JP2005340509A - Information processing system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To prevent system-down and the burglary of an information processor by making the information processor removable only at the time of switching the information processor due to the configuration change of an information processing system when the failure of the information processor is generated. <P>SOLUTION: This system is configured so that a plurality of modules 21 and 22 can be stored from the front side and back side of a case body 12 in one case body 12. The modules 21 and 22 show such a status that they are mounted in the case body 12, and that they are removed from the case body 12. The width dimensions of the modules 21 and 22 are made different according to a function included as an information processor. The modules 21 and 22 are respectively provided with automatic lock mechanisms. When the modules are mounted on the case body 12 and put in an operating status, the modules are locked so as not to be removed from the case body 12. When any failure is generated in the modules, or any instruction is issued from a management terminal, the modules are unlocked so as to be removed from the case body. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an information processing system, and more particularly, to an information processing system configured by mounting a plurality of information processing devices in a single housing at high density.

  In recent years, broadband networks have become widespread and Web services have been expanded. For this reason, information processing systems used for such services have dramatically increased the number of servers and other information processing devices installed, and there is a need to reduce management costs through system consolidation. Yes.

  As a conventional technique related to an information processing system that can meet the above-described requirements, for example, a technique described in Patent Document 1 is known. In this conventional technology, an information processing system is configured by mounting a plurality of high-density mounting information processing devices in a rack-mounted cabinet, and the plurality of information processing devices are centrally managed, so that effective use of space and high maintainability are achieved. Can be secured. The information processing system as described above is configured so that each mounted information processing device can be hot-plugged, and when replacing a failed device or when changing the system configuration, etc. It can respond flexibly.

  When a large-scale system is constructed using the information processing system as described above, it is necessary to efficiently manage all of the mounted high-density mounting type information processing apparatuses. However, as the number of devices to be managed increases, it becomes more difficult to isolate the information processing device where the failure has occurred, and when the device is replaced, not the device to be replaced but the operating device is accidentally removed. There is a risk of system down. For this reason, there is a demand for an information processing system that can always monitor the state of an installed information processing apparatus and can clearly distinguish between an information processing apparatus in operation and an information processing apparatus to be exchanged.

  As conventional techniques that can meet the above-described requirements, for example, techniques described in Patent Document 2, Non-Patent Document 1, and the like are known.

  The prior art described in Patent Document 2 is an information processing system that allows a user to recognize the state of an information processing device that can be hot-plugged and inserted based on the display state of a light emitting unit provided in the device. As a result, the system administrator can distinguish between a normally operating device and a device in which a failure has occurred.

The prior art described in Non-Patent Document 1 relates to a standard interface specification for enabling hardware mounted on an information processing apparatus such as a server to be monitored via a network. The state of the CPU, cooling fan, power supply, etc. in the apparatus can be monitored via a network. In actual operation, an IPMI-compliant chip is mounted on the information processing apparatus, and the state of the information processing apparatus is monitored via a network from a terminal in which management software is installed. The IPMI-compliant chip constantly monitors hardware, and when a failure occurs in any part, notifies the management terminal of failure information via the network.
JP 2001-356841 A JP 2000-22357 A "IPMI (Intelligent Platform Management Interface) 1.5" March 2001 held by Intel Corporation "Intel Developer Forum"

  As described above, an information processing system in which a plurality of information processing devices that can be hot-swapped is installed and operated is used when a failure occurs in one information processing device or when the system configuration is changed. Since only the information processing apparatus to be exchanged can be easily exchanged without stopping the system, it has high maintainability.

  However, in the prior art described in Patent Document 2, it is necessary to determine the state of the device based on the display state of the light emitting unit, and the system administrator recognizes in advance all the display specifications of each light emitting unit of the information processing device. Therefore, there is a problem that it becomes difficult to maintain high maintainability as the number of devices to be managed increases.

  In addition, since the prior art described in Non-Patent Document 1 is a method of monitoring a state from a management terminal via a network, when an apparatus is actually replaced in an environment where there are a large number of information processing apparatuses to be managed, an error occurs. There is a possibility that the apparatus in operation is removed, and this method causes a problem that it is difficult to maintain high maintainability.

  Furthermore, while all of the above-described conventional technologies can easily replace the information processing device even during system operation, the information processing device is stolen by a malicious third party other than the system administrator, and the system is down. If a large-scale system is constructed using a large number of information processing devices that can be hot-plugged, the number of information processing devices that need to be managed increases. Therefore, there is a problem in that when an apparatus is replaced when a failure occurs in the information processing apparatus, the operating apparatus may be mistakenly removed to cause a system down.

  The object of the present invention is to solve the problems of the prior art described above, and to maintain high maintenance not only when a failure occurs in the information processing apparatus but also when the information processing apparatus is replaced due to a change in the configuration of the information processing system. It is an object of the present invention to provide an information processing system composed of a plurality of highly maintainable information processing devices that can maintain the reliability and reliability, and can prevent the system from going down and theft of the information processing device.

