Abstract: According to one embodiment, a magnetic disk device includes a disk having a first region in which a plurality of tracks is written and a second region that is positioned with a gap in a first direction of the radial direction of the first region, and in which a plurality of tracks is overwritten in the first direction, a head, and a controller that offsets part of a plurality of tracks which is overwritten in the second region in a second direction opposite the first direction to perform rewriting.

Abstract: The present invention discloses an erasure code calculation method, including the following steps: S1) splitting original data, and building an original encoding matrix M; S2) acquiring a transverse exclusive OR encoding matrix M1; S3) acquiring a longitudinal exclusive OR encoding matrix M2; S4) acquiring an exclusive OR encoding matrix M3 according to the transverse exclusive OR encoding matrix M1 and the longitudinal exclusive OR encoding matrix M2; S5) transforming a data position of the transverse exclusive OR encoding matrix M1 to acquire a storage matrix M4; S6) judging whether storage nodes at which the last column of data of the storage matrix M4 is stored are damaged; S7) restoring the lost data according to a position 1 of the damaged node; and S8) restoring the lost data according to a position 2 of the damaged node. In the present invention, the operation is rapid, and calculation efficiency is high.

Abstract: An indication that a selected data set stored on a selected tape storage media is associated with a failure is received. An error correction data set group for the selected data set is identified, and wherein the error correction data set group includes a plurality of data sets and each data set included in the error correction data set group is stored on a different tape storage media. One or more data sets other than the selected data set that are included in the error correction data set group are selectively obtained from one or more corresponding tape storage media other than the selected tap storage media without reading entire contents of the one or more corresponding tape storage media. The obtained one or more data sets are utilized to recover at least a portion of the selected data set associated with the failure.

Abstract: A data storage device is disclosed comprising at least one head configured to access a magnetic tape. Data is read from the magnetic tape to generate a read signal which is processed to decode a first M blocks of low density parity check (LDPC) type codewords using a LDPC type decoder. First un-converged codewords out of the first M blocks are decoded using a first M-blocks parity, and second un-converged codewords out of the first M blocks are decoded using an erasure code.

Abstract: A storage system communicatively coupled to a database management system (DBMS performs storage-side scanning of data sources that are not stored in native database storage format of the DBMS. Data sources for external tables are accessible in a storage system referred to as a distributed data access system (DDAS), e.g. a Hadoop Distributed File System. To execute a query that references an external table, a DBMS first generates an execution plan. The DDAS supplies the DBMS with information that specifies each portion of the data source, and specifies which data node to use to access the portion. The DBMS sends a request for each portion to the respective data node, requesting that the data node generate rows from data in the portion. The request may specify scanning criteria, specifying one or more columns to project and/or filter on, and code modules for the data node to execute to generate records.

Abstract: In one embodiment, a method includes receiving data to store to a magnetic tape medium using a tape drive, the data being organized into a predetermined number of two-dimensional arrays, each two-dimensional array including a plurality of narrow-spread (NS) codewords positioned orthogonally to a plurality of wide-spread (WS) codewords (relative spread referring to space occupied on the magnetic tape medium when written). The method also includes successively writing each of the NS codewords onto a plurality of tracks of the magnetic tape medium using a first number of channels of the tape drive capable of writing data to tape tracks simultaneously. A first NS codeword from a first two-dimensional array is completely written to the magnetic tape medium prior to starting to write a second NS codeword from the first two-dimensional array. Also, the plurality of NS codewords are protected with a stronger encoding than the plurality of WS codewords.

Abstract: In one embodiment, a method includes reading a plurality of narrow-spread (NS) codewords from M tracks of a magnetic tape medium using a plurality of read elements, the plurality of NS codewords collectively comprising data logically organized as a predetermined number of two-dimensional arrays. Each two-dimensional array includes a predetermined number of NS codewords positioned orthogonally to a predetermined number of wide-spread (WS) codewords, with a first NS codeword from a first two-dimensional array being read in its entirety from the magnetic tape medium prior to reading a second NS codeword from the first two-dimensional array. The method also includes laterally decoding each NS codeword from the first two-dimensional array read from the M tracks of the magnetic tape medium in succession prior to decoding any WS codewords from the first two-dimensional array. The NS codewords are protected with a stronger encoding than the WS codewords within each two-dimensional array.

Abstract: A technique includes in a data storage device, sensing a plurality of data streams from a track of storage media as the media moves in a given direction using a plurality of read elements such that at least one of the read elements is redundant. The technique includes combining the data streams to generate a data stream indicating data read from the track.

