JPH11327808A - Retry controller for external storage device controller - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a retry controller for an external storage device controller, which is suitable for shortening processing time in an environment where command execution time is severe. SOLUTION: Total time required for processing a received command is set 102 in an external storage device controller. Access time to a sector where a retry processing to a magnetic disk device is not generated and time required for the retry processing are sequentially subtracted 106 and 110 from the total which is set. It is judged 104 and 108 whether access and retry processings can be executed in subtracted time or not. When they can be executed, processing of the command is compulsorily interrupted 118. It is judged whether an automatic exchange processing is required or not as the result of the retry processing. When the automatic exchange processing is judged to be required, it is confirmed that the automatic exchange processing can be executed within the subtracted time before the automatic exchange processing is executed.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a retry control device for an external storage device controller for a magnetic disk device or the like, and more particularly to a retry control device for an external storage device controller suitable for reducing processing time in an environment where command execution time is severe. .

[0002]

2. Description of the Related Art In a conventional external storage device controller,
Using the table of retry processing time corresponding to the number of retries for reading and writing (retry processing time required for each number of retries is tabulated for each read or write), the controller monitors the time for the magnetic disk device. I was

When monitoring the time for each retry process using the retry process time table relating to read and write, it is effective for monitoring the retry process when a read error or a write error occurs. The above table alone is not sufficient for monitoring the time of command processing over sectors.

[0004]

When the controller monitors the time of the magnetic disk device by using the retry processing time table relating to read and write in the prior art, the retry when a read error or a write error occurs is performed. Although effective for monitoring the processing time, the monitoring of the command processing over a plurality of sectors does not consider the access time to the normal sector where retry processing does not occur. When the time is specified, the command may not be completed within the specified time.

That is, when a command is issued to a controller while managing command processing over a plurality of sectors by a host device with a prescribed time for the processing, the controller performs only time management using a retry processing table. The time required for the first retry processing is detected in a table, such retry processing is repeated a plurality of times, the time is detected in the table each time, and the integrated value has not yet reached the specified time of the higher-level device. However, in practice, when accessing a large number of normal sectors, the time for the access (the normal access time for each sector is very small compared to the retry processing time, but there are many normal sectors). In this case, the integrated value of the normal access time is a time that cannot be ignored.) Since the scan time is not managed by the controller), there can be when they've reached the specified time. In this case, the command does not end within the specified time.

[0006]

In order to solve the above problems, the present invention employs the following configuration.

In the external storage device controller for the magnetic disk device, the processing time for processing the command received from the host device is monitored, and the processing of the command including the retry process for the magnetic disk device is completed within a specified time. A retry control device of the external storage device controller for forcibly interrupting the processing of the command when there is no command.

In the external storage controller for the magnetic disk drive, the total time required for command processing received from the host device is set, and the access time to the sector where no retry processing is performed on the magnetic disk drive and the retry processing are performed. The required time is sequentially subtracted from the set total time, and it is determined whether the access and the retry process can be executed within the subtracted time. If the access and the retry process cannot be executed, the process of the command is forcibly performed. The retry control device of the external storage device controller to be interrupted.

In the retry control device of the external storage device controller, as a result of the retry processing, it is determined whether or not the automatic replacement processing is necessary. A retry control device for an external storage device controller that executes the automatic replacement processing after confirming that the automatic replacement processing can be executed within the specified time or within the subtraction time.

[0010]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Retry control of a magnetic disk drive according to an embodiment of the present invention will be described below with reference to FIG. FIG. 1 is a flowchart showing a flow of processing of the external storage device controller according to the embodiment of the present invention.

When a command is received from the host computer (101), the external storage device controller changes the value of the specified time specified by the host computer to the processable time T.
It is set to 1 (102), and time monitoring is performed using the processable time T1. The time monitoring of the entire command is started (103), and it is checked whether the access process to the commanded target sector is within the processable time T1 (10).
4) If it is outside the processable time (T for normal processing)
However, there is also a case where a magnetic disk device that requires an access processing time longer than the command processing time expected by the higher-level device but is slower than 1 is used.) An error is reported (118) and the command is terminated. .

