JP2013250732A - Blade server system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a blade server system for suppressing the installation space and costs of equipment, and for increasing the reliability of a system.SOLUTION: The blade server system includes: a first blade server for providing a first function; a second blade server for providing a second function which is different from the first function; a third blade server which functions as the redundant system of the first blade server; a fourth blade server which functions as the redundant system of the second blade server; and a management device which manages the first to fourth blade servers. The management device selects either the first blade server or the second blade server when detecting that a preliminary blade server has been added to the blade server system, and reconfigures the preliminary blade server so that it can function as the redundant system of the selected blade server.

Description

  The present invention relates to a blade server system.

  Conventionally, a dual system comprising a regular system and a standby system is known. For example, in Patent Document 1, when a spare blade server is put on standby in a symmetric multiprocessor blade server system and a failure occurs in a blade server constituting the symmetric multiprocessor blade server system, A blade server system is described in which the power is turned off and a spare blade server is turned on.

JP 2010-102523 A

  In the prior art, in a system having a plurality of nodes each providing different functions, when increasing the number of standby systems and improving the reliability of the system, it is necessary to prepare a plurality of standby systems for all of those nodes. There has been a problem that the installation space and cost of the equipment are greatly increased.

  The blade server system according to claim 1, a first blade server that provides a first function, a second blade server that provides a second function different from the first function, and a redundant system of the first blade server A blade server system comprising: a third blade server that functions as: a fourth blade server that functions as a redundant system of the second blade server; and a management device that manages the first to fourth blade servers, When the management device detects that a spare blade server has been added to the blade server system, the management device selects either the first blade server or the second blade server and functions as a redundant system of the selected blade server. The spare blade server is reconfigured as described above.

  ADVANTAGE OF THE INVENTION According to this invention, the reliability of a system can be improved, restraining the installation space and cost of an apparatus.

It is a block diagram which shows the whole structure of the system | strain monitoring control system to which the blade server system of this invention is applied. 2 is a schematic diagram illustrating a physical configuration of a system monitoring control system 100. FIG. 2 is a block diagram illustrating a configuration of a maintenance server 104. FIG. 2 is a block diagram showing a configuration of a blade server module 111. FIG. 10 is a flowchart of processing executed when a blade server module 111 is inserted into an empty slot.

(First embodiment)
FIG. 1 is a block diagram showing the overall configuration of a system monitoring control system to which the blade server system of the present invention is applied. The system monitoring control system 100 is a system that monitors and controls a power system such as a power plant, a substation, a transmission line, and a transformer. The system monitoring control system 100 includes a plurality of servers and clients connected to each other by a control LAN 110 that is a telecommunication line.

  The system monitoring server 103 is a server that monitors the operating status of individual servers that constitute the system monitoring control system 100. The remote control server 107 is a server that exchanges data with a distant power facility such as a substation. The I / F server 105 is a server that exchanges data with other systems and manages input / output between the system monitoring control system 100 and other systems. The information system server 101 is a server that manages a database in which information on each base, past power transmission history, and the like are stored, for example. The console client 102 is a client operated by an operator. The operator operates the console client 102 to input a predetermined control instruction or the like to each server. In FIG. 1, only one console client 102 is shown, but a plurality of console clients 102 may actually exist. The support server 106 is a server that supports the operation of the system by an operator, and performs, for example, power system power flow calculation. The monitoring control server 108 is a server that monitors each power system and performs control such as conduction and interruption of power as necessary.

  In the present embodiment, among the various servers described above, the information server 101, the support server 106, the remote control server 107, and the monitoring control server 108 have two servers having the same configuration, thereby adopting a redundant configuration. It has been. For example, the information system server 101 includes a first information system server 101a and a second information system server 101b having the same configuration. Normally, the first information system server 101a is in an operating state, and the second information system server 101b is on standby. It is in a state (a state that can be changed to an operating state at any time). When the first information server 101a stops due to a failure or the like, the second information server 101b transitions from the standby state to the operating state after a certain period of time or manual operation. However, the operation of the grid monitoring control system 100 is not hindered. Similarly, the support server 106 is the first support server 106a and the second support server 106b, the distance control server 107 is the first distance control server 107a and the second distance control server 107b, and the monitoring control server 108 is the first monitoring control server 108a and It consists of a second monitoring control server 108b. Although details will be described later, the control LAN 110 has a redundant configuration.