Abstract: In a method of operating a memory device, a command and a first address from a memory controller are received. A read code word including a first set of data corresponding to the first address, a second set of data corresponding to a second address and a read parity data is read from a memory cell array of the memory device. Corrected data are generated by operating error checking and correction (ECC) using an ECC circuit based on the read cord word.

Abstract: A semiconductor memory device which performs a refresh operation. The semiconductor memory device may include an information detection unit suitable for detecting a refresh characteristic of a memory cell, a control signal generation unit suitable for generating a refresh control signal having a refresh cycle corresponding to the refresh characteristic, and a refresh driving unit suitable for driving a refresh operation on the memory cell with the refresh cycle in response to the refresh control signal.

Abstract: A non-binary layered low density parity check decoder includes a variable node processor operable to generate variable node to check node messages and to calculate perceived values based on normalized check node to variable node messages and on normalized decoder inputs, and to output normalized decoded values, and a check node processor operable to generate the check node to variable node messages based on normalized variable node to check node messages.

Abstract: In one embodiment, an apparatus includes an amplifier configured to receive an asymmetric signal. A first resistance is coupled between an input node and an output node of the amplifier. A second resistance is coupled to the input node of the amplifier. A first switch is configured to be controlled during a first interval to couple the second resistance to a positive resistance to increase a gain of the amplifier to correct the asymmetric signal. The gain is a function of the first resistance and a combination of the second resistance and the positive resistance. A second switch is configured to be controlled during a second interval to couple the second resistance to a negative resistance to decrease the gain of the amplifier to correct the asymmetric signal. The gain is a function of the first resistance and a combination of the second resistance and the negative resistance.

Abstract: In one embodiment, an apparatus includes an amplifier configured to receive an asymmetric signal. A first resistance is coupled between an input node and an output node of the amplifier. A second resistance is coupled to the input node of the amplifier. A first switch is configured to be controlled during a first interval to couple the second resistance to a positive resistance to increase a gain of the amplifier to correct the asymmetric signal. The gain is a function of the first resistance and a combination of the second resistance and the positive resistance. A second switch is configured to be controlled during a second interval to couple the second resistance to a negative resistance to decrease the gain of the amplifier to correct the asymmetric signal. The gain is a function of the first resistance and a combination of the second resistance and the negative resistance.

Abstract: A method of protecting information written to a recording medium includes magnetizing the recording medium to form a first magnetic pattern corresponding to information to be stored, and magnetizing the recording medium to form a protective magnetic pattern having a phase difference of 180° from the first magnetic pattern at a position adjacent to where the first magnetic pattern is formed, with adjacent bits of the first magnetic pattern opposite and the protective magnetic pattern opposite to each other.

Abstract: A magnetic tape device comprises a plurality of magnetic heads for recording or reproducing data to a magnetic tape including a data recording area, a tape running mechanism for making the magnetic tape run on the magnetic heads, and a controller for controlling a recording and reproducing operation of the magnetic heads for the magnetic tape while running. The controller includes a recording controller to control the recording and reproducing operation of the magnetic heads so as to record recording target data including a redundancy without recording the recording data and data which is the same as at least a part of the recording target data as redundant data, these data not being recorded in data recording areas adjacent with each other.

Abstract: A method to decode linear position (“LPOS”) information encoded in a sequential information storage medium, by detecting each of one or a plurality of sequential first LPOS servo patterns encoded in a first servo band using a first servo sensor in communication with a first servo channel. While detecting that one or a plurality of first LPOS servo patterns, the method fails to detect (n) second LPOS servo patterns encoded in a second servo band. The method determines a value for (n), and then syncs the first servo channel with the second servo channel using that value of (n).

Abstract: A magnetic disk apparatus that uses a head to write data in a plurality of unit areas arranged in an adjacent relationship to each other on a surface of a magnetic disk, and includes: a first recording section that records, in a RAM, the position of a unit area which is a target of data write; a first counting section that counts the number of times of data write operation performed for the unit area corresponding to the position recorded by the first recording section; and a second recording section that records, in a non-volatile storage medium different from the RAM, the side position which is the position of a unit area adjacent to the unit area whose number of records counted by the first counting section is more than a predetermined value.