  According to the present invention, in an information processing system configured by storing a plurality of information processing devices in a single housing, each of the plurality of information processing devices includes a lock mechanism that electrically locks, By automatically locking the device when mounted on the housing, the device cannot be removed from the housing.

  According to the present invention, when a failure occurs in the information processing device, only the device in which the failure has occurred can be removed, so that it is possible to prevent human errors such as accidentally removing the device that is operating. , System down, theft of the information processing apparatus can be prevented, and the maintainability of the entire information processing system can be improved.

  Embodiments of an information processing system according to the present invention will be described below in detail with reference to the drawings. The embodiment of the present invention described below is information called a blade server that operates a plurality of information processing devices having functions such as a server, a switching hub, a cooling fan, and a PCI slot in a rack mount cabinet. A processing system is taken as an example, and an information processing apparatus that can be mounted on such a blade server is hereinafter referred to as a module.

  FIG. 1 is a perspective view showing a configuration example of an information processing system according to an embodiment of the present invention. In FIG. 1, 11 is a blade type server, 12 is a housing, 21 and 22 are modules, and 23 is a pull-out handle.

  A blade type server as an information processing system according to an embodiment of the present invention is configured by storing a plurality of modules 21 and 22 from the front side and the back side of a case 12 in one case 12. A module 21 shown in FIG. 1 shows a module mounted on the housing 12, and a module 22 shows a state where the module is removed from the housing. Each of the modules 21 and 22 has a width dimension that differs depending on a function incorporated as an information processing apparatus, and is configured to have a width dimension that is an integral multiple of a certain reference dimension. In the illustrated example, modules having width dimensions of 1 times, 2 times, and 3 times the standard dimension are mixed and stored in the width direction of the housing 12. Such blade type servers 11 can be configured as a larger system by stacking a plurality of the blade servers 11 in the height direction.

  The size of the blade server shown in FIG. 1 is generally about 44.1 cm in width, 70 cm in depth, and 9 cm to 44 cm in height as shown in FIG. The width of each module is 4.4 cm as described above.

  Each of the plurality of modules 21 and 22 has functions of a server, a switching hub, a cooling fan, a PCI slot, and the like, and is configured to be hot-plugged. Further, each of the plurality of modules 21 and 22 has an automatic locking mechanism as will be described later. When the module is mounted on the housing 12 and is in an operating state, the module cannot be removed from the housing 12. When the module is locked and a failure or the like occurs, or, as will be described later, the lock is released by an instruction from the management terminal, and the module can be removed using the pull-out handle 23.

  By having the configuration as described above, the embodiment of the present invention makes it possible to easily replace a module when replacing a failed module or when changing the system configuration, etc. Maintainability can be ensured.

  FIG. 2 is a side view of the modules constituting the blade server 11 shown in FIG. 1, and a lock mechanism used in the embodiment of the present invention will be described with reference to FIG. In FIG. 2, 31 and 32 are lock mechanisms, 41 is a lever, and other numerals are the same as those in FIG.

  A lock mechanism 31 for controlling insertion / removal of the module is provided at an opening on the front surface of each module 21, 22, and a lever 41 having a lock mechanism 32 for fitting with the lock mechanism is provided. As shown in FIG. 2A, the lock mechanism 31 is unlocked and the module can be removed while the upper portion of the lever 41 is tilted toward the front side. Further, as shown in FIG. 2B, when the module is inserted into the housing 12 and the lever 41 is closed in a vertical state, the locking mechanism 31 on the front surface of the module and the locking mechanism 32 provided on the lever 41 are As a result, the module is electrically locked, and the module cannot be removed. In FIG. 2, the lock mechanism 31 is shown separately in white and black, but white indicates unlocking and black indicates locking. Moreover, since a general-purpose electric lock mechanism well known as an electromagnetic lock or the like can be used as the lock mechanism, a detailed description of its structure is omitted.

  As already described, the modules 21 and 22 can be hot-plugged, and the module having the server function is configured by a circuit on which a CPU, a memory, and the like are mounted. In the circuit, a monitoring chip that conforms to or corresponds to IPMI is mounted in order to monitor the state of the module. In addition to the server function, the module having the function of substituting the switching hub, cooling fan, PCI slot, etc., also monitors the status inside the module in the same manner as described above, so that it conforms to IPMI or equivalent monitoring. Chip is mounted.

  When the lock mechanism 31 mounted in the opening of the modules 21 and 21 having the functions described above is closed by the lever 41 and the module is mounted in the housing 12 constituting the blade server 11. The module is automatically and electrically locked so that the module cannot be removed from the housing 12 when the module is mounted. Thereby, the maintainability of the whole system can be improved. Note that the lock is released only when the module is replaced when a failure occurs in the module or when the configuration of the blade server system is changed.