  The system monitoring control system 100 of this embodiment includes a maintenance server 104 and a maintenance LAN 109 in addition to the above-described units. The maintenance LAN 109 is connected to the first remote control server 107a, the second remote control server 107b, the first monitoring control server 108a, the second monitoring control server 108b, and the maintenance server 104. The configuration of the maintenance server 104 will be described in detail later.

  FIG. 2 is a schematic diagram illustrating a physical configuration of the system monitoring control system 100. The system monitoring control system 100 includes a first server chassis 1 and a second server chassis 2. The first server chassis 1 has four slots (first slot 11, second slot 12, third slot 13, and fourth slot 14) into which blade server modules can be inserted. Similarly, the second server chassis 2 includes four slots (fifth slot 21, sixth slot 22, seventh slot 23, and eighth slot 24) into which blade server modules can be inserted.

  The first monitoring control server 108a, the second monitoring control server 108b, the first distance control server 107a, and the second distance control server 107b of this embodiment are each configured by a single blade server module. A first monitoring control server 108a and a first distance control server 107a are inserted into the first slot 11 and the second slot 12 of the first server chassis 1. In the fifth slot 21 and the sixth slot 22 of the second server chassis 2, the second monitoring control server 108b and the second distance control server 107b are inserted.

  The control LAN 110 includes a first control LAN 110a and a second control LAN 110b that function independently. The eight slots provided in the first server chassis 1 and the second server chassis 2 are connected to the first control LAN 110a and the second control LAN 110b, respectively. Therefore, the blade server modules inserted into these eight slots can continue to communicate with the second control LAN 110b even when the first control LAN 110a stops functioning due to failure or the like. In addition, the server connected to the control LAN 110 is also connected to each of the first control LAN 110a and the second control LAN 110b in the same manner as the eight slots described above.

  Each of the eight slots described above is further connected to the maintenance LAN 109. That is, the blade server module inserted into one of these eight slots is connected to the maintenance LAN 109 and can perform data communication via the maintenance LAN 109.

  FIG. 3 is a block diagram showing the configuration of the maintenance server 104. The maintenance server 104 includes a maintenance server control device 120 including, for example, a CPU and its peripheral circuits. The maintenance server control device 120 controls the entire maintenance server 104 by reading and executing a control program stored in advance in the ROM 122 which is a nonvolatile storage medium. This control program uses the DRAM 121, which is the main storage device, as a temporary storage area.

  The maintenance server 104 further includes an HDD 123 that is a large-capacity auxiliary storage device. The HDD 123 stores distance control server data 30 a corresponding to the distance control server 107 and monitoring control server data 30 b corresponding to the monitoring control server 108. The remote server data 30a includes system environment data 31a, application data 32a, and operation data 33a. Similarly, the monitoring control server data 30b includes system environment data 31b, application data 32b, and operation data 33b. Instead of the HDD 123, it is also possible to use a storage device such as an SSD (Solid State Drive) configured by a nonvolatile semiconductor memory, for example.

  The system environment data 31a and 31b include data necessary for starting the blade server module inserted into the empty slots of the first server chassis 1 and the second server chassis 2, such as IP addresses and host names. The application data 32a and 32b are data including application programs executed in the blade server module, application data used by the application programs, and the like, and are data prepared by the manufacturer of this system. The operation data 33a, 33b is user data used by the application program, and includes settings of the application program, for example. That is, the operation data 33a and 33b are data set by the user who uses the system.

  The remote server data 30 a is data necessary for the blade server module to behave as the remote server 107. That is, the system environment data 31a, the application data 32a, and the operation data 33a include a program for realizing the function provided by the remote control server 107 and data required by the program. Similarly, the monitoring control server data 30b is data necessary for the blade server module to behave as the monitoring control server 108.