Abstract: A hard disk includes: a plurality of servo areas which are elongated radially from a center of the hard disk to an outward thereof over tracks thereof; and a plurality of data areas which are respectively provided between the plurality of servo areas; wherein alternative operation is conducted per servo sector containing one of the plurality of servo areas and a data area provided subsequent to the one of the plurality of servo areas and having a defect created therein when a user sector containing at least a portion of the data area and functioning as an access unit has the defect therein.

Abstract: The present disclosure pertains to multiple-platter disk drive digital data storage with integrated redundancy operations for improved reliability. Within a single disk drive assembly (300), one or more individual storage platters (304) can be used to store redundant data, enabling recovery of user data in the event that another platter (302) is defective, fails or is otherwise unavailable. “On-the-fly” redundancy operations (FIGS. 6A,6B) can be made transparent to the user or host, and impose no significant speed penalty. A data path switch (26) can reconfigure mappings between logical ports and platter interfaces (210) as needed.

Abstract: A method and device of recording RRO correction information including recording first RRO correction information on a first track in one of servo frames of a recording medium and recording second RRO correction information on a second track different from the first recording track at a position different from a position of the first RRO correction information with respect to a peripheral direction.

Abstract: A random access memory (RAM) in a programmable logic device (PLD) supports error correction as well as a configurable data width. The number of bits in a user data word varies by the selected configuration of the RAM, while the number of bits in the error correction code (ECC) is unvarying, and is based on the total width of the memory. In some embodiments, separate ports are provided for the user data and the ECC data. Thus, ECC data can be written to an ECC portion of the RAM array at a given RAM address, while at the same time user data is written to or read from a configurable user data portion of the RAM array at the same RAM address. In other embodiments, a single memory access port is used for both user data and ECC data.

Abstract: A method for testing the integrity of a memory with defective sections under a plurality of operating environments includes testing the memory with defective sections under a plurality of operating environments, recording results of each operating environment test, and comparing the results of the tests. If the results of are the same, the memory with defective sections is declared to have integrity. If not, the memory with defective sections is declared to not have integrity.

Abstract: A method of controlling a drive of a recording medium according to one embodiment includes receiving data from a host, and writing the data to the recording medium. A difference in volume between the data at the time of being received from the host and the data at the time of being written to the recording medium is detected. An adjustment is made to at least one of a first transfer rate with the recording medium, and a second transfer rate with the host, based on the difference.

Abstract: A device comprises a memory array in which a plurality of codewords is stored. Each codeword comprises an error correction code and a data block that comprises a plurality of units of data. The device further comprises an error code correction module coupled to the memory array. When multiple units of data are to be read from the device for an address, a codeword stored in a location associated with the address is fetched from the memory array, the error code correction module decodes the codeword and corrects any errors in the data block for that codeword, and the multiple units of data are read from the corrected data block.

Abstract: A random access memory (RAM) in a programmable logic device (PLD) supports error correction as well as a configurable data width. The number of bits in a user data word varies by the selected configuration of the RAM, while the number of bits in the error correction code (ECC) is unvarying, and is based on the total width of the memory. In some embodiments, separate ports are provided for the user data and the ECC data. Thus, ECC data can be written to an ECC portion of the RAM array at a given RAM address, while at the same time user data is written to or read from a configurable user data portion of the RAM array at the same RAM address. In other embodiments, a single memory access port is used for both user data and ECC data.

Abstract: Essentially all of the processing parameters which control processing of a source audio signal to produce an encoded audio signal are stored in an audio processing profile. Multiple audio processing profiles are stored in a processing profile database such that specific combinations of processing parameters can be retrieved and used at a later time. Audio processing profiles are organized according to specific delivery bandwidths such that a sound engineer can quickly and efficiently encode audio signals for each of a number of distinct delivery media. Synchronized A/B switching during playback of various encoded audio signals allows the sound engineer to detect nuances in the sound characteristics of the various encoded audio signals.

Abstract: In a reproducing device (1) for the reproduction of reproduction data (WD) recorded on a magnetic tape (2), having transport means (7) for the transport of the magnetic tape (2) with a normal-play speed (VNS, VNLS2, VNLS3, VNLS5, VNLS7, VNHS), a first trick-play speed (VT1) and at least a second trick-play speed (VT2, VT3), the normal-play speed (VNS, VNLS2, VNLS3, VNLS5, VNLS7, VNHS) corresponding to a recording speed during the recording of the reproduction data (WD) on the magnetic tape (2), and having reproducing means (13) for the reproduction of normal-play reproduction data (NP1, NP2, NP3, NP4, NP5) recorded on the magnetic tape (2) during transport of the magnetic tape (2) with the normal-play speed (VN), of first trick-play reproduction data (TPD1, TPD6) recorded during transport of the magnetic tape (2) with the first trick-play speed (VT1), and of second trick-play reproduction data (TPD2, TPD3, TPD4, TPD5, TPD7) recorded during transport of the magnetic tape (2) with the second trick-play speed (V

