JP2508589B2 - Server operation method - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a server operating system in a server / client type integrated office system, and more particularly to a server operating system capable of reducing the load on the system.

[0002]

2. Description of the Related Art A server that requests various services from a client to a server, such as sending a document, retrieving an incoming mail, and searching a document, and executing the service requested by the client in a server process.・ Client-type integrated office systems have been put to practical use.

By the way, in the conventional server / client type integrated office system, one server is configured to execute a plurality of types of services. When a client requests a service, the server process is a part of it. The service was executed using only (the part that executes the requested service).

[0004]

As described above, conventionally, since one server is configured to execute a plurality of types of services, there are the following problems. In other words, it is generally unlikely that all services will be used with the same frequency. Therefore, when multiple types of services are executed by one server, the part of the server that is used less frequently is given to the system. There was a problem that the load increased.

An object of the present invention is to provide a server operation system which can reduce the load on the system.

[0006]

In order to achieve the above object, the present invention provides: (A) In an integrated office system capable of simultaneously activating a plurality of server processes, a plurality of servers each executing one type of service. A rewritable server management table in which control information of each server is set, a communication management unit that requests allocation of a server corresponding to the service request in response to a service request from a client, and the communication management A server management unit is provided for controlling the server requested to be allocated by the unit according to the corresponding control information set in the server management table.

Further, in order to achieve the above object, the present invention (B) is an integrated system capable of simultaneously starting a plurality of server processes.
Each service of the Fiss system
A plurality of servers that run Sir corresponding to respective servers
Rewritable, which sets the upper limit of the number of boot processes
Server management table and service from client
Of the server corresponding to the service request in response to the service request.
Ri and communication management unit that requests against said only when the starting number of server processes corresponding to the server allocation is requested from the communication management unit does not exceed the upper limit value set in the server management table assignments And a server management unit for activating a server process corresponding to the requested server and allocating it to the requesting client .

Further, in order to achieve the above object and to reduce the time required to start the server process, the present invention provides (C) an integrated option capable of simultaneously starting a plurality of server processes.
Each service of the Fiss system
A plurality of servers that run Sir corresponding to respective servers
The maximum number of processes that can be started and the hold time are set.
Rewritable server management table and client
Responding to a service request from a client
A communication management unit for requesting allocation of a server that, one of which cases the server process not running a service in the booted server process corresponding to the server allocation is requested from the communication management unit is present Are assigned to the requesting client, and if there are no server processes that are not executing services among the started server processes, only if the number of started processes does not exceed the upper limit set in the server management table. The server process corresponding to the server for which the allocation is requested is started and assigned to the requesting client. If there is no started server process, the server process is started and assigned to the requesting client, and the service is executed. The hold time set in the server management table has passed since the end Server tube to terminate the Bapurosesu
It has a science department.

In addition to achieving the above object, the present invention accurately knows the usage status of each server, and sets the upper limit value of the number of server process activations and the hold time set in the server management table to the usage status of each server. (D) a time storage table in which the service start time and service end time of each server are stored in order to be combined.
A statistical processing unit that creates statistical information based on the contents of the time storage table and a setting processing unit that rewrites the contents of the server management table are provided.

[0010]

In the configuration (A), a plurality of servers are provided, each of which executes one type of service, and the control information of each server is set in the rewritable server management table.

When there is a service request from the client,
The communication management unit requests the server management unit to allocate the server corresponding to the service request, and the server management unit controls the server requested to be allocated according to the control information set in the server management table.

In the configuration of (B), the upper limit of the number of server processes to be started corresponding to each server is set in the server management table as control information of each server.

When the client makes a service request, the communication management unit requests the server management unit to allocate a server corresponding to the service request, and the server management unit responds to the allocation request and responds to the requested server. If the number of server processes to be started does not exceed the upper limit set in the server management table, the server process corresponding to the server for which allocation is requested is started and allocated to the requesting client.