  The maintenance server 104 includes a first communication unit 124 for connecting to the maintenance LAN 109, a second communication unit 125 for connecting to the first control LAN 110a, and a third communication unit 126 for connecting to the second control LAN 110b. It has. The maintenance server control device 120 performs data communication with nodes (servers and clients) connected to each LAN via these communication units.

  FIG. 4 is a block diagram showing the configuration of the blade server module 111. In the present embodiment, the remote control server 107 and the monitoring control server 108 are each implemented by a blade server module 111 that is hardware having the same configuration. In other words, the remote control server 107 and the monitoring control server 108 have the same hardware, although the programs executed in them are different.

  The blade server module 111 includes, for example, a module control device 130 including a CPU and its peripheral circuits, a DRAM 131 as a main storage device, a ROM 132 as a nonvolatile storage medium, and an HDD 133 as a large-capacity auxiliary storage device. . The module control apparatus 130 controls the entire blade server module 111 by reading and executing a predetermined control program stored in advance in the ROM 132.

  The HDD 133 stores one system environment data 31c, one application data 32c, and one operation data 33c. The HDD 133 further stores version data 50 corresponding to these three data. The version data 50 is data for specifying the contents of the system environment data 31c, the application data 32c, and the operation data 33c. For example, when the contents of the application data 32c are updated, the contents of the version data 50 are also changed. It is updated accordingly. The blade server module 111 includes a first communication unit 134 for connecting to the maintenance LAN 109, a second communication unit 135 for connecting to the first control LAN 110a, and a third communication unit 136 for connecting to the second control LAN 110b. It has. The module control device 130 performs data communication with nodes (servers and clients) connected to each LAN via these communication units.

  Next, the redundancy recovery function when a failure occurs by the maintenance server 104 will be described. In this embodiment, a spare blade server module 111 is inserted into an empty slot of the first server chassis 1 or the second server chassis 2 so that the blade server module behaves as the remote control server 107 or the monitoring control server 108. be able to. Hereinafter, this function will be described.

  FIG. 5 is a flowchart of processing executed by the maintenance server control device 120 and the module control device 130 when the blade server module 111 is inserted into the empty slot. The process of FIG. 5 executed by the maintenance server control device 120 is a process included in the control program stored in the ROM 122 in advance. Similarly, the process of FIG. 5 executed by the module control apparatus 130 is a process included in a control program stored in the ROM 132 in advance. Hereinafter, the process (the process shown on the left side of FIG. 5) executed by the maintenance server control apparatus 120 will be described first, and then the process (the process shown on the right side of FIG. 5) executed by the module control apparatus 130 will be described. .

  First, in step S100, the maintenance server control apparatus 120 receives an insertion notification indicating that the blade server module 111 has been inserted. In response to this, the maintenance server control device 120 determines (selects) whether the blade server module 111 behaves as the monitoring control server 108 or the remote control server 107 in step S110. That is, the role of the blade server module 111 is determined (selected). This determination (selection) of the role is performed, for example, by the operator operating the console client 102 and transmitting information specifying the role to be assigned to the blade server module 111 from the console client 102 to the maintenance server 104. The maintenance server control device 120 determines a role to be assigned to the blade server module 111 according to this information received from the console client 102. The role determination method is not limited to this. For example, the maintenance server 104 may count the number of monitoring control servers 108 and the number of remote control servers 107 and select the smaller number.

  Next, the maintenance server control device 120 receives the version data 50 from the blade server module 111 and temporarily stores it in the DRAM 121 in step S120. The version data 50 includes version information for specifying the system environment data 31c, application data 32c, and operation data 33c stored in the HDD 133. In step S130, the maintenance server control device 120 determines whether or not the system environment data 31c of the blade server module 111 needs to be updated based on the version data 50 stored in the DRAM 121. This determination is performed, for example, when the operator of the console client 102 refers to the above version data 50 and transmits information indicating whether update is necessary from the console client 102 to the maintenance server 104. The maintenance server control device 120 may determine whether or not the update is necessary by a method different from the above method using the information from the operator.