Abstract: A method for performing full optimization of most of the files on a volume in accordance with a composed optimization plan, is performed, including the separation of less frequently accessed files from those whose number or location of clusters is being more frequently modified by user applications. Optimization does not continue indefinitely—it reaches a final state even if a small percentage of its file data is still out of place under the optimization plan and therefore not in the planned part of a plan-defined, Placed Files Area. Each time a Push of out-of-place file data is attempted, or a Pull of file data into a correspondingly planned free space within the Placed Files Area is attempted, a copy of the current volume bitmap is made in order to determine what is the largest free space currently available in the Placed Files Area. Once determined, the size of the largest free space is compared to the size of the largest out-of-place range of corresponding clusters in the Placed Files Area.

Abstract: When a DRAM enters an operation mode in which only a data storing operation is performed, a check bit for error detection and correction for plural data is generated and stored. Refresh operation is performed in a refresh cycle which is made long within an allowable range of an error occurrence by an error correcting operation using the check bit. Before the DRAM returns to the normal operation mode from the data holding operation mode, an error bit is corrected by using the data and the check bit.

Abstract: A data reader is arranged to read data comprising user data 30 and non-user data 32, 34 written across at least two channels of a data-holding medium 10, said data being arranged into a plurality of data items 26 each containing user data and non-user data, with said non-user data holding information relating to said user data, said data reader having a read head 12 for reading a respective said channel of said data-holding medium 10 to generate a data signal 14 comprising said data items, and processing circuitry 250 arranged to receive and process said data signals to identify a set CCPset1 of said data items written at the same time onto different said channels. Identifying a set of data items written at the same time gives rise to the possibility of correcting header information for the data items in a set.

Abstract: A system (100) has a storage medium (110) that contains a number of storage units (108). A multimedia title is stored in the storage medium (110) as a number of blocks, whereby some of the blocks are stored multiple times such that different storage units (108) contain a copy of that block. Retrieving blocks from the storage medium (110) for supply to a user is executed in cycles and the system has for some of the blocks the freedom to choose from which of the storage units (108) the block is retrieved in order to balance the load on the storage units (108). Furthermore, the system (100) is arranged to look ahead a number of cycles to determine what blocks in the near future are necessary and to retrieve these future blocks earlier if the storage units are not yet fully occupied. The additional freedom to schedule retrieval of a block at earlier moments in time provides a further mechanism to balance the load on the storage units.

Abstract: A collision recording apparatus records times in a memory, when a collision waveform provided accompanying a collision of a mobile unit and a waveform derived from the collision waveform reach predetermined values. The collision recording apparatus also records waveform types in association with the times, respectively.

Abstract: A single disk drive uses data mirroring techniques to store data within the drive. Each piece of data received by the disk drive for storage is written to at least two different storage locations within the drive. Preferably, each of the different storage locations occurs on a different disk surface, but within the same cylinder, of the drive. In one embodiment, each disk surface in the disk drive is divided into N distinct regions that are each 360/N degrees out of phase with one another. Each region on a disk surface is aligned with a corresponding region on each other disk surface in the disk drive. Data stored in each region on a disk surface is then mirrored within a different region on a different disk surface in the drive. Using such a technique allows a significant reduction in latency to be achieved during subsequent read operations.

Abstract: When a master medium and a slave medium are brought into close contact and a transfer magnetic field is applied thereto to perform magnetic transfer, the leakage magnetic field intensity of the transfer magnetic field is regulated to within a permissible range, making possible the transfer of an exact magnetic pattern. After a magnetic field is applied to the slave face in the track direction thereof to initially magnetize the slave medium in the track direction, the master medium and the slave medium are brought into close contact. When applying a transfer magnetic field in the track direction of the slave face to perform a magnetic transfer, the magnetic field intensity component in the direction opposite that in which the transfer magnetic field is to be applied is regulated to less than or equal to ½ the magnetic coercive force Hcs of the slave medium over the entire slave surface.