  As described above, according to the embodiment of the present invention, by providing the module with the lock mechanism, it is possible to prevent the system from being down due to erroneous removal of the operating module, and to prevent the module from being stolen as the information processing apparatus. Therefore, an information processing system with high maintainability can be provided.

  FIG. 3 is a cross-sectional view of the blade server 11 shown in FIG. 1 as viewed from the side, and will be described next.

  As shown in FIG. 3A, the blade type server 11 has a configuration in which modules are housed in the housing 12 from both the front and back sides of the housing 12, as described with reference to FIG. When the module is operating normally as indicated by the module 21, the module 21 has the lever 41 closed and the lock mechanism 31 provided in the opening of the module is in an activated state. As a result, the module 21 in normal operation is held by the housing 12 so that it cannot be removed from the housing 12, and can be prevented from being removed during module operation.

  When a failure occurs in the module 21 shown on the right side of FIG. 3A, a state monitoring chip (not shown) provided in the module 21 detects the failure and determines information on the failure in the module 21. Then, the lock of the lock mechanism 31 is released. As a result, the faulty module can be removed with the lever 41 opened and the entire module moved slightly in the direction of removing the housing 12, as shown as module 22 in FIG. 3B. It becomes a state. A mechanism for slightly moving the module in the direction of removing the housing 12 when the lock mechanism 31 is unlocked is not shown, but for example, when the electromagnetic lock as the lock mechanism 31 is released, a spring or the like It may be a known mechanism that pushes out the module. As a result, since only the module in which the failure has occurred can be removed, it is possible to prevent a human error such as erroneously removing an operating device and to improve maintainability of the entire system.

  FIG. 4 is a flowchart for explaining processing in the module when a failure occurs in the module. Next, this will be described.

  When the module is inserted and mounted in the housing, the lock mechanism is electrically locked (step 401). Thereafter, the above-described state monitoring chip provided in the module monitors the operation state of the module, and when detecting that a failure has occurred in the module (step 402), unlocks the module (step 403). .

  FIG. 5 is a diagram for explaining a configuration example of an information processing system according to another embodiment of the present invention, which is an example provided with a management terminal that manages each module constituting a blade type server. In FIG. 5, 50 is a management terminal, 60 is a network, and other symbols are the same as those in FIGS.

  The blade type server 11 shown in FIG. 5 (a) shows a cross section viewed from the side when the module is mounted, as in the case shown in FIG. 3 (a), and FIG. )), That is, the module is in a normally operating state. In the example described here, a management terminal 50 in which software for managing each module is installed is connected to each module constituting the blade server 11 via the network 60. The management terminal 50 can detect the failure information of the module at any time via the network 60 by a monitoring chip that complies with IPMI or is equivalent to the IPMI mounted on each module.

  When a failure occurs in the module 21 shown on the right side of FIG. 5A, a state monitoring chip (not shown) provided in the module 21 detects the failure information and is set in advance from the management device 50. Whether or not to release the lock is determined in the module 21 according to the unlocking condition, and whether or not to release the lock of the lock mechanism 31 is controlled. As a result, the faulty module can be removed with the lever 41 opened and the entire module moved slightly in the removal direction of the housing 12, as shown as module 22 in FIG. 5B. It becomes a state. In the example illustrated in FIG. 5, even when a failure has occurred in the module, the lock of the designated module is released when the management device 50 designates the module and gives an unlock instruction. .

  In the above description, the management terminal 50 can also be configured in one of the modules constituting the blade server 11.

  FIG. 6 is a flowchart for explaining processing in a module when a failure occurs in the module in another embodiment of the present invention shown in FIG. 5, which will be described next.

  In the processing of the other embodiment of the present invention described here, the failure information obtained in the module 21 is a known problem of the system administrator and is determined to be acceptable, When it is desired not to be able to release the lock by judgment, it is not necessary to release the lock of the module 21, so it is necessary to make the lock release condition flexible. Therefore, first, a lock release condition is set for each of the plurality of modules 21 from the management terminal 50, and the release condition is reflected on the lock mechanism 31 in each module via the network 60.

(1) When the module is inserted and mounted in the housing, the lock mechanism electrically locks the module, and then the above-described state monitoring chip provided in the module monitors the operation state of the module. The module is operated as a system (step 601).

(2) When it is detected that a failure has occurred in the module, the module notifies the management terminal 50 of the failure information of the module via the network, and the unlocking conditions set in advance are: It is determined whether or not the lock is released regardless of the content of the failure (steps 602 to 604).

(3) If it is determined in step 604 that the lock release condition is not to release the lock regardless of the content of the failure, whether or not the content of the failure information matches the lock release condition. If it does not match, the process here is terminated without releasing the lock (step 605).