  If it is determined that the system environment data 31c needs to be updated, the process proceeds to step S140. In step S140, the maintenance server control device 120 reads the latest system environment data corresponding to the role determined in step S110 from the HDD 123, and transmits the system environment data to the blade server module 111, thereby storing the system stored in the HDD 133. The environmental data 31c is updated. On the other hand, if it is determined that updating of the system environment data 31c is unnecessary, the process proceeds to step S150 without executing the process of step S140.

  In step S <b> 150, the maintenance server control device 120 determines whether or not the application data 32 c stored in the HDD 133 of the blade server module 111 needs to be updated based on the version data 50 stored in the DRAM 121. For example, if the content of the application data 32c does not correspond to the role determined in step S110 or is different from the latest version stored in the HDD 123, it is determined that the update is necessary. If it is determined that the update is necessary, the process advances to step S 160, and the maintenance server control device 120 reads the latest application data from the HDD 123 and transmits it to the blade server module 111. On the other hand, if it is determined that the update of the application data 32c is not necessary, the process proceeds to step S170 without executing the process of step S160.

  In step S <b> 170, the maintenance server control device 120 determines whether or not the operation data 33 c stored in the HDD 133 of the blade server module 111 needs to be updated based on the version data 50 stored in the DRAM 121. For example, if the contents of the operation data 33c do not correspond to the role determined in step S110, or if it differs from the latest version stored in the HDD 123, it is determined that updating is necessary. If it is determined that the update is necessary, the process advances to step S180, and the maintenance server control device 120 reads the latest operation data from the HDD 123 and transmits it to the blade server module 111. On the other hand, if it is determined that the operation data 33c need not be updated, the process proceeds to step S190 without executing the process of step S180.

  In step S190, the maintenance server control apparatus 120 transmits a restart request to the blade server module 111. In response to this restart request, the blade server module 111 starts to act with the role determined in step S110. That is, for example, it functions as the monitoring control server 108 or the remote control server 107.

  Next, processing executed by the module control apparatus 130 will be described. When the blade server module 111 is inserted into any of the empty slots, the module control device 130 first activates the blade server module 111 in the maintenance mode (preparation mode) in step S200. The maintenance mode is a mode for reconfiguring the blade server module 111. When activated in the maintenance mode, the blade server module 111 is disconnected from the control LAN 110. That is, only the first communication unit 134 operates and the second communication unit 135 and the third communication unit 136 are not operating.

  Next, the module control device 130 transmits an insertion notification indicating that the blade server module 111 has been inserted to the maintenance server 104 in step S210. The blade server module 111 may be configured so that this insertion notification is not sent. In that case, when the operator operates the console client 102, an insertion notification is transmitted from the console client 102 to the maintenance server 104.

  In subsequent step S220, the module control device 130 transmits the version data 50 stored in the HDD 133 to the maintenance server 104. The version data 50 includes version information for specifying the system environment data 31c, application data 32c, and operation data 33c stored in the HDD 133.

  Next, in step S230, the module control apparatus 130 determines whether or not system environment data has been received from the maintenance server 104. If the system environment data is received, the process proceeds to step S240, and the system environment data 31c stored in the HDD 133 is updated with the received system environment data. On the other hand, if the system environment data has not been received, the process proceeds to step S250 without executing the process of step S240.

  In step S250, the module control apparatus 130 determines whether application data has been received from the maintenance server 104. When the application data is received, the process proceeds to step S260, and the application data 32c stored in the HDD 133 is updated with the received application data. On the other hand, if the application data has not been received, the process proceeds to step S270 without executing the process of step S260.

  In step S <b> 270, the module control device 130 determines whether operation data has been received from the maintenance server 104. When the operation data is received, the process proceeds to step S280, and the operation data 33c stored in the HDD 133 is updated with the received operation data. On the other hand, if the operation data has not been received, the process proceeds to step S290 without executing the process of step S280.