In the configuration of (C), the upper limit value and the holding time of the number of server processes corresponding to each server are set in the server management table as control information of each server.

When a client makes a service request,
The communication management unit requests the server management unit to allocate the server corresponding to the service request.

If there is a server process that is not executing a service among the started server processes corresponding to the server requested to be allocated by the communication management unit, one of them is the request source. If a server process that is assigned to a client and has not started is among the server processes that have been started, the requested server is assigned only if the number of started processes does not exceed the upper limit set in the server management table. Start the server process corresponding to and allocate it to the requesting client. If there is no started server process, start the server process and allocate it to the requesting client. Terminate the server process for which the hold time set in the table has elapsed.

In the configuration of (D), the statistical processing unit causes each server to perform a predetermined period of time based on the contents of the time storage table in which the service start time and service end time of each server are stored. Statistical information such as the number of times the server is used, the average value of the difference between the service end time and the service start time of each server, the average value of the service execution time of each server once, and the number of times each server is used per day.

Then, the system administrator calculates the upper limit value and the holding time of the number of startup processes of each server according to the usage status of each server based on the statistical information created by the statistical processing unit, and calculates each calculated value. The upper limit value and the hold time of the number of server startup processes are set in the server management table using the setting processing unit.

[0019]

Embodiments of the present invention will now be described in detail with reference to the drawings.

FIG. 1 is a block diagram of an embodiment of the present invention. The figure shows a case where the present invention is applied to an integrated office system, which is composed of a server system 1 and a plurality of clients 2-1 to 2-m.

The server system 1 includes a communication management unit 11 and
Server management unit 12 and server management tables 13 and 14
And a time storage table 15, a setting processing unit 16, an input unit 17, a statistical processing unit 18, and a screen display unit 19. Further, in FIG. 1, S1 'to Sn' indicate server processes corresponding to the servers having the server names S1 to Sn.

FIG. 2 is a diagram showing an example of the contents of the server management table 13, in which the server names of all the servers included in the server system 1 and the upper limit value of the number of startup processes of each server,
The hold time of each server (the time from the end of the service of the corresponding server process to the end of the server process itself) is set.

In the example of FIG. 2, the server system 1 has n servers having server names S1 to Sn, and server names S1 and S are provided.
The upper limit of the number of startup processes of each server of 2, ..., Sn is 3, 1, ..., 1 and the hold time is 1: 0, respectively.
0, 0:30, ..., 0:15.
In this embodiment, the server with the server name S1 only sends the document and the server with the server name S2 only retrieves the document, such that the servers with the server names S1 to Sn provide different types of services. Is configured to run. Further, the total upper limit value of the number of startup processes for each server is set to the same value as the maximum number of processes that the server system 1 can simultaneously start.

FIG. 3 is a diagram showing an example of the contents of the server management table 14, in which the server names S1 to Sn of each server are set by the number of upper limit values of the corresponding server names set in the server management table 13. At the same time, startup information indicating whether or not the server process corresponding to each server name has been activated, and service execution information indicating whether or not the server process corresponding to each server name is executing a service are set.

FIG. 4 is a diagram showing an example of the contents of the time storage table 15, in which a server name, a service start time, and a service end time are set.

The tables 13 to 15 shown in FIGS.
Exists as a file, for example, and the data can be stored in a buffer (not shown) in the server management unit 12 as needed.

Next, the operation of this embodiment will be described.

First, at the time of introduction, the system administrator uses the input unit 17 to set the server name set in the server management table 13,
Enter the upper limit value for the number of started processes and the hold time. At this time, the system administrator increases the upper limit value and the hold time value to be input, as the server is expected to be used more frequently.

The server name, the upper limit of the number of startup processes, and the hold time input from the input unit 17 are set in the server management table 13 by the setting processing unit 16.

After that, when the system manager gives an instruction to start the operation, the communication manager 11 and the server manager 12 are activated.

