Systems and methods for logically organizing data for storage and recovery on a data storage medium using a multi-level format are described. Embodiments include systems and methods for protecting data stored on a data storage medium so that the data may be recovered without errors.

Upon receipt of an instruction to access a logical address of a storage medium, an information processing apparatus controls access to its corresponding physical address of the storage medium. A management unit manages mapping between a continuous series of logical addresses and discrete physical addresses skipping a predetermined number of replacement areas. A controller identifies to which physical address the received logical address is mapped, and controls access to the storage medium using the identified physical address. When a defect occurs in a storage area indicated by a physical address, the information processing apparatus remaps its corresponding logical address to a replacement area adjacent to the defective physical address.

A data storage device is disclosed comprising a non-volatile storage medium (NVSM) having a plurality of data sectors and a plurality of reserve sectors. A map-out value is generated for each of a first plurality of the data sectors based on a read latency of each of the first plurality of data sectors, and when the map-out value of a first data sector in the first plurality of data sectors exceeds a threshold, a first logical block address (LBA) is mapped from the first data sector to a first reserve sector. When the map-out value of a second data sector in the first plurality of data sectors exceeds the map-out value of the first data sector, the first LBA is mapped from the first reserve sector back to the first data sector, and a second LBA is mapped from the second data sector to the first reserve sector.

A data storage device is disclosed comprising a non-volatile storage medium (NVSM) having a plurality of data sectors and a plurality of reserve sectors. A map-out value is generated for each of a first plurality of the data sectors based on a read latency of each of the first plurality of data sectors, and when the map-out value of a first data sector in the first plurality of data sectors exceeds a threshold, a first logical block address (LBA) is mapped from the first data sector to a first reserve sector. When the map-out value of a second data sector in the first plurality of data sectors exceeds the map-out value of the first data sector, the first LBA is mapped from the first reserve sector back to the first data sector, and a second LBA is mapped from the second data sector to the first reserve sector.

A magnetic disk device includes a magnetic disk, which includes, in a track, a plurality of first sectors each recording user data and a second sector recording parity data for restoring the user data recorded in the first sectors, and a disk controller configured to read data from a plurality of sectors including the first sectors and the second sector based on an instruction from a host, and to detect whether there is an error in each of the first sectors when the user data is read from the first sectors. The disk controller continues reading the data recorded in the plurality of sectors even when an error is detected in one of the first sectors by the disk controller.

According to one embodiment, a disk device includes a disk, a head that performs data read/write processing on a recording region of the disk, a controller that performs a media scan processing for detecting the presence or absence of a defect in a sector in the recording region of the disk in track unit. When the controller performs the media scan processing on a first sector and a second sector arranged in the track, and a third sector arranged between the first sector and the second sector, the controller performs skip processing in which the controller scans the first sector and the second sector, and does not scan the third sector.

A magnetic disk device includes a magnetic disk, which includes, in a track, a plurality of first sectors each recording user data and a second sector recording parity data for restoring the user data recorded in the first sectors, and a disk controller configured to read data from a plurality of sectors including the first sectors and the second sector based on an instruction from a host, and to detect whether there is an error in each of the first sectors when the user data is read from the first sectors. The disk controller continues reading the data recorded in the plurality of sectors even when an error is detected in one of the first sectors by the disk controller.

According to one embodiment, a defect registration method includes measuring, performing first extracting, and registering. The measuring is measuring, on a unit-area basis, performance of a storage area of a magnetic disk including a plurality of unit areas. The storage area includes a redundant area of a set capacity. The first extracting includes second extracting one or more unit areas in order of increasing the performance. The registering is registering the extracted unit areas of the set capacity as defect locations.

According to one embodiment, a disk device includes a disk, a head that performs data read/write processing on a recording region of the disk, a controller that performs a media scan processing for detecting the presence or absence of a defect in a sector in the recording region of the disk in track unit. When the controller performs the media scan processing on a first sector and a second sector arranged in the track, and a third sector arranged between the first sector and the second sector, the controller performs skip processing in which the controller scans the first sector and the second sector, and does not scan the third sector.

According to one embodiment, a defect registration method includes measuring, performing first extracting, and registering. The measuring is measuring, on a unit-area basis, performance of a storage area of a magnetic disk including a plurality of unit areas. The storage area includes a redundant area of a set capacity. The first extracting includes second extracting one or more unit areas in order of increasing the performance. The registering is registering the extracted unit areas of the set capacity as defect locations.