  In step S290, the module control apparatus 130 receives the restart request transmitted from the maintenance server 104, and restarts the blade server module 111 in response to this request. In subsequent step S300, the blade server module 111 is activated not in the maintenance mode but in the normal mode. The normal mode is a mode that can act as the role determined in step S110 based on the system environment data 31c, application data 32c, operation data 33c, and the like stored in the HDD 133. Unlike the maintenance mode, This mode is connected to the control LAN 110. At this time, the blade server module 111 enters a standby state if there is an active server with the same role.

According to the system monitoring control system according to the first embodiment described above, the following operational effects can be obtained.
(1) The system monitoring and control system 100 is configured to monitor the power systems with a first remote control server 107a that provides a function of exchanging data with a distant power facility such as a substation. The first monitoring control server 108a that provides a function for performing control, etc., the second distance control server 107b that functions as a redundant system of the first distance control server 107a, and a second function that functions as a redundant system of the first monitoring control server 108a And the four monitoring server 108b, and a maintenance server 104 for managing these four blade server modules 111. When the maintenance server 104 detects that the spare blade server module 111 has been added by the insertion notification, the maintenance server 104 selects either the remote control server 107 or the monitoring control server 108 and functions as a redundant system of the selected server. Thus, the spare blade server module 111 is reconfigured. As described above, in order to make the remote control server 107 and the monitoring control server 108 have a triple redundant configuration, normally two blade server modules 111 are required. According to the present invention, the spare blade server By preparing only one module, it can be used as the third redundant system of the remote control server 107 and the monitoring control server 108 at any time. Therefore, the reliability of the system can be improved while suppressing the installation space and cost of the device.

(2) The maintenance server 104 includes the remote server data 30a necessary for the spare blade server module 111 to provide the functions of the remote server 107, and the spare blade server module 111 functions as the monitoring control server 108. The HDD 122 stores the monitoring control server data 30b necessary for providing the data. When the remote control server 107 is selected, the remote control server data 30a is transmitted to the spare blade server module 111. When the supervisory control server 108 is selected, the supervisory control server data 30b is transmitted to the spare blade server module 111. The spare blade server module 111 is reconfigured. Since it did in this way, it becomes possible to make it operate | move as a standby system of arbitrary servers whatever the state of the spare blade server module 111 is.

(3) The remote server data 30a necessary for the spare blade server module 111 to provide the functions of the remote server 107 includes system environment data 31a, application data 32a, and operation data 33a. The monitoring control server data 30b necessary for the spare blade server module 111 to provide the function of the monitoring control server 108 includes system environment data 31b, application data 32b, and operation data 33b. Since it did in this way, the range of the data which the manufacturer which provides a system, and the user who uses a system handle can be clarified, and the accident which changes the data which should not be changed accidentally can be prevented. Further, for example, when only application data is changed, only application data is transmitted to the blade server module 111, and it is not necessary to transmit operation data or the like, so that the time required for reconfiguration of the blade server module 111 is reduced. Can do.

(4) The spare blade server module 111 is capable of receiving and storing the remote control server data 30a and the monitoring control server data 30b transmitted from the maintenance server 104 during reconfiguration by the maintenance server 104. When the remote control server data 30a is received in response to receiving and storing one of the remote control server data 30a and the monitoring control server data 30b, the remote control server 107 functions. When the monitoring control server data 30b is received, the functions of the monitoring control server 108 are restarted in the normal modes that can be provided. Since it did in this way, about the added blade server module 111, the reconfiguration | reconstruction by the server 104 for maintenance can be performed reliably.

(5) The first server chassis 1 has a first slot 11 in which the first monitoring control server 108a is inserted, and a second slot 12 in which the first distance control server 107a is inserted. The second server chassis 2 has a fifth slot 21 in which the second monitoring control server 108b is inserted, and a sixth slot 22 in which the second distance control server 107b is inserted. The first server chassis 1 further has a third slot 13 into which a spare blade server module 111 can be inserted. Similarly, the second server chassis 2 has a seventh slot 23 into which a spare blade server module 111 can be inserted. Have. Since it did in this way, even if one server chassis itself fails, it can avoid that the whole system stops because of it. Further, the redundancy of the monitoring control server 108 and the remote control server 107 can be recovered by inserting the spare blade server module 111 into an empty slot of the server chassis that has not failed.