As a result, the communication management unit 11 enters a waiting state waiting for service requests from the clients 2-1 to 2-m.

When started, the server management unit 12 first refers to the server management table 13 and acquires each server name set in the server management table 13 and the corresponding upper limit value.

After that, the server management unit 12 sets the acquired server names in the server management table 14 by the number indicated by the acquired upper limit value. In the example of FIG. 2, server names S1 and S
Since the upper limit values corresponding to 2 are “3” and “1”, the server management unit 12 sets three server names S1 and one server name S2 in the server management table 14.

When the setting of the server name is completed, the server management unit 12 sets all the start information indicating whether or not the server process of the server management table 14 has been started to “not yet” indicating that the server process is not started and the service is being executed. The service execution information indicating whether or not all is set to "not executed" indicating that the service has not been executed.

Next, the client 2-i (1≤i≤m)
The operation when a service request is sent from will be described.

When requesting a service from the server, the client 2-i sends a service request including the type of service requested to the communication management unit 11.

When a service request is sent from the client 2-i, the communication management unit 11 identifies the type of service requested and assigns a server corresponding to the type of service to the server management unit 12. Request.

Now, for example, when the service request from the client 2-i is a request for sending a document,
The communication management unit 11 requests the server management unit 12 to allocate the server with the server name S1, and when the service request from the client 2-i requests a document search, the communication management unit 11 11 requests the server management unit 12 to allocate the server with the server name S2.

When the server management unit 12 is requested by the communication management unit 11 to allocate a server, the server management unit 12 performs the following processing.

Now, for example, the server name Sj (1≤j≤n)
If the server allocation is requested, the server management unit 12 first refers to the server management table 14 and investigates the usage status of the server with the server name Sj.

Then, when the server process Sj 'corresponding to the server having the server name Sj is not activated at all (all the activation information set in association with the server name Sj in the server management table 14 is "not yet". If)
Activates one server process Sj ′ and sets one of the activation information set corresponding to the server name Sj in the server management table 14 to “completed”.

Thereafter, the server management unit 12 notifies the communication management unit 11 that the service can be executed, and indicates the activated server process Sj 'as information indicating the server process to be allocated to the requesting client 2-i. Information is passed, and further, among the service execution information of the server management table 14, the one corresponding to the start information changed to "completed" is set to "running".

Further, when the server process Sj 'corresponding to the server name Sj is activated (in the activation information set corresponding to the server name Sj in the server management table 14, "completed" is indicated. (If there is one), it is determined whether or not some of the activated server processes Sj ′ are not executing the service.

Then, the activated server process S
If some j'do not execute the service (the start information is "completed" and the service execution information is "not executed" in the set of the start information and the service execution information corresponding to the server name Sj. "When there is a message"), the server management unit 12 notifies the communication management unit 11 that the service can be executed, and at the same time sets the above conditions as information indicating a server process to be allocated to the requesting client 2-i. Information indicating one of the satisfied server processes Sj ′ is passed, and further, of the service execution information of the server management table 14,
The one corresponding to the server process Sj ′ assigned to the requesting client 2-i is set to “in execution”.

The server process S that has been activated
If all j's are executing services (server name Sj
If some of the startup information corresponding to the above is "completed", and the service execution information corresponding to the completed startup information and the corresponding service execution information are all "running"), The management unit 12 determines whether or not the number of started server processes Sj ′ corresponding to the server name Sj has reached the upper limit set in the server management table 13.

If it is determined that the upper limit has not been reached (if there is "not yet" in the startup information corresponding to the server name Sj in the server management table 14), it is determined that the server has the server name Sj. One corresponding server process Sj ′ is activated, and one of the activation information corresponding to the server name Sj that is “not yet” is changed to “completed”.

Thereafter, the server management unit 12 notifies the communication management unit 11 that the service can be executed, and indicates the activated server process Sj 'as information indicating the server process to be assigned to the requesting client 2-i. Information is passed, and further, among the service execution information of the server management table 14, the one corresponding to the start information changed to "completed" is set to "running".