Abstract: A redundant recording disk device includes a first logical sub-disk and a second logical sub-disk. Mirrored redundant data are recorded such that a recording position on the first logical sub-disk and a recording position on the second logical sub-disk are positions on a corresponding radius and have a phase difference of 180 degrees around respective rotation axes of the disks with respect to one data. Therefore, the practical use of the disk device can be continued even when one recording surface is not usable because of a fault, and the access speed of the disk can be increased when there is no fault since the mirrored data which have been written with a predetermined phase-offset into the different recording surfaces may be read out in a parallel manner.

Abstract: A method of storing values that involves splitting each value into a n-bit value and an overflow value, and storing, in a main table, the n-bit values in order of increasing magnitude of the values. For each overflow value, the position of the smallest n-bit value is stored in an overflow table. To retrieve a value, the position of the corresponding n-bit value is compared to the positions stored in the overflow value to determine the overflow value of the n-bit value. The actual value is then obtained from the n-bit value and its overflow value.

Abstract: A method and apparatus are provided for preemptive head crash avoidance in a direct access storage device (DASD). At least two actuators are provided in the DASD. Each actuator services at least one transducer head. Monitoring for a predicted head crash is performed. Responsive to an identified predicted head crash, a transducer head for the predicted head crash is unloaded and unloading each additional transducer head being serviced by an actuator for the transducer head for the predicted head crash. A data backup process is performed. Responsive to an identified predicted head crash, data are transferred from a disk surface associated with the transducer head for the predicted head crash and transferring data from each disk surface associated with each additional transducer head being serviced by the actuator for the transducer head for the predicted head crash.

Abstract: A method and system that map the defects on the storage media of a disc drive. The method and system calculate the position of the defect and from that position determine the sector that must be flagged as unusable. The method and system also detect the distance the defect lies from an adjacent sector and flags the adjacent sector as unusable if the defect is too close because the defect may grow over time and eventually invade the adjacent sector. Also a method and system that efficiently utilize the space between all of the servo bursts by filling the space with sectors during the defect mapping process and marking the sectors with defects and those too close to the defect as unusable rather than shifting the sectors within the track area to avoid the defect.

Abstract: A method of redundancy coding of user data received from a host apparatus and storage of said coded data on a magnetic tape data storage medium comprises inputting a byte stream of user data into a buffer and assembling a plurality of data sets in the buffer; for each data set assembling a data set into a two-dimensional data array and (1103) applying a second redundancy coding algorithm (C2 parity) to the two-dimensional data set in a second dimension; applying (1105) a first redundancy coding (C1 parity) algorithm to the second redundancy coded data array in a first dimension to form a two-dimensional data frame having second and first redundancy coding in respective second and first dimensions, the two-dimensional data frame comprising a plurality of rows, each row comprising a first codeword and a plurality of columns, each column comprising a second codeword; partitioning the two-dimensional data frame into a plurality of logical track blocks (1106) each comprising a plurality of first codewords; and rec

Abstract: A method of managing information to be stored on a storage disc in a disc drive system to achieve improved system performance. Some of the data on the disc may be mirrored on the same surface of the disc to reduce latency and/or seek time. Compression techniques may be employed to limit reductions in the storage capacity of the disc. In another embodiment, data is preferentially stored on the outer tracks of the disc because system performance is greater when accessing outer tracks. Compression techniques may also be employed with this embodiment to limit reductions in the storage capacity of the disc. Compressed data stored on the disc may be transferred to another storage medium in its raw compressed form to reduce the number of instructions that would be required to transfer the data if it were decompressed before transfer.

Abstract: A method and apparatus for managing blocks of information in a disc drive storage system. One embodiment is directed to a disc drive storage system that employs a plurality of disc drives, and that mirrors blocks of information on different physical locations within at least two of the disc drives. Preference may be given in each of the mirrors to reading blocks of information that are closer to an outer edge of the disc to improve system performance. Another embodiment is directed to the handling of a hot spot by reading blocks of information within the hot spot from a mirror in the system that is not read during normal operation. A further embodiment addresses hot spots by temporarily employing another disc drive as a temporary mirror to enable parallel access to blocks of information within the hot spot.

Abstract: A method of managing information to be stored on a storage disc in a disc drive system to achieve improved system performance. Some of the data on the disc may be mirrored on the same surface of the disc to reduce latency and/or seek time. Compression techniques may be employed to limit reductions in the storage capacity of the disc. In another embodiment, data is preferentially stored on the outer tracks of the disc because system performance is greater when accessing outer tracks. Compression techniques may also be employed with this embodiment to limit reductions in the storage capacity of the disc. Compressed data stored on the disc may be transferred to another storage medium in its raw compressed form to reduce the number of instructions that would be required to transfer the data if it were decompressed before transfer.

