JP2002268780A - Computer system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To inhibit the erroneous insertion/extraction of a module in a computer system constituted of at least one module. SOLUTION: This computer system is provided with a fixing means capable of fixing an arbitrary module to a slot, a fixing releasing means capable of releasing the fixing, and a fixing control means capable of controlling them. Also, the fixing control means is controlled by a managing means for managing the state of the device equipment. Also, the fixing control means is controlled by a remote management computer to be operated by a maintenance staff.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a maintenance and management technique for a computer system composed of a plurality of modules.

2. Description of the Related Art A backbone computer system, which is responsible for a social infrastructure,
Since high reliability is required, many maintenance costs are incurred. In addition, maintenance costs are increasing for the following reasons. The reason is that the computer system becomes large-scale and the types of modules constituting the computer system increase (the specifications of modules tend to be open, and the types of modules are becoming diversified. And computer systems that have been distributed and independent at remote locations are becoming increasingly cooperative.

[0002] In order to reduce the above-mentioned maintenance cost, automation of maintenance management of computer systems and expansion of remote maintenance are required.

Against this background, in a computer system composed of a plurality of module groups, techniques related to maintenance efficiency, particularly management of module groups and control of insertion / removal of individual modules, will be described below. The technology is known.

The first technique disclosed in Japanese Patent Application Laid-Open No.
In the prior art, in a computer system composed of a plurality of hot-swappable modules, the modules are duplicated to automatically eject a failed module and automatically install a module to replace a module after installation. Storage is possible.

A second conventional technique disclosed in Japanese Patent Application Laid-Open No. H11-85649 discloses a method in which a management computer that manages a group of computer nodes to be managed is remotely managed via a network. Have been. For example, the management computer can detect and change the state of any module on the computer system to be managed.

In the third conventional technique disclosed in Japanese Patent Application Laid-Open No. H11-296375, stored modules can be automatically recognized by plug-and-play, and device devices included therein can be validated.

[0007] By combining the second and third conventional techniques, it is possible to remotely support replacement of a failed module that has failed or is expected to occur. Further, when these are combined with the first conventional technique, much of the replacement work can be automated.

For example, consider a case where a failure occurs in a module of a computer system at a customer site where there is no maintenance staff, and a remote maintenance staff supports the module replacement work.

Hereinafter, a description will be given with reference to FIGS. FIG. 1 shows a configuration of a computer system 10 requiring replacement of a module. The computer system 10 has a plurality of slots 40, and one module 50 is inserted in each slot 40. In FIG. 1, a failure occurs in one of the inserted modules 50a (hereinafter referred to as a failure module 50a), and an alternative module 5a
0b (hereinafter, referred to as a substitute module 50b) is inserted (not stored yet).

The procedure of the replacement work is as follows. First, the substitute module 50b is inserted into the computer system (Step 1-1). FIG. 1 shows a state immediately after the processing of step 1-1. Next, the substitute module 50
b is stored in the computer system (step 1-2), and the operation is confirmed by activating the substitute module 50b (step 1-3). Further, the failed module 50a is invalidated (step 1-4), the failed module 50a is ejected (step 1-5), and the failed module 50a is removed (step 1-6).

FIG. 2 shows the computer system 10 immediately after the processing of step 1-5. In FIG. 2, unlike FIG. 1, the replacement module 50b is stored in the slot 40, and the failed module 50a is ejected.

Here, the work required by the worker at the customer site where the computer system 10 is installed is only steps 1-1 and 1-6. The remaining processing can be performed by maintenance personnel at a remote maintenance center or management means built in the computer system 10. That is, the worker at the customer site can perform the replacement work without operating the module stored in the group of modules 50.

[0013]

The above-mentioned prior art is capable of supporting the exchange of the faulty module 50a and the replacement module 50b and is useful for preventing erroneous insertion and removal, but prohibits erroneous insertion and removal of the module 50 itself. It is not possible.
In the example shown in FIGS. 1 and 2, it is possible for a worker to accidentally remove another module 50. Then, a degenerate operation of the computer system 10 newly occurs, and further maintenance work is required. Further, an operator can intentionally remove an arbitrary module. As described above, since the safety and reliability of the replacement work depend on the skills and morals of the worker, not everyone can perform such replacement work safely and reliably.

