JPH02259854A - Information processor - Google Patents

Abstract

PURPOSE:To improve reliability and fault resistance by monitoring the using time of a line and the counted value of the number of times of faults, deciding whether the faults more than a specified value is generated or not and managing the line control. CONSTITUTION:Fault history display parts 20 and 21, counting parts 22 and 23 of the number of times of the use of line counting parts 24 and 25 of the number of times of the use of the line are provided. Based on the information of the number of times of the use of the line and the information of the number of times of the occurrence of the fault corresponding to respective lines 31-34, the effectivity of the line is decided according to a fault generation rate in the number of times of the use of the line to be specified in advance. Further, the line of the high fault generation rate is defined as a fault line and the use of such a line is suppressed according to the fault history information. Thus, defect that the line can not be used by information transfer abnormality generated by the temporary fault is removed and the information processing system with the high reliability and the fault resistance can be constructed.

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Field of Application] The present invention is applied to an information processing device. Particularly, the present invention relates to means for controlling an information transfer path between devices connected through a plurality of paths.

〔overview〕

The present invention provides an information transfer route control means between devices connected through a redundant configuration route, in which even if the number of retry failures reaches a predetermined value, the route is not recognized as a failed route, and the retry failure is prevented. By recognizing faulty routes according to the ratio between the number of times and the number of information transfers, it is possible to avoid inhibiting usable routes as much as possible.

[Conventional technology]

In recent years, information processing systems with excellent reliability and fault tolerance have been particularly anticipated. In order to improve reliability and fault tolerance, the devices that make up the information processing system are connected through multiple routes, and even if a failure occurs in one route, the other route is used as an alternative route. Create a redundant configuration so that overall processing can continue.

Conventionally, in this type of information processing system, when a failure is detected during information transfer processing in an information processing device, retry processing is performed several times on the same route, but retry processing is performed a predetermined number of times. If the transfer is not executed normally, that route is recognized as a failed route, and the information transfer is retried via an alternative route without using that route. If information cannot be transferred via an alternative route, it is generally treated as a failure of the route or the device itself.

FIG. 2 is a block diagram showing the configuration of a conventional example. Information processing device 1 and peripheral devices 2 and 3 each have two paths 3
1 and 33 and paths 32 and 34. The information processing device 1 includes a control unit 11 that performs overall control of route selection and information transfer, a storage unit 12 that stores transfer information, and peripheral devices 2 and 3 via a route 31.33 or 32.34. Transfer control unit 1 that controls the interface of
3 and 14, and a transfer status monitoring unit 15 that monitors whether the interface control performed by the transfer control units 13 and 14 is normal. .33 or 32
．． 34, and retry counters 18 and 19 for counting the number of retries on the route 31.33 or 32.34. In such an information processing device 1, when the control section 11 transfers information in the storage section 12 to the peripheral device 2, the control section 11 first checks the states of the path activation control sections 16 and 17. The states of the route activation control units 16 and 17 are the routes 31 and 3.
3 indicate that both are valid, the control unit 11 selects a route based on a predefined procedure. When path 31 is selected, control section 11 instructs transfer control section 13 to transfer the information in storage section 12 to peripheral device 2 via path 31. The transfer status in the transfer control unit 13 is monitored by the transfer status monitoring unit 15, and the control Ill checks the transfer status monitoring unit 15 to determine whether the information transfer using the path 31 has been performed normally.

If the information on the path 31 is abnormal, the control unit 11
updates the retry counter 18 and checks whether it has reached a predefined number of times. The control unit 11 repeatedly retries the information transfer process until the retry counter 18 reaches a predetermined number of times. If the information transfer is successful during the retry,
Control all initializes the retry counter 18 and performs the next process. If normal information transfer is not performed during the retry, an instruction is issued to the route activation control unit 16 to inhibit future use of the route 31 as a failed route. Next, the control unit 11 performs the same processing on the route 33 as on the route 31. Route 3
If both 1 and 33 are inhibited from use, the peripheral device 2 itself is at fault due to a path failure, and the information transfer process is treated as abnormal.