  The following modifications are also within the scope of the present invention, and one or a plurality of modifications can be combined with the above-described embodiment.

(Modification 1)
In the embodiment described later, the second distance control server 107b and the second monitoring control server 108b operate as standby systems for the first distance control server 107a and the first monitoring control server 108a. That is, while the first distance control server 107a and the first monitoring control server 108a are operating normally, the second distance control server 107b and the second monitoring control server 108b are in a standby state. The present invention is not limited to such an embodiment. For example, the second remote control server 107b and the second monitoring control server 108b may always be in an active (operating) state.

(Modification 2)
The roles that can be assigned to the blade server module 111 are not limited to the two described above (distance server 107 and monitoring control server 108). For example, it is possible to assign three or four roles. In order to do this, a plurality of types of data such as system environment data may be stored in the HDD 122 of the maintenance server 104. For example, if three or four types of servers are duplicated, each of which has a triple or quadruple redundant configuration requires a huge amount of cost and installation space, but the present invention is applied. Thus, if only one spare blade server module 111 is prepared, it is possible to ensure the same reliability as when a triple redundant configuration is adopted for all of the servers.

  As long as the characteristics of the present invention are not impaired, the present invention is not limited to the above-described embodiments, and other forms conceivable within the scope of the technical idea of the present invention are also included in the scope of the present invention. .

DESCRIPTION OF SYMBOLS 100 ... System monitoring control system, 101 ... Information system server, 102 ... Console client, 103 ... System monitoring server, 104 ... Maintenance server, 105 ... I / F server, 106 ... Support server, 107 ... Remote control server, 108 ... monitoring control server, 109 ... LAN for maintenance, 110 ... control LAN, 111 ... blade server module

Claims (6)

A first blade server providing a first function;
A second blade server providing a second function different from the first function;
A third blade server functioning as a redundant system of the first blade server;
A fourth blade server functioning as a redundant system of the second blade server;
A blade server system comprising a management device for managing the first to fourth blade servers,
When the management device detects that a spare blade server has been added to the blade server system, the management device selects either the first blade server or the second blade server and functions as a redundant system of the selected blade server The blade server system is characterized by reconfiguring the spare blade server. The blade server system according to claim 1,
The management device stores first data necessary for the spare blade server to provide the first function and second data necessary for the spare blade server to provide the second function. The first data is transmitted when the first blade server is selected, and the second data is transmitted when the second blade server is selected. A blade server system, wherein a spare blade server is reconfigured. The blade server system according to claim 2,
Each of the first data and the second data includes system environment data including setting data necessary for starting the spare blade server, application data including a user program executed in the spare blade server, and the user program Blade server system characterized by comprising operational data read and written by The blade server system according to claim 3,
The spare blade server is activated in a maintenance mode in which the first data and the second data transmitted from the management device can be received and stored in the reconfiguration by the management device, and the first data and the In response to receiving and storing any one of the second data, the first function is received when the first data is received, and the second function is received when the second data is received. A blade server system that is restarted in a normal mode that can be provided. In the blade server system according to any one of claims 1 to 4,
A first housing having a first slot into which the first blade server is inserted, and a second slot into which the second blade server is inserted;
A third slot in which the third blade server is inserted, and a second housing having a fourth slot in which the third blade server is inserted,
At least one of the first casing and the second casing includes a fifth slot into which the spare blade server can be inserted. In the blade server system according to any one of claims 1 to 4,
The management apparatus includes an input unit that inputs a selection operation of either the first blade server or the second blade server from the outside, and the first blade server and the first blade according to an input to the input unit A blade server system, wherein either one of two blade servers is selected.

Images