Error detection/correction and fault detection/recovery – Data processing system error or fault handling – Reliability and availability
Reexamination Certificate
1999-11-10
2003-09-23
Baderman, Scott (Department: 2184)
Error detection/correction and fault detection/recovery
Data processing system error or fault handling
Reliability and availability
Reexamination Certificate
active
06625755
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates to a control technique of a storage apparatus and more particularly to such a technique which is effectively applied to retrying control for error or the like in the storage apparatus such as a magnetic disk apparatus.
Recently, a need for storing image and audio data and management data for managing each of the image and audio data in a single magnetic disk apparatus is increased due to the high-speed operation and the large capacity of the magnetic disk apparatus and the development of the multi-media technique. The magnetic disk apparatus of this kind generally adopts an interface of a SCSI (Small Computer System Interface) standard or the like to be connected to a host.
In the magnetic disk apparatus, when an access error to a storage medium occurs while reading and writing requests from the host are being performed, a retrying operation for recovery of the error is performed.
In the above SCSI standard, it is stipulated that the number of times of retrying operations can be specified for each of a reading error and a writing error. However, the stipulation is effective for the whole of the single magnetic disk apparatus and there is a technical problem that the number of times of the retrying operations cannot be specified for an area unit or data unit.
More particularly, in reading or writing of the image data or the audio data, there is a technical problem that image or audio is interrupted when the reading or writing operation is not completed within a time prescribed by the host computer, so that the reproduction quality of the audio or the image is deteriorated. Accordingly, it is general that a reading or writing command is completed within the prescribed time by reducing the number of times of retrying operations at the slight sacrifice of reliability, although reading or writing operation of the management data must be retried as much as possible so that the reliability thereof is not deteriorated.
To this end, in the conventional interface, when the image and audio data and the management data are stored in a single magnetic disk apparatus, it is necessary to issue a command for setting the number of times of retrying operations each time the area for storing each data is accessed. This becomes an overhead in case where the image and audio data and the management data are accessed alternately, so that a technical problem causing deterioration of performance arises.
Referring now to
FIG. 7
, such a technical problem in the prior art is described.
FIG. 7
schematically illustrates the retrying processing in a magnetic disk apparatus of a conventional reference technique.
In
FIG. 7
, numeral
310
denotes a host computer,
320
a magnetic disk apparatus,
321
a magnetic disk controller having the magnetic disk combined therewith integrally for controlling the magnetic disk apparatus,
323
a magnetic disk,
322
a connection interface for connecting the magnetic disk controller to the magnetic disk, and
330
a connection interface conforming to the standard such as the SCSI for connecting the host computer to the magnetic disk apparatus.
In the retrying processing of the magnetic disk apparatus
320
of such a reference technique, the host computer
310
issues a command for setting the number of times of retrying operations to the magnetic disk apparatus
320
so as to set the number of times of retrying operations for reading and writing error. The magnetic disk apparatus
320
records the number of times of retrying operations for reading and writing error as a table for the number of times of retrying operations in the magnetic disk controller
321
and controls to perform the retrying operation with reference to the number of times of retrying operations recorded in the table in the retrying processing.
On the other hand, when data having different properties are recorded in the single magnetic disk apparatus
320
, it is desirable to perform the recording by the number of times of retrying operations coincident with the property thereof as described above. Such data include image data and audio data which are to be read and written at a high speed and management data requiring sufficient retrying operations.
In the configuration of the reference technique of
FIG. 7
, however, the host computer
310
can set the number of times of retrying operation for the whole storage area of the single magnetic disk apparatus
320
and the host computer cannot set the number of times of retrying operations for an area unit or data unit.
With respect to the retrying operation in reproduction of information, a technique disclosed in, for example, JP-A-8-95717 is known. That is, when the number of times of retry operations for reproduction of a sector requiring the retrying operation of reproduction exceeds a predetermined value upon reproduction of information, reproduced data is stored in a memory of a disk apparatus and means for reading out the reproduced data upon subsequent access is provided to thereby shorten the reproduction time. However, this art does not quite disclose that the number of times and the time of retrying operations for each of a plurality of area and for each of a plurality of kinds of data in a medium are managed individually.
SUMMARY OF THE INVENTION
It is an object of the present invention to provide a control technique of a storage apparatus capable of realizing the optimization of retrying operation for access operation by setting a retrying condition of the access operation for each of a plurality of storage areas set in a storage medium or for each of a plurality of kinds of data stored in the storage medium variously.
It is another object of the present invention to provide a control technique of a storage apparatus capable of completing a reading or writing command within a prescribed time required in data such as image data or audio data of which processing in real time is important and capable of performing retrying operation as much as possible in data such as management data of which reliability is important so that reading and writing can be made without deterioration of reliability.
According to the present invention, in a control method of a storage apparatus including a storage medium in which data transmitted to and received from a higher-rank apparatus are stored, a limit value for at least one of the number of times of retrying operations for access operation and a time required for access operation to a storage area or data is set individually for each of a plurality of storage areas set in the storage medium or for each of a plurality of kinds of data stored in the storage medium.
As described above, as the method of setting the limit value for at least one of the number of times of retrying operations for access operation and the time required for access operation for each of the plurality of storage areas or the plurality of kinds of data individually, the following two methods are considered.
First, parameters for setting the number of division for dividing an area of a storage medium, a size of the divided area and the number of times of retrying operations for an area unit thereof are added to the standard of interface for connecting the higher-rank apparatus and the storage apparatus, and the higher-rank apparatus is adapted to be able to set the parameter to any value. The storage apparatus performs the retrying operation in accordance with the limit value for the number of times of retrying operations or the time required for retrying operation set in the area unit accessed in response to a reading or writing request from the higher-rank apparatus.
Second, an access instruction such as a reading or writing instruction issued by the higher-rank apparatus has a field in which the number of times of retrying operations can be set. The storage apparatus performs the retrying operation in accordance with the limit value for the number of times of retrying operations or the time required for retrying operation designated by the reading or writing instruction when
Hirata Keiichiro
Kojima Akira
Baderman Scott
Hitachi Software Engineering Co. Ltd.
Mattingly Stanger & Malur, P.C.
LandOfFree
Storage apparatus and control method thereof does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Storage apparatus and control method thereof, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Storage apparatus and control method thereof will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3044841