[Problem that the invention seeks to solve]

As described above, in the conventional example, even if a path abnormality occurs due to a temporary failure, the route may be inhibited and become unusable, which has the drawback of inhibiting usable routes.

The present invention aims to improve these drawbacks, and provides an information processing device with excellent reliability and fault tolerance by providing means for more detailed route control management.

[Means for solving problems]

The present invention includes a first counting means for counting the number of retries of information transfer via a plurality of routes connected to each of the peripheral devices, and a first counting means for counting the number of times information transfer via the route is retried for each of a plurality of routes connected to each of the peripheral devices; In the information processing apparatus, the information processing apparatus includes a route activation control means for inhibiting the information transfer, and a second counting means for counting the number of times information is transferred via the route, and a second counting means for counting the number of failed retries related to the information transfer via the route. a third counting means for counting; a fault history registration means for registering the counted value of the second counting means and the counted value of the third counting means; Features: The route activation control means includes a control means for specifying a route for which information transfer is prohibited.The route activation control means counts the number of uses and the number of failures of the route used during information transfer, and calculates the number of uses and the number of failures. The count of the number of failures is monitored to determine whether failures exceeding a specified value have occurred on the route being used, and for routes with many failures, a failure history is registered to prevent use.

〔Example〕

Hereinafter, one piece of information about the present invention will be explained with reference to the drawings. FIG. 1 is a block configuration diagram showing the configuration of this embodiment.

In this embodiment, as shown in FIG.
4, where the information processing device 1 includes peripheral devices 2 and 3 connected via paths 31, 32, 33 and 34.
a control unit 11 that performs overall control of selection and information transfer; a storage unit 12 that stores transfer information; and a transfer control unit that controls interfaces with peripheral devices 2 and 3 via paths 31, 32, 33, and 34. 13 and 14, a transfer status monitoring unit 15 that monitors whether the interface control performed by the transfer control units 13 and 14 is performed normally, and a route 31. 3
2. Route activation control units 16 and 17 that indicate the validity of routes 3132.33 and 34, retry counters 18 and 19 that display the number of retries on routes 3132.33 and 34, and failure history that is a feature of the present invention. It is composed of display units 20 and 21, route usage count units 22 and 23, and route failure count units 24 and 25.

Retry counters 18 and 19 are designated as first counting means for counting the number of times information transfer via this route has been retried for each of the plurality of routes connected to each of the peripheral devices 2 and 3. The present invention is further equipped with route activation control units 16 and 17 for prohibiting information transfer via the route, and further includes a second counting unit for counting the number of times information is transferred via the route. route usage frequency counting units 22 and 23, which are the route usage frequency counting units 22 and 23, and route failure frequency counting units 24 and 25, which are third counting means that count the number of failed retries related to information transfer via the above-mentioned route.
and failure history display sections 20 and 21 including failure history registration means for registering the counts of the second counting means and the third counting means, and the values registered in the failure history registration means. A control unit 1 including a control means for specifying a route for which information transfer is prohibited to the route activation control means based on
1.

Next, the operation of this embodiment will be explained.

When the control section 11 transfers information in the storage section 12 to the peripheral device 2, the states of the path activation control sections 16 and 17 are first checked. When the states of route activation controllers 16 and 17 indicate that both routes 31 and 33 are valid,
The control unit 11 selects a route based on a predefined procedure. When path 31 is selected, control section 11 instructs transfer control section 13 to transfer the information in storage section 12 to peripheral device 2 via path 31. The transfer status in the transfer control unit 13 is monitored by the transfer status monitoring unit 15,
The control unit 11 checks the transfer status monitoring unit 15 to determine whether information transfer using the route 31 has been performed normally.