Further, according to the above-mentioned conventional technique, since the insertion of the module 50 into the computer system 10 cannot be prohibited, the following problem also occurs. That is, when a maintenance person has not authorized the insertion of the module 50 into the computer system 10, the operator can intentionally insert the module 50. This also causes a malfunction of the computer system 10 and is not preferable. However, also in this case, removal of the faulty module 50a should be permitted, and it is desirable that control of insertion and removal of the module 50 can be performed.

The problem to be solved by the present invention is to control the fixing of the module so as to prohibit the above illegal insertion and removal.

A first object of the present invention is to control insertion / removal of an arbitrary module and to prevent erroneous insertion / removal. Further, a second object is to perform the above control depending on the state of the module group of the computer system. The third purpose is
The above control is performed from a remote management computer. 4th
The purpose is to allow removal of only the failed module. A fifth object is to reliably insert a replacement module and remove a failed module in this order.

[0017]

For the first purpose, a fixing means and a fixing releasing means for releasing its function are provided in each of the slots, and the fixing means and the fixing releasing means are provided for each of the slots. Is provided with fixed control means that can be controlled independently.

For the second purpose, the fixed control means monitors the state of the device group and controls the fixed state of an arbitrary slot in accordance with an instruction from the managing means for changing.

For the third purpose, the fixed control means controls the fixed state of an arbitrary slot in accordance with an instruction from the management computer.

[0020] For the fourth purpose, the securing device permits only removal of the module. Specifically, the fixing device is
Once the module is unlocked by the unlocking device and the module is removed, insertion of the module is prohibited again. Further, the fixing control means activates fixing releasing means for releasing the fixing of the module after the failure of the faulty module is detected or predicted.

For the fifth purpose, the fixed control means activates the fixing means for fixing the replacement module after the replacement module has been activated, and the fixed control means operates after the failure module has been disabled for the fault. Activating the fixing releasing means for releasing the fixed state of the module.

[0022]

Embodiments of the present invention will be described below with reference to the drawings.

First, a first embodiment will be described with reference to FIGS.

FIG. 3 is an overall configuration diagram of the first embodiment of the present invention.

The computer system 10 includes a slot 40, a module 50, a fixing unit 30, a fixing releasing unit 31, a fixing control unit 20, and a first network 110. The module 50 has a plurality of device devices 51 therein. The slot 40 has an interface 41. Each device 51 can input and output arbitrary information via the interface 41. The first network 110 connects the interface 41 and the fixed control means 20, and enables mutual communication.

It should be noted that the present invention can be carried out by using a bus in which the interface 41 and the fixed control means 20 can communicate with each other, instead of the first network 110. Further, the housing outer wall 120 shields all of the above except the opening of the slot 40 and the module 50 from the outside.

A pair of fixing means 30 and fixing releasing means 31 are fixed to the slots 40, respectively. The slot 40 stores at most one module 50. The fixing means 30 and the fixing releasing means 31 follow the instructions of the fixing control means 20. When the fixing means 30 operates, insertion and removal of the module 50 stored in the slot 40 is prohibited. When the unlocking means 31 operates, the module 50 stored in the slot 40 can be freely inserted and removed.

The fixed control means 20 is connected to the first network 1
Information can be input to and output from the device 51 via the device 10.
That is, the fixed control unit 20 can detect whether any device 51 is enabled or disabled by the above input / output. In the first embodiment, if all device devices 51 included in a certain module 50 are invalidated, the module 50 can be removed. In that case, the fixed control means 20
0 insertion / removal is permitted. This permission method will be described later in detail. The number of slots 40 is arbitrary. Similarly, the number of modules 50 and the number of device devices 51 included in one module 50 are also arbitrary.

FIGS. 4 and 5 are enlarged views of the module 50, the slot 40, the fixing means 30, and the fixing releasing means 31. FIG. In each of the figures, the fixing means 30 includes a flange 301
And the spring 302, and the fixing release means 31 has an actuator 311.

The spring 302 pushes the flange 301 to push the fixing means 30 leftward in the figure. The actuator 311 is connected to the fixed control means 20, and expands and contracts according to the instruction. The left end of the actuator 311 is connected to the fixing releasing means 31, and the right end is the flange 3
It touches 01. The actuator 311 is normally contracted, and the flange 301 is pushed to the left end by the spring 302.

FIG. 4 shows a state in which the fixing means 30 is operated to prevent the module 50 from being inserted and withdrawn, and a state in which the fixing means 30 is released.
1 indicates that the module 1 has been activated to release the module 50 from being fixed.

