摘要:
A method for detecting the location of falsely detected 'good' data, or 'rogue', packets in a data buffer (10) is presented. A segemnt-level CRC is generated (14, 614) over, and associated with, a buffer segment (11), and recorded along with the segment data (15) onto a storage medium. During data recovey, only packets (5) that pass a packet-level error detection test are allowed in the data buffer (10). Once a data segment (11) is complete, a segment-level CRC test (14) is performed over the recovered segment-level CRC and the entire recovered segment data. The segment (11) contains a rogue packet if the segment-level CRC test fails. Reed-Solomon syndromes are generated and used to locate and optionally correct the rogue packets.
摘要:
A method for monitoring or adjusting a magnetic tape drive using central control packets (30) that are recorded on the tape (2) is presented. Central control packets (30) are used for adjusting the timing of the read head signal over the tape, adapting the reel count to that of the drive at the time the tape was recorded, and monitoring filemarks for logical tape positioning.
摘要:
A method and apparatus for varying track recording speed to maximize host-to-tape data transfer rates is presented. Variable data transfer rates of host systems and networks are accommodated by continually adjusting the tape speed to match the tape drive to the host's actual transfer rate. The speed of the tape is adjusted according to both the level of data present in the tape drive data buffer and whether the current mode of the drive is write mode or read mode. In the preferred embodiment, when the tape speed is accelerated or decelerated while writing a number of data track pairs to tape, a predetermined number of dummy track pairs are first written to tape at the current speed. On the next write head phase after the speed is increased or decreased as appropriate while the read heads are over the track, a predetermined number of dummy track pairs followed by more data track pairs are written to the tape at the new speed.
摘要:
A multi-level error correction technique for maintaining data integrity in a data recording/recovery device is presented. Data is stored in buffer packets in a segment of a data buffer before being recorded to a storage medium. Error correction codes are generated over a plurality of instances of a plurality of buffer packet patterns in the segment, and are recorded along with all of the buffer packets onto a storage medium. During data recovery from the storage medium, recovered error correction code packets are used to perform multi-level segment correction on the recovered data packets.
摘要:
A helical scan drum design for use in non-tracking tape devices which assures 70% coverage of a track to be read by overscanning with at least two read heads at approximate 1X speed. The present invention further provides a simulation method for evaluating potential drum designs for such overscan applications. The preferred drum design uses pairs of like-azimuth read heads positioned on the rotating drum such that in combination they overlap the scan of a track by 130% the track width. These dimensions assure at least 70% coverage of each track by at least one of the pair of heads at up to 1X speed while assuring no overlap with another like-azimuth recorded track. The simulation method allows for evaluation of potential drum designs by accepting parameters describing the intended drum application and then simulating track read operations over a plurality of simulated tracks to determine the efficacy of the design over a range of tape speeds and gap widths. Designs that simulate successful reading of a sufficient threshold number of tracks over a sufficiently broad range of tape speeds may then be selected for further test and evaluation.
摘要:
A method for logically rejecting previously recorded track residue from magnetic media is presented. A session ID (52) unique to a given recording session is encoded into track packet error check (16) and error correction codes (18) but is not itself actually written to tape. During a data recovery session, a reference session ID (78) for the original recording session is acquired by reconstructing the packet session ID (52) from the first few track packets and verifying that a predetermined number of consecutive track packets (8) have identical packet session IDs (52). Once the reference packet session ID (78) is acquired, it is preloaded into error detection (82) and correction (84) hardware. When a residue track encoded with a previously recorded session ID is recovered by the tape drive track packet detection circuitry, it is inherently rejected because the error detection and correction hardware detects an error and it is therefore never allowed into the data buffer.
摘要:
A helical scan drum design for use in non-tracking tape devices which assures 70% coverage of a track to be read by overscanning with at least two read heads at approximate 1X speed. The present invention further provides a simulation method for evaluating potential drum designs for such overscan applications. The preferred drum design uses pairs of like-azimuth read heads positioned on the rotating drum such that in combination they overlap the scan of a track by 130% the track width. These dimensions assure at least 70% coverage of each track by at least one of the pair of heads at up to 1X speed while assuring no overlap with another like-azimuth recorded track. The simulation method allows for evaluation of potential drum designs by accepting parameters describing the intended drum application and then simulating track read operations over a plurality of simulated tracks to determine the efficacy of the design over a range of tape speeds and gap widths. Designs that simulate successful reading of a sufficient threshold number of tracks over a sufficiently broad range of tape speeds may then be selected for further test and evaluation.
摘要:
A method and apparatus for reacquiring synchronization of a clock synchronization signal (58) with a data signal (60) in a non-tracking storage device is presented. A packet error detector (46) detects errors in data packets (55) from a data signal (60) and generates a packet error status (56) for each reconstructed packet (55). A read quality detector (42) monitors the packet error status (56) and determines whether the quality of the data signal (60) is of an acceptable versus unacceptable read quality condition. The read quality condition is based on the number of consecutive packets in which an error is detected. Upon detection of an unacceptable read quality condition, the read channel is disabled and the clock synchronize signal (58) is relocked to a known reference frequency (54). The read channel is then reenabled and the clock synchronization signal (58) is relocked to the data signal (60).