JP2009032120A - Process history database system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a process history database system capable of preventing generation of time-out due to the delay of client's response during the updating of statistical information. <P>SOLUTION: Two history servers for collecting and distributing process variables as process history in time series and a client 40 for executing display, analysis and other data processing by using the data of the process history are connected by network communication. The two history servers have a duplex configuration of a master and a slave, respectively independently collect process data and store the process history, and asynchronously update the statistical information of a database. The history servers for acquiring the process history are switched so that the client 40 acquires the process history from the master history server 21 usually, and in updating the statistical information by the master history server 21, the process history is acquired from the slave history server 22. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a process history database system including a history server that collects and distributes process variables as process history in time series.

  2. Description of the Related Art A process history database system including a history server that collects and distributes process variables as a process history in time series has been conventionally used for controlling various plants (see, for example, Patent Document 1).

Such a process history database system normally performs display, analysis, and other data processing using two history servers including a master and a backup that collect and distribute process variables as a process history in time series, and data of the process history. The client is connected via network communication.
JP 2005-182681 A

In such a process history database system, when general-purpose database software is used, it is necessary to update so-called statistical information for the optimization processing of the database, and the amount of table updates reaches the change threshold. Then, when the update of statistical information is automatically started, a client response may be delayed while the statistical information is being updated (up to about 15 minutes), and a timeout may occur (see Non-Patent Document 1).
http://support.microsoft.com/kb/195565/en (3 pages, Fig. 2)

  However, if automatic updating of statistical information is stopped, database performance will eventually deteriorate.

  Since the conventional process history database has a small hard disk capacity (for example, about 40 gigabytes (GB)), the influence of the update of statistical information is small, but the hard disk capacity is large (for example, 100 GB or more). With the increase in the size of the database that expands to (3), there is a concern that the above-mentioned client response at the time of automatic update of statistical information will be delayed and the frequency of occurrence of timeout will increase in the acquisition of actual process history.

  To deal with such problems, it is possible to reduce the size of the process history database so that response delays are not noticeable while updating statistical information, or to optimize the database during regular maintenance. Is an emergency that will fail if the capacity is further increased, and could not be a fundamental measure.

  An object of the present invention is to provide a process history database system in which a client response is not delayed during a statistical information update and a timeout does not occur.

The process history database system according to claim 1 of the present invention comprises:
Two history servers that collect and distribute process variables as process history over time;
In a process control system in which a client that performs display, analysis and other data processing using the process history data is connected by network communication,
The two history servers have a dual master / slave configuration, and the two history servers independently collect process data and store the process history of the process data, respectively, and the database statistical information is asynchronous with each other. It is supposed to be updated to
Normally, the client acquires the process history from the master history server, and when the master history server updates the statistical information, the process history acquisition destination so that the process history is acquired from the slave history server. It is characterized by switching.

  In the process history database system according to claim 1 of the present invention, when the history servers schedule the statistical information of the databases asynchronously with each other, and the master history server updates the statistical information, the process from the slave history server is performed. By switching the acquisition source so that the history is acquired by the client, the response to the client is not delayed or a timeout does not occur while updating the statistical information.

The process history database system according to claim 2 of the present invention is the process history database system according to claim 1,
The two history servers are shared by a node connected to a network by a personal computer, and are adjusted to each other so that update of statistical information does not overlap between the master history server and the slave history server.

  Since the process history database system according to the second aspect of the present invention has such a configuration, the operation based on the first aspect can be more reliably achieved.

The process history database system according to claim 3 of the present invention is the process history database system according to claim 1 or 2,
The client is configured such that the client program switches the acquisition destination of the process history to the slave history server by looking at the update status of the statistical information of the master history server, or the client program executes the environment management program. The environment management program is called to execute one of the process history acquisition destinations.

  In the process history database system according to claim 3 of the present invention, the client may view the update status of the master history server in this way, and the acquisition destination of the process history by calling the environment management program and checking the update status of the statistical information The degree of freedom in design increases. In addition, in the latter case, the load on the client program is reduced and management becomes easier.

A process history database system according to claim 4 of the present invention is the process history database system according to any one of claims 1 to 3,
The two history servers can selectively switch a role as a master history server and a role as a slave history server.

  Since the process history database system according to the fourth aspect of the present invention has such a configuration, the master history server and the slave history server can be freely selected and switched appropriately.

  According to the present invention, it is possible to provide a process history database system in which a client response is delayed during a statistical information update and a timeout does not occur.

  Hereinafter, a process history server database system according to an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a system schematic configuration diagram to which a process history server database system 1 according to an embodiment of the present invention is applied. This system schematic configuration diagram shows a history of plant data of process data such as a PV value in process control using a field device 10 such as a differential pressure transmitter 11, an electromagnetic flow meter 12, and a gas component analyzer 13. Data is accumulated individually by the two history servers 20, the master history server 21 and the slave history server 22. The control history can be monitored by the client.

  For example, in this system, the field device 10 is connected to the control LAN via the process controllers 31 and 32, and the programmable controllers 33 and 34 are connected to the LAN via the gateway 35 that enables communication between different protocols. It has become so. A master history server 21 and a slave history server 22 are connected to the LAN, and other servers 23 are connected to the LAN. The master history server 21, slave history server 22, or other server 23 is connected to each client 40 (41, 42, 43) via Ethernet.

  Both the master history server 21 and the slave history server 22 each store the process history collected from the field device 10, and the history servers 21 and 22 schedule the accumulated statistical information asynchronously with each other.

  Each client 41, 42, 43 normally acquires the process history from the master history server 21, but acquires the process history from the slave history server 22 when the master history server 21 updates the statistical information. As described above, the process history acquisition destination is switched.