If the information transfer on the route 31 is abnormal, the control unit 11 updates the retry counter 18 corresponding to the route 31 and checks whether the retry counter 18 has reached a predetermined number of times. Control part 1
1 retries such information transfer processing until the retry counter 18 reaches a specified number of times. If the information transfer is successful or if the information transfer is successful after a retry, the retry counter 18 is initialized, the route use count counter 22 counts, and then the next process is continued. . If normal information transfer is not performed during retry, conventionally route 3 is used.
However, in this information processing device 1, the control unit 11 counts a counter indicating that the retry has failed in the route failure count unit 24, and prevents the route 31 from being used again. is not carried out. Next, the control unit 11 performs information transfer processing on the alternative route 33 in the same manner as the processing on the route 31. When the information transfer on the route 33 is performed normally, the control unit 11 determines that there is a possibility of a failure on the route 31,
The failure of the route 31 is registered in the failure history display section 20. In this state, the control unit 11 does not inhibit the use of the route 31 and continues the valid state. When the control unit 11 selects the route 31 again to transfer information in another information transfer process, if a transfer error occurs and does not recover even after retrying the route 31, the route usage count unit 22 The count is updated and the count is updated in the route failure frequency counting unit 24. However, if a failure of the route 31 is registered in the failure history display unit 20, the usage frequency information and the route failure frequency in the route usage frequency counting unit 22 are updated. The control unit 11 refers to the failure frequency information of the counting unit 24 and investigates the frequency of failure occurrence with respect to the number of uses.

In the case of the information processing apparatus of this embodiment, if a failure occurs eight times out of 32 times of use, it is determined that the path is abnormal. When the control unit 11 determines that the route 31 is a failed route, it registers route unusable information for the route 31 in the failure history display unit 20. When the control unit 11 detects a route abnormality again on the route 31 for which this route unusable information is registered, it instructs the route activation control unit 16 to inhibit the use of the route 31. The processing of this control unit 11 is carried out by another route 32.3.
3 and 34, but if the two routes 31 and 33 to the peripheral device 2 are inhibited, the transfer of the peripheral device 2 becomes impossible, so the first route 31 is inhibited from being used. However, the use of route 33 is not inhibited.

If a failure occurs 8 or more times out of 32 times the route 33 has been used, the failure of the route 33 is registered in the failure history display section 20. Route usage count units 22 and 23 have a count value of “0”
The count value is incremented by "1", and when the count value exceeds "32", the count value returns to "0" again. The route failure display information in the failure history display units 20 and 21 is set and canceled by the control unit 11, and the setting is performed when a failure occurs on one route and the transfer process is successfully completed on the alternative route. Further, the cancellation is performed when the count values of the route use count units 22 and 23 reach “16” and the count values of the route failure count units 24 and 25 are “3” or less. The failure count information of the route failure count units 24 and 25 is incremented by 1 by the control unit 11 when the retry process on that route becomes abnormal. When the count value of the route usage count units 22 and 23 reaches “32”, the control unit 11 refers to the count value of the route failure count units 24 and 25 to determine the frequency of failure occurrence. When the determination of the frequency of failure occurrence is completed, the failure frequency information of the route failure frequency counting units 24 and 25 is initialized.

〔Effect of the invention〕

As explained above, the present invention determines the effectiveness of a route based on the failure occurrence rate within a predetermined number of route uses based on route use number information and failure occurrence number information corresponding to each route, Furthermore, since the use of routes with a high frequency of failures as failure routes is inhibited based on failure history information, the drawback that routes become unusable due to information transfer abnormalities that occur due to temporary failures is eliminated, resulting in high reliability and fault tolerance. It has the effect of building an information processing system.

... Retry counter, 20.21 ... Failure history display section, 22.23 ... Route usage count section, 24.25
...Route failure count counter, 31.32.33.34.
··route.

Claims (1)

[Claims] 1. A first counting means for counting the number of times information transfer is retried via each of a plurality of routes connected to each of the peripheral devices; a second counting means for counting the number of times information has been transferred via the above-mentioned route; and a retry regarding the information transfer via the above-mentioned route. a third counting means for counting the number of failures; a failure history registration means for registering the counted value of the second counting means and the counted value of the third counting means; An information processing apparatus comprising: control means for specifying a route for which information transfer is prohibited to the route activation control means based on the value.