At the time of fixing, the fixing means 30 is inserted into the hole penetrating the wall of the slot 40 and the hole formed in the module 50 by the action of the spring 302. Module 50
Can be removed by moving it upward in FIG. 4, but it cannot be removed at the time of fixation because the fixing means 30 is in a bar shape. As shown in FIG. 3, the computer system is shielded by the casing outer wall 120 except for the slot opening and the module, so that the fixing control means 20 issues an instruction, and the fixing means 30 and the fixing releasing means There is no way to activate 31.

At the time of releasing the fixing, the fixing means 30 is pushed rightward by the action of the actuator 311, and comes out of the holes of the module 50 and the slot 40. For this reason, the module 50 can be freely inserted and removed when the fixing is released. As described above, since the flange 301 is normally pushed to the left end, the fixing means 30 is in a fixed state, and insertion and removal of the module 50 is prohibited.

FIG. 5 shows the operation of the fixing means 30 and the fixing releasing means 31 when only the removal of the module 50 is permitted.

When only the removal of the module 50 is permitted, the fixed control means 20 does not extend the actuator 311 to the right end but to the position shown in FIG. At the time of removal, since the left end of the fixing means 30 has an oblique notch, the fixing means 30 is pushed to the right by sliding at the entrance of the hole of the module 50. Flange 301 is actuator 31
Since the fixing means 30 is not fixed to 1, the fixing means 30 moves rightward as it is, and passes through the hole of the module 50. Thus, the first removal of the module 50 is successful. Then, once the module 50 is removed, the fixing means 30 is pushed to the left by the action of the spring 302. Then, since it comes into contact with the lower part of the module 50 to be inserted again, it cannot be inserted again.

As described above, in the first embodiment,
Only removal of the module 50 can be permitted.

Next, a second embodiment will be described with reference to FIG.

FIG. 6 is an overall configuration diagram of the second embodiment.

The difference between FIG. 6 and FIG.
0 is to include the management means 200 connected to the first network 110. The management unit 200 can monitor or change the state of the device 51 via the first network 110. Further, the management means 200 and the fixed control means 20 can communicate with each other. This communication is performed via the first network 110 in the second embodiment, but the present embodiment can be performed even if this communication is performed using another means.
Furthermore, in the second embodiment, the fixed control means 20 does not need to monitor the state of the device 51, and therefore, if the fixed control means 20 can communicate with the management means 200,
There is no need to connect to the first network 110.

When all the device devices 51 included in a certain module 50 are invalidated, the management means 200 determines that the module 50 can be removed, and instructs the fixed control means 20 to make the module 50 operate. Allow insertion and removal. Further, the fixing control means 20 operates the fixing releasing means 31 to bring the module 50 into a state at the time of releasing the fixing.

In the second embodiment, the module 5
The method of permitting only the removal of 0 is the same as in the first embodiment.

Next, a third embodiment will be described with reference to FIGS.

FIG. 7 is an overall configuration diagram of the third embodiment.

The difference between FIG. 7 and FIG. 6 is that a second network 11 and a management computer 12 are provided, and that the computer node 10 is connected to one or more computer node groups connected to the second network 11. It is one. Although only one computer node 10 is illustrated in FIG. 7, the present invention can be implemented even if there are a plurality of computer nodes 10. FIG. 8 is a timing chart showing the processes of the fixed control unit 20, the management unit 200, and the management computer 12 in the third embodiment in parallel.

In the initial state, the management means 200 is waiting for an event. When the management unit 200 detects an abnormality of the device 51, the information identifying the device 51 and the information identifying the module 50 including the device 51 are transmitted to the management computer 12 through the second network 11.
(Step 200-1). Thereafter, the management means 200 waits for an event again (step 200-
2). The management computer 12 that has received the notification predicts a failure of the device 51 from the content of the notification (step 12-
1). If a failure can be expected in the device device 51 (here, a failure can be expected), information for identifying the module 50 including the device device 51 and an instruction to release the fixation of the module 50 are transmitted through the second network 11. This is sent to the management means 200 (step 12-2), and the process is completed. After receiving the instruction, the management means 200 waiting for the event transmits information for identifying the module 50 and an instruction to release the fixation of the module 50 to the fix control means 20 (step 200-40), and the process is completed.

