Error detection/correction and fault detection/recovery – Data processing system error or fault handling – Reliability and availability
Reexamination Certificate
1999-12-23
2004-02-24
Beausoliel, Robert (Department: 2184)
Error detection/correction and fault detection/recovery
Data processing system error or fault handling
Reliability and availability
C714S005110, C714S047300
Reexamination Certificate
active
06697958
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a technique for driving a recording medium; such as continuous real-time data such as moving image data; and the medium itself.
2. Description of the Related Art
Hard disk drives (HDDs) are widely used as magnetic disk devices. With advancements in magnetic head technology and signal processing technology, the surface storage density has increased an average of about 60% per year since 1990. It is expected that a single disk with a diameter of 3.5 inches will have a storage capacity of 10 to 20 GB by the year 2000 or soon thereafter and thus a single HDD including a plurality of disks will have a storage capacity of 100 GB or more. Such an increase in storage capacity in conjunction with high efficiency digital moving picture compression technology such as DV (Digital Video) technology (transmission rate=29 Mbps) or MPEG-2 (Moving Picture Experts Group phase-2) technology (transmission rate=15 Mbps) will make it possible to concurrently record or reproduce plural channels of moving image information on an HDD. That is, a practical multi-channel moving image disk recorder will be provided.
Historically, the HDD has been developed as a primary storage device for use in a computer, and thus the technology of the HDD has been developed so as to randomly access discrete text data in a highly reliable fashion as quickly as possible. This means that operations are performed discretely with respect to time. That is, commands such as a record command or a reproduce command supplied from a host device to the HDD are separately and discretely executed one by one. In order to assure that a recording or reproducing operation is completed within a predetermined period of time (that is, in order to assure real-time operation), the HDD must be designed such that it completes its recording or reproducing processing within the predetermined period of time.
One factor which can prevent the assurance of real-time operation is a retry of a data recording/reproducing operation, that is, the same operation is re-executed if an error is detected. Typically, the recording/reproducing of data is performed in units of small data sectors (each data sector consists of, for example, 512 bytes) corresponding to the packets that are used in communication technology. When the HDD head is moved to a track for recording or reproducing, but that track cannot be found (that is, if a seek error occurs), the moving of the head to that data sector is retried (a seek retry is performed). This is because if a seek error occurs, data cannot be recorded or reproduced correctly for all of the data sectors of interest and, thus, a seek error can cause a large burst error which cannot be corrected by means of an error correction code (ECC). Hence, it is desirable to perform a seek retry. However, each seek retry needs several to several tens of msec, and thus it becomes impossible to assure real-time operation.
If, during a data recording operation, the position of the head deviates from a data track because of mechanical shock or the like, due to external influences, from the outside, the data recording operation is retried (that is, a write retry is performed) after the head has returned to the correct position on the track and the sector of interest once again rotates into position for the recording operation. Therefore, if a write retry occurs, a break occurs in the otherwise continuous data recording operation, and a disk rotation waiting time (for example, 11 msec when the rotation speed is 90 Hz) is required. This too makes it difficult to assure real-time operation.
If a larger amount of errors occurs when data is reproduced from a data sector, and if it is determined that the amount of errors is too large to be corrected by ECC correction, the data reproducing operation is re-executed (that is, a read retry is performed) after the sector once again rotates into position to be read. Thus, a break occurs in the otherwise continuous data reproducing operation, and a disk rotation waiting time is required, thereby making it difficult to assure real-time operation. Furthermore, if an error which cannot be corrected occurs in a first read retry, a second read retry is executed; and if there are hard errors due to noise or other random factors, a large number of read retries can fail. For example, if ten or more successive read retries fail, 100 msec or more are needed for read retries until data is correctly read.
Usually, a maximum allowable number of retries for various operations is designed into the HDD. Retries are fully enabled or disabled according to, for example, the ATA (AT-Attachment) interface standard recommended by the ANSI (American National Standard Institute). However, if retries are fully enabled, the amount of non-useful write or read time increases and it becomes difficult to assure real-time operation. Conversely, if retries are fully disabled, a non-correctable burst error owing to a seek error may arise, and thus significant degradation occurs in the quality of the reproduced data.
Therefore, it is an object of the present invention to provide a technique of executing retries of various operations within a period of time which allows the HDD to continuously record and reproduce data, such as a moving image, in real time while maintaining the data quality and reliability of the HDD.
SUMMARY OF THE INVENTION
According to an aspect of the present invention, there is provided a technique for driving a recording medium in which it is the control signal that is output from the information processing apparatus with which the present invention is used that functions to control the re-execution of a read or write operation.
The information processing apparatus manages the delay time caused by the re-execution (or retry) which is performed when an error occurs in seeking, recording, or reproducing data; and generates the re-execution control signal in accordance with that delay time.
If an error occurs in a seeking, recording, or reproducing operation, the failed operation is re-executed in accordance with the re-execution control signal which, in turn, is related to the delay time caused by such re-execution.
REFERENCES:
patent: 5388254 (1995-02-01), Betz et al.
patent: 5544339 (1996-08-01), Baba
patent: 5634037 (1997-05-01), Sasaki et al.
patent: 5721816 (1998-02-01), Kusbel et al.
patent: 5812752 (1998-09-01), Yamada
patent: 5872905 (1999-02-01), Ono et al.
patent: 6125459 (2000-09-01), Andoh et al.
patent: 2002/0010833 (2002-01-01), Yoshihiro
Kanno Hiroshi
Yada Hiroaki
Frommer William S.
Frommer & Lawrence & Haug LLP
Savit Glenn F.
LandOfFree
Method and apparatus for driving a recording medium, method... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for driving a recording medium, method..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for driving a recording medium, method... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3345917