If it is determined that the number of server processes Sj 'corresponding to the server name Sj has started up to the upper limit value set in the server management table 13, the server management unit 12 executes the server processes Sj'. Waiting until the service being performed in any one of them is completed, and when the service is completed, the communication management unit 11 is notified that the service can be executed, and information indicating the server process Sj ′ in which the service is completed. Is passed, and the corresponding service execution information in the server management table 14 is changed to “in execution”.

When notified by the server management unit 12 of the server process Sj 'to be allocated to the requesting client 2-i, the communication management unit 11 passes the data received from the client 2-i to the server process Sj'.

The server process Sj 'starts the service according to the data. At that time, the server process S
j ′ passes the start time of the service to the server management unit 12. Further, the server process Sj ′ passes the service end time to the server management unit 12 when the execution of the service ends.

The server management unit 12 uses the server process S.
When the service start time and the service end time are passed from j ′, the server name Sj corresponding to the server process Sj ′,
The service start time and service end time are added to the time storage table 15.

When the service end time is passed from the server process Sj ′, the server management unit 12 selects the server end time from the service execution information set in the server management table 14 corresponding to the server name Sj. The service execution information corresponding to the server process Sj ′ that has output is set to “not executed”. Further, even after the hold time of the server having the server name Sj set in the server management table 13 has elapsed since the service execution information was set to "not executed", the client was not assigned to the server process Sj '. In this case, the server process Sj ′ is terminated, and the corresponding activation information in the server management table 14 is changed to “not yet”. In this way, even if the service by the server process Sj ′ is terminated, the server process Sj ′ is not terminated until the hold time elapses, and when a service request is made from the client, the server process Sj ′ is terminated. By allocating to the client, the time required to start the server process can be reduced.

Next, the operation when the system administrator confirms the operating state of the system and changes the upper limit value and the holding time in the server management table 13 according to the operating state of the system will be described.

The system administrator regularly inputs a statistical information creation instruction from the input unit 17 (eg, one week, one day).

When the setting processing section 16 receives a statistical information creation instruction from the input section 17, the setting processing section 16 transfers it to the statistical processing section 18 and displays the contents of the server management table 14 on the screen display section 19.

When the statistical processing unit 18 receives the instruction to create the statistical information, the statistical processing unit 18 uses the contents of the time storage table 15 to determine the number of times each server has been used during a certain period (for example, the last week and the previous week), Average value of one server service execution time (for example, 30 minutes / time), average number of times each server is used for a certain period of time (for example, 10 times / day), and next service after the service of each server is terminated Statistical information such as the average value of the time (eg, 30 minutes) until the start is created and displayed on the screen display unit 19.

Based on the statistical information displayed on the screen display section 19 and the contents of the server management table 14, the system administrator determines the upper limit value and the holding time according to the usage status of each server.

For example, the upper limit value and the holding time are increased for a server whose number of times of use in the last week has increased compared with the number of times of use for the previous week, and conversely, the upper limit value and a hold time for a server whose number of times of use decreases. Obtain the upper limit value and the hold time that are suitable for the usage status of each server by reducing the time. However, if the upper limit value for a certain server is increased so that the total upper limit value of the number of startup processes of each server does not exceed the number of server processes that the server system 1 can execute at the same time, It will be necessary to reduce the upper limit for the server.

Then, when the upper limit value and the holding time according to the usage status of each server are obtained, the system administrator can enter the input section 17
From, enter the server name of the server for which you want to change the maximum value, hold time, the new maximum value, and the hold time.

When the above-mentioned information is input from the input unit 17, the setting processing unit 16 changes the upper limit value and the holding time set in the server management table 13 accordingly.

In the above-described embodiment, the system administrator activates the statistical processing section 18 to create statistical information, and the system administrator creates the statistical information by using the server management table 13 based on the created statistical information. The upper limit value and the holding time are changed, but it goes without saying that the processes performed by the system administrator may be automatically performed.