Upon receiving the instruction, the fixing control means 20 releases the fixing of the module 50 in the same manner as in the second embodiment (step 20-10), and completes the processing. In addition,
Similarly, when the management unit 200 detects not a failure but a failure of the device 51 in step 200-2, similarly, in step 12-2, issues an instruction to release the fixation of the module 50 to the management unit 200. I do. Therefore,
Thereafter, similarly, Step 200-40 and Step 20-1
0 is executed.

In this way, the fix control means 20 releases the fixation of the module 50 after the management computer 12 detects or predicts a failure of the module 50.

Next, a fourth embodiment will be described with reference to FIGS.

FIG. 9 is an overall configuration diagram of the fourth embodiment.

The difference between FIG. 7 and FIG. 9 is that two sets of slots 40, a module 50,
0 means that the fixing release means 31 is distinguished by a and b, respectively. Hereinafter, it is assumed that a failure has occurred in the module 50a (hereinafter, referred to as a failure module), and the module 50b is a module that substitutes for the module 50a (hereinafter, referred to as a substitution module). In the initial state, the replacement module 50b is not stored in the slot 40b,
It is assumed that the fixing releasing means 31b is operating. Details of the operation will be described below.

FIG. 10 shows the fixed control means 20, the management means 200, and the management computer 12 according to the fourth embodiment.
5 is a timing chart showing the processing in parallel.

The difference between FIG. 8 and FIG.
1 to Step 20-2 and Step 200-10
To 200-11, steps 200-20 to 200-
22, step 200-30, steps 12-10 to 12-13, step 12-30, and not including step 12-2.

Hereinafter, the fixed control means 20 will be described with reference to FIG.
The processing of the management unit 200 and the management computer 12 will be described.

Step 200-1 to step 200-
Steps 2 and 12-1 are the same as in the third embodiment. Next, the management computer 12 stores the failure module 50a that has predicted a failure in step 12-1 (step 12-10), and waits for an event (step 12-1).
1). Next, when the management unit 200 waiting for an event detects that the substitute module 50b has been inserted,
Information for identifying the module 50b is reported to the management computer 12 (step 200-10).

Thereafter, the management means 200 waits for an event again (step 200-11). The management computer 12 having received the notification, based on the content of the notification, substitute module 50b,
The compatibility with the failure module 50a stored in step 12-10 is confirmed, and if the compatibility can be confirmed (step 1)
2-12) The management module 20 stores an instruction to enable and fix the alternative module 50b and information for identifying the module 50b.
0 (step 12-13). Subsequently, the management computer 12 transmits an instruction to invalidate and release the failure module 50a and information for identifying the module 50a to the management unit 200 (Step 12-30). FIG.
In the case of 0, the processing is divided into two in and after step 12-13, but the present invention can be implemented even if the processing is performed sequentially with synchronization.

Next, the management means 200, having received the instruction of step 12-13, activates the alternative module 50b (step 200-20), and fixedly controls the instruction to fix the alternative module 50b and the information for identifying the module 50b. It instructs the means 20 (step 200-21) and waits for an event again (step 200-22). Steps 200-21 and 200-40 need to be performed in this order.

Next, the fixing control means 20, which has received the instruction in step 20-22, activates the fixing means 30b to fix the substitute module 50b (step 20-1).
The management unit 200 receiving the instruction of step 12-30,
Disable the failure module 50a (step 200-
30). After that, similarly to the third embodiment, steps 200-40 and step 20-10 are executed to release the failure module 50a from being fixed.

As described above, the fixing means 20 fixes the substitute module 50b after the substitute module 50b is activated by the instruction of the management computer 12.
0b, and after the failure module 50a is invalidated by the instruction of the management computer 12, the failure module 50a
Then, the fixing releasing means 31a for releasing the fixing is operated.

Next, a fifth embodiment will be described with reference to FIG.

FIG. 11 is an overall configuration diagram of the fifth embodiment.

The computer system 10 is a 19-inch rack mount system that is generally used as a server system. The rack 120 includes a slot 40, a thin server 50, a fixing unit 30, a fixing releasing unit 31, and a first network 1.
10. A management unit 200 is provided. The thin servers 50 and the management unit 200 can communicate with each other via the first network 110. Further, the management means 200
Can communicate with the management computer 12 via the second network 11.

In the fifth embodiment, the insertion / removal of the thin server 50 is controlled in the above-described configuration in the same manner as in the first to fourth embodiments.