(4) If the content of the failure information matches the condition for releasing the lock in the determination in step 605, or the determination in step 604 releases the lock regardless of the content of the failure. , The module is unlocked, and the processing here ends (step 606).

  FIG. 7 is a flowchart for explaining the processing in the module when there is an unlocking instruction from the management terminal in the other embodiment of the present invention shown in FIG.

(1) When the module is inserted and mounted in the housing, the lock mechanism electrically locks the module, and then the above-described state monitoring chip provided in the module monitors the operation state of the module. The module is operated as a system (step 701).

(2) When receiving an instruction from the management terminal, the module receives the instruction and determines whether or not the received instruction is unlocking (steps 702 and 703).

(3) If it is determined in step 703 that the received instruction is unlocked, the module is unlocked, the processing here is terminated, and the received instruction is another instruction that is not unlocked. The processing here is terminated without releasing the lock (step 704).

  Each processing in the above-described embodiment of the present invention can be configured as a processing program, and this processing program is stored in a recording medium such as HD, DAT, FD, MO, DVD-ROM, and CD-ROM and provided. can do.

  The process described with reference to FIG. 6 can be used to unlock the failed module. When the process described with reference to FIG. 7 is used, the locked module 21 is replaced due to a system configuration change or the like. The lock release instruction command can be transmitted from the management terminal 50 via the network 60, and the lock mechanism 31 in the module 21 can receive this command to unlock the module. As a result, it is possible to remove only a failed device or a module that is to be replaced due to a system configuration change, etc., thus preventing human error such as accidentally removing a device that is operating, The maintainability of the entire system can be improved.

  FIG. 8 is a diagram for explaining an example of a display screen on the management terminal when a lock release instruction is issued from the management terminal.

  On the display device of the management terminal 50, as shown as a display screen 81, an arrangement 82 of modules constituting the blade server 11 and an unlock button 83 are displayed. Each module is displayed with an identifier of the module according to the arrangement order or the like in the housing 12, and in the example shown in FIG. 8, a number indicating the arrangement order is given. Each module may be displayed by, for example, color-coding according to an operation state such as normal operation, rest, and failure. Then, as explained in the flow shown in FIG. 7, the system administrator selects a module to be unlocked with a mouse or the like using the display screen shown in FIG. 8 when issuing an instruction to unlock the module from the management terminal. Then, after clicking, the unlock button 83 may be clicked. Thereby, the process of the flow shown in FIG. 7 is started, and the lock of the designated module can be released.

  According to the above-described embodiment of the present invention, the module can be electrically locked at the time of mounting the module, and the removal of the module can be restricted. Therefore, the system down and the module can be prevented from being stolen. Maintainability can be improved.

  Further, according to the embodiment of the present invention described above, when a failure occurs in the module, the lock is automatically released based on the judgment of the module, and the failed module can be removed. It is possible to prevent human error such as erroneously removing an operating device and to improve maintainability of the entire system.

  Furthermore, according to the above-described embodiment of the present invention, the lock release condition and the lock release when a failure occurs in the module can be controlled via the network from the management terminal. It is possible to flexibly set a condition for automatically releasing the lock when a failure occurs, and it is possible to release the lock even when it is desired to replace the apparatus by changing the system configuration.

It is a perspective view showing an example of composition of an information processing system by one embodiment of the present invention. It is a side view of the module which comprises the blade type server 11 shown in FIG. It is sectional drawing seen from the side surface side of the blade type | mold server 11 shown in FIG. It is a flowchart explaining the process in a module when a failure generate | occur | produces in a module. It is a figure explaining the structural example of the information processing system by other embodiment of this invention. In other embodiment of this invention, it is a flowchart explaining the process in a module when a failure generate | occur | produces in a module. In other embodiment of this invention shown, it is a flowchart explaining the process in the module when there exists an instruction | indication of unlocking from the management terminal. It is a figure explaining the example of the display screen in a management terminal in the case of instruct | indicating lock release from a management terminal.

Explanation of symbols

DESCRIPTION OF SYMBOLS 11 Blade type server 12 Case 21, 22 Module 23 Pull-out handle 31, 32 Lock mechanism 41 Lever 50 Management terminal 60 Network

Claims (3)

  In an information processing system configured by storing a plurality of information processing apparatuses in a single casing, each of the plurality of information processing apparatuses incorporates a lock mechanism that electrically locks, and the casing of the information processing apparatus An information processing system characterized in that it is made impossible to be removed from the housing by automatically locking when mounted on the housing.   2. The information processing system according to claim 1, wherein each of the plurality of information processing apparatuses constantly monitors the state of the own apparatus and releases a lock when a failure occurs in the apparatus. 3. The management terminal for managing the plurality of information processing apparatuses, the management terminal controlling a lock release condition and a lock release of each of the plurality of information processing apparatuses. Information processing system.

Images