[0062]

As described above, according to the present invention, each server executes one type of service, and a server management table in which control information for controlling each server is set is provided. Each server is controlled, and the size of the server can be reduced compared to the conventional method in which one server executes multiple services, so the load on the system is reduced. There is an effect that can be.

Further, according to the present invention, since the upper limit value of the number of server processes to be started corresponding to each server is set as the control information of the server, the operation of the server according to the usage status of each server can be performed. There is also an effect that becomes possible.

Further, according to the present invention, the hold time of each server is set in the server management table, and even if the service ends, the server process is not ended until the hold time elapses, and the service request is issued during that time. In that case, the service can be executed without starting the server process, which has the effect of reducing the time required to start the server process.

Further, according to the present invention, a time storage table in which the service start time and the service end time of each server are stored, and statistical information such as the number of times each server is used in a certain period based on the contents of the time storage table. Since it has a statistical processing unit that creates it, it is possible to accurately grasp the usage status of each server, and as a result, the usage status of each server can be accurately determined based on the upper limit value and hold time set in the server management table. There is also an effect that can be reflected.

[Brief description of drawings]

FIG. 1 is a block diagram of an embodiment of the present invention.

FIG. 2 is a diagram showing an example of contents of a server management table 13.

FIG. 3 is a diagram showing an example of contents of a server management table 14.

FIG. 4 is a diagram showing an example of contents of a time storage table 15.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Server system 11 ... Communication management part 12 ... Server management part 13 ... Server management table 14 ... Server management table 15 ... Time storage table 16 ... Setting processing part 17 ... Input part 18 ... Statistics processing part 19 ... Screen display part S1 ' ~ Sn '... Server process 2-1 to 2-m ... Client

Claims (5)

(57) [Claims] 1. A plurality of servers each executing one type of service, a rewritable server management table in which control information of each server is set, and the service request in response to a service request from a client. A communication management unit for requesting allocation of a server corresponding to the above, and a server management unit for controlling the server requested by the communication management unit for allocation according to corresponding control information set in the server management table. Server operation method characterized by. 2. Each executes one type of service
Multiple servers and server processes corresponding to each server
Rewritable server with the upper limit of the number of startup
Management table and the service in response to the service request from the client
Communication management unit that requests the allocation of the server corresponding to the request
If the server corresponding to the allocation is requested the server only if not exceeding the upper limit number of active server process corresponding to the server allocation is requested from the communication management unit is set to the server management table A server management unit that starts a process and assigns it to the requesting client
A server operation method characterized by having . 3. Each executes one type of service
Upper limit of the number of startups of multiple servers and server processes corresponding to each server
And rewritable server management, which is set with the hold time
Tables and services in response to service requests from clients
Communication management unit that requests the allocation of the server corresponding to the request
If, assigned to the client requesting the one of which if the server process is not running a service in the booted server process corresponding to the server allocation is requested from the communication management unit is present, start If there is no server process that is not executing a service among the already-existing server processes, the allocation is requested only if the number of startup processes does not exceed the upper limit value set in the server management table. Starts the server process and allocates it to the requesting client, and if there is no started server process, starts the server process and allocates it to the requesting client, and sets it in the server management table after the execution of the service ends. support to terminate the server process that the hold time has been elapsed
A server operation method characterized by having a server management unit . 4. A time storage table in which the service start time and service end time of each server are stored, a statistical processing unit that creates statistical information based on the contents of the time storage table, and the contents of the server management table. The server operation method according to claim 3, further comprising a setting processing unit that rewrites 5. The statistical information created by the statistical processing unit includes:
Number of times each server was used in a given period of time, average value of the difference between the service end time and service start time of each server, average value of one service execution time of each server, number of times each server was used per day 5. The method according to claim 4, wherein
Described server operation method.