Abstract: A plurality of digital signals having total data rates of n bit/s or less are input to a digital recording device recording digital signals having a data rate of n bit/s. The digital signals are subjected to format-conversion for the digital recording device to be recorded therein. Furthermore, signals having a data rate of n/i.sub.1 bit/s or less are subjected to format-conversion to be recorded i.sub.1 times in the digital recording device. In the case where the data rate of the digital signals to be input is n/j bit/s, the digital signals are subjected to format-conversion so that the amount of data capable of being recorded per unit time is decreased by i/j; thus, the digital signals are recorded for a long period of time under the condition that the data recording time is prolonged j times.

Abstract: A dot code consists of a group of plurality of blocks arranged in a matrix manner, and each block has markers, an array direction detecting code, a block address, and block recorded data. A dot code in which two rows and (N/2) columns of N blocks having block addresses 1 to N are arranged and which has an effective code width CEW is doubly provided, i.e., two such dot codes are recorded parallel on an information recording medium. A recording code width CRW in this case is twice the effective code width CEW.

Abstract: A method for controlling a data read/write operation of a disk drive recording device provided with re-allocated reserve sectors as a substitute for defective sectors, includes the steps of: reading data written on the reserve sectors when turning the power on and further storing the data, and disposing the data stored after being retrieved from the re-allocated reserve sectors as a substitute for defective sectors to the rear of data retrieved from a sector preceding the defective sector when accessing a random data section containing the defective sector, and further transferring the resulting data.

Abstract: A method and apparatus for locating more accurately the beginning of data cells stored on a medium with embedded servo sectors and improving the efficiency with which data can be stored on a medium as a consequence of more accurately locating the beginning of data cells stored on the medium. A counter counts the number of cycles of a clock. Initially, the counter is reset to zero upon recognition of a valid servo identification pattern (SID). When a terminal count value representative of the elapse of time until the end of a servo sector is reached, the counter is loaded with a negative value and continues to count. Upon detecting the next SID, the value of the counter should be equal to zero if an estimated time between detection of the last SID and the current SID was equal to the actual time between detection of the two SIDs. If the value of the counter is positive, then the current SID arrived under the head later than expected.

Abstract: A storage device is described for reading data from a hard disk drive wherein delays due to rotational latency are substantially reduced. The storage device includes a platter having a substantially planar annular surface with a geometrical center. On the platter are a plurality of generally arcuate concentric tracks. Each of the tracks is divided into a plurality of sectors that are radially outwardly disposed from the geometrical center of the platter and are bounded by radii extending from the geometrical center. Original data is written on at least one of the sectors. Further, a copy of the original data is maintained at a fixed azimuth angle from the original data during rotation of the platter about the center.

Abstract: A device for recording and reproducing data on a recording media includes an error detecting unit for detecting a defective sector by reading the data from the media, a data copying unit for copying the data of the defective sector to at least one temporary replacing sector, a test unit for testing whether the defective sector has a media defect, and a replacement unit for replacing the defective sector with a spare sector by copying the data of the at least one temporary replacing sector to the spare sector when the defective sector has the media defect.

Abstract: A data recording disk drive is provided with a reassignment architecture that employs both an "in-line" approach for small disk defects and an "off-site" reassignment for multiple-sector defects. As a further aspect of the invention, the defective sites that have been off-site reassigned are periodically reinspected to determine whether they in fact remain defective. If it is determined upon inspection that the defect is no longer present, the data in the off-site reassign area is returned to its original location, freeing spares for reuse. As a further aspect of the invention, reinspection is performed at disk drive spin-up or during idle periods.

Abstract: A magnetic disk drive includes redundant data written at a plurality of out of phase angular locations to reduce the latency and enhance performance during a read operation. The loss of recording capacity is reduced by increasing the data density to achieve the same soft error rate standard required for single recording. Dual recording also allows different recording codes to be used at the duplicated locations to thereby have the more highly stressed code words occur at different locations in the data to further reduce the possibility of an error. The redundant recording can be used in one portion of the media and normal recording used in another media portion to enable selection of the recording technique in accordance with the type of data being stored. The size of the normal and redundant recording portions can be controlled by the format operation and the user of the disk drive can intervene to designate the size of the redundant and normal media recording portions effected during the format operation.