[0063]

According to the present invention, erroneous insertion and removal of modules can be prohibited in a computer system including a plurality of modules.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of a module replacement operation according to a conventional technique.

FIG. 2 is a configuration diagram of a module replacement operation according to a conventional technique.

FIG. 3 is an overall configuration diagram of the first embodiment.

FIG. 4 is an enlarged view of a module 50, a slot 40, a fixing means 30, and a fixing releasing means 31.

5 is an enlarged view of the module 50, the slot 40, the fixing means 30, and the fixing releasing means 31. FIG.

FIG. 6 is an overall configuration diagram of a second embodiment.

FIG. 7 is an overall configuration diagram of a third embodiment.

FIG. 8 is a timing chart of a process according to the third embodiment.

FIG. 9 is an overall configuration diagram of a fourth embodiment.

FIG. 10 is a timing chart of a process according to the fourth embodiment.

FIG. 11 is an overall configuration diagram of a fifth embodiment.

[Explanation of symbols]

10: computer node, 11: second network, 1
2: management computer, 20: fixed control means, 30: fixed means, 31: fixed release means, 40: slot, 41: interface, 50: module, 51: device,
110: first network, 120: outer wall of housing, 20
0: Management means

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Kento Sakuraba 1099 Ozenji Temple, Hitachi, Ltd.System Development Laboratory, Hitachi, Ltd. Address: Hitachi, Ltd.System Development Laboratory (72) Inventor Hiroshi Shimabara 5030, Totsuka-cho, Totsuka-ku, Yokohama-shi, Kanagawa Prefecture, Hitachi, Ltd.Platform Solution Development Division

Claims (6)

[Claims] 1. A module group comprising modules each having one or more device units, and a slot comprising a slot each storing one of the modules and having an interface for inputting and outputting information to and from the device units. A computer system connected to the interface and having communication means for allowing each of the device units to communicate with each other; provided in correspondence with each of the module groups stored in each of the slots; A fixing device group that prohibits the insertion and removal of the device; a fixing release device group that is provided corresponding to each of the fixing device groups and releases the function of the corresponding fixing device; and the device device and the information via the communication unit. A computer having a fixing control device for inputting and outputting data and operating the fixing device and the fixing releasing device. Stem. 2. A module group comprising modules each having one or more device units, and a slot comprising slots each storing one module and having an interface for inputting and outputting information to and from the device units. And a communication unit connected to the interface, for communicating with each other, and each of the device units manages the state of each module and each device unit through the communication unit, and enables the device unit. A computer system having management means for invalidating or invalidating, a fixing device group provided corresponding to each of the module groups stored in the respective slots, and prohibiting insertion / removal of the corresponding module; and each of the fixing device groups , A group of fixing release devices for releasing the function of the corresponding fixing device, Computer system the input and output device apparatus and information, characterized by having a fixed control device for operating the locking device and the unlocking device Te. 3. A module group including modules each having one or more device units, and a slot including slots each storing one module and having an interface for inputting and outputting information to and from the device units. A first communication unit connected to the interface and the interface, and each of the device units communicates with each other; and via the communication unit, managing the state of each module and each device unit; One or more computer nodes having management means for enabling or disabling a device, a management computer for managing the computer node group, and a second communication means for communicating between the computer node group and the management computer with each other The computer system is provided corresponding to each of the module groups stored in the respective slots, and A fixing device group for inhibiting insertion / removal of a module; a fixing release device group provided corresponding to each of the fixing device groups to release the function of the corresponding fixing device; and the device device and the information via the communication unit. A computer system, comprising: a fixing control device that inputs and outputs data and operates the fixing device and the fixing releasing device. 4. The computer system according to claim 1, wherein said fixing device releases said module once by said fixing release device, and then operates again when said module is removed, and inserts said module. Computer system characterized by prohibition. 5. The computer system according to claim 2, wherein after the management unit detects a module in which a failure has occurred or is predicted to have occurred, the controller releases the fixation of the module. A computer system characterized by having 6. The computer system according to claim 2, wherein a second module which is compatible with the first module in which a fault has occurred or is expected to have occurred is inserted into an arbitrary empty slot. After the management means detects that the second module has been activated,
The fixing means for fixing the module of the second module is operated, and after the management means invalidates the first module, the fixing of the first module is operated by the release fixing releasing means. Computer system.