If the check in the process (104) is within the processable time T1, access to the target sector is performed (1).
05), the processable time is reduced by the time required to access the target sector, and the processable time is set to T2 (1).
06). If an error occurs during this access (1
07), after confirming that the retry processing time is within the processable time T2 (108), the retry processing (109) is performed. Then, the processable time is reduced by the time required for this retry, and the processable time is set to T3 (1
10) Then, check whether the error has been recovered (1)
11). At the time of checking whether the retry processing time is within the processable time T2 of (108), if it is outside the processable time T2, an error report is made (118) and the command is terminated.

If the error has not been recovered at the time of the check in the process (111), it is confirmed whether or not the number of retries specified by the host computer has been executed (112)
If the specified number of retries has not been performed and the processing is within the processable time, the retry processing is performed again.

When the error recovery or the retry process for a specified number of times is completed, it is determined whether or not the automatic replacement process is necessary based on the type of error and other set values (11)
3). Here, the automatic replacement process refers to a process of recording the data in a replacement sector due to the number of retries reaching a specified number or the cause of a retry error when writing or reading data.

When the automatic replacement process is performed, it is checked whether the time required for the automatic replacement process is within the processable time T3, and it is determined whether the automatic replacement process is temporally possible (11).
4). If it is determined that the automatic replacement process is temporally impossible, the command is terminated after an error report (118). Conversely, if it is determined that the automatic replacement process is possible, the automatic replacement process is performed (115). Then, the processable time is reduced by the time required for the automatic replacement process, and is set as the processable time T4 (116).

Finally, it is checked whether the access to the data requested by the host computer has been completed (step 11).
7) It repeats while monitoring the time until the data access required by the host computer is completed, and terminates the time monitoring when the command ends (119).

[0017]

As described above, according to the present invention, the external storage device controller monitors the entire command processing by using the actual time, so that the read error can be reduced.
It is not necessary to monitor the time according to the type of error such as a write error or a seek error (monitoring the retry processing time for each error using a table). In addition to the retry processing time, the access time to a sector where retry processing does not occur, The command can be terminated within the time specified by the host computer with respect to the processing time of the entire command including the command.

Further, before performing the automatic replacement process, it is checked whether or not the automatic replacement process can be temporally executed, and then the automatic replacement process is performed. In this case, it is possible to avoid interruption of the automatic replacement process that occurs when the error occurs, and the risk of occurrence of a failure is reduced.

[Brief description of the drawings]

FIG. 1 is a flowchart illustrating a process flow of an external storage device controller according to an embodiment of the present invention.

[Explanation of symbols]

 101-119 Processing Flow

Claims (3)

[Claims] An external storage device controller for a magnetic disk device monitors a processing time for processing a command received from a host device, and performs processing of the command including a retry process for the magnetic disk device within a specified time. A retry control device for an external storage device controller, wherein the processing of the command is forcibly interrupted when not completed. 2. An external storage device controller for a magnetic disk device, wherein an overall time required for command processing received from a host device is set, and an access time to a sector in which no retry process is performed on the magnetic disk device and a retry process are performed. The required time is sequentially subtracted from the set total time, and it is determined whether or not the access and the retry processing can be executed within the subtracted time. If the access and the retry processing cannot be executed, the processing of the command is forcibly performed. A retry control device for an external storage device controller, wherein the retry control is interrupted. 3. The retry control device for an external storage device controller according to claim 1, wherein as a result of said retry processing, it is determined whether or not automatic replacement processing is necessary, and said automatic replacement processing is required. An external storage device controller that executes the automatic replacement process after confirming that the automatic replacement process can be performed within the specified time or within the subtraction time when it is determined that there is the automatic replacement process. Retry control device.