as the replacement sector. If all of the plurality of spare sectors have been previously assigned as replacement sectors, a spare sector is instead located from a plurality of spare sectors allocated to a neighboring group. After locating the spare sector, its physical address is returned and the spare sector is logged in the defective sector list and in the replacement list.

Reassigning defective sectors on a disk

An optical disk has four copies or sets of its control data structures. The control data structure includes a disk definition structure DDS, a primary defect list PDL, and a secondary defect list SDL. Procedures are described for checking to whether or not all copies of the data structures have congruence. If not, all the tables are then made congruent and re-recorded on a disk. In the procedure, the copy having the greatest number of entries is first examined for determining whether or not congruence is possible. If congruence is not possible from the longest copy, then the next longest copy is used.

Growing community of inventors

Tucson, AZ, United States of America

John E Bish