  Next, features of the present embodiment will be described in more detail with reference to FIG. Each of the master history server and the slave history server shown in FIG. 2 can store the process history continuously in the database and perform a statistical information update service.

  Note that the upper left of FIG. 2 shows a client PC 1 that performs form management based on, for example, a continuous history database. The client PC 1 directly monitors the statistics information update service of the master history server, and switches to acquire the process history from the slave history server while preparing or updating the statistics information of the master history server in the procedure described later. ing.

  On the other hand, a client PC 2 having functions of a remote trend, an open form, and an environment management program (SE) is provided in the upper right of FIG. In this client PC 2, the remote trend program usually obtains the process history from the master history server, but the master history server is preparing or updating the statistical information from the environment management program according to the procedure described later. If confirmed, the process history acquisition destination is switched from the master history server to the slave history server. Similarly, for the open form program, if the environment management program confirms that the master history server is preparing or updating statistical information, the process history acquisition destination is switched from the master history server to the slave history server. ing.

  Next, the statistical information update execution procedure of the process history server database system having the configuration shown in FIG. 2 will be described.

  Along with the function of the statistical information update service operating on the master history server side, processing is performed in the following procedure.

  As the first step, before the master history server updates the statistical information, it confirms the statistical information update status of the slave history server, and if the slave history server is not preparing or updating the statistical information, The status of updating statistical information is set in preparation (see (1) in FIG. 2).

  Subsequently, as a second step, the form program of the client PC 1 directly confirms the update status of the statistical information of the master history server, and the remote trend program and open form program of the client PC 2 execute the master history server via the environment management program. Check the update status of the statistical information (see (2) in Fig. 2).

  Subsequently, as a third step, when the master history server changes the status of statistical information update during preparation, the statistical information is updated after 5 minutes (this time value is an example and can be changed) ( (See (3) in FIG. 2).

  Subsequently, as a fourth step, when the master history server is preparing for or updating statistical information, the form program of the client PC 1, the remote trend program of the client PC 2, and the open form program acquire process history. The history server connected to is switched from the master history server to the slave history server (see (4) in FIG. 2).

  Here, when the master history server finishes updating the statistical information, the status of the statistical information update is ended.

  Subsequently, as a fifth step, the form program of the client PC 1 directly confirms the update status of the statistical information of the master slave server, and the remote trend program and open form program of the client PC 2 execute the master slave server via the environment management program. After confirming the end of the update status of the statistical information (see (5) in FIG. 2), as the sixth step, the form program of the client PC 1, the remote trend program of the client PC 2, and the open form program acquire the process history. Therefore, the connected history server is switched again from the slave history server to the master history server (see (6) in FIG. 2).

  3 and 4 are statuses showing the update status of statistical information in the master history server and the slave history server. The status transitions are as shown in FIG. 3, and the status of each transition is as shown in FIG.

  In this status transition, READY (0) indicates a situation in which statistical information update processing is not performed. In addition, in the case of INITIATE (1), a state is shown in which the statistical information is being updated, that is, in the case of this embodiment, waiting for 5 minutes. In the case of UPDATING (2), it indicates a situation in which statistical information is being updated. Further, in the case of COMPLETE (3), the status of the update of statistical information is indicated. The status indicating the update of the statistical information is monitored by the above-mentioned form program of the client PC 1, the remote trend program of the client PC 2, and the open form program, and when the status status is INITIATE (1) and UPDATING (2) Since the master history server is preparing for or updating the statistical information, the history data is switched to be acquired from the slave history server.

  In addition, as in the above-described embodiment, it may be directly monitored whether the form program of the client PC 1 is preparing or updating the statistical information update service of the master history server, such as a remote trend program or an open form program of the client PC 2. In addition, it may be confirmed whether the master history server is preparing or updating the statistical information through an environment management program that it has. However, when there are many client programs, the latter has an advantage that the work load of each program of the client is lightened and can be easily managed because it can be centrally managed by the environment management program.

  As described above, the process history server database system according to the present invention has the configuration as described above, whereby the history servers schedule the statistical information of the databases asynchronously with each other, and the master history server updates the statistical information. In this case, by switching the acquisition destination so that the client acquires the process history from the slave history server, the response to the client is not delayed or a timeout does not occur during the update of the statistical information.

  The master history server and slave history server described above can serve as a master history server and as a slave history server instead of either one serving as a master history server permanently and the other serving as a slave history server. It may be configured to selectively switch.

  Further, the process history server database system according to the present invention described above is not limited to accumulation of statistical information as long as it has an asynchronous history server, for example, and the database system is equipped within the scope of the present invention. Needless to say, the present invention can be applied to other optimization processes.

It is a system schematic block diagram to which a process history server database system according to an embodiment of the present invention is applied. FIG. 2 is a system configuration diagram for explaining the operation of the process history server database system of the present invention shown in FIG. 1. It is a transition diagram which shows the status regarding the statistical information update in the master history server of this invention, and a slave history server. It is a table | surface explaining the specific content of each status corresponding to FIG.

Explanation of symbols

1 Process history server database system
DESCRIPTION OF SYMBOLS 10 Field apparatus 11 Differential pressure transmitter 12 Electromagnetic flow meter 13 Gas component analyzer 20 History server 21 Master history server 22 Slave history server 23 Other server 31, 32 Process controller 33, 34 Programmable controller 35 Gateway 40 (41, 42, 43) Client

Claims (4)

Two history servers that collect and distribute process variables as process history over time;
In a process control system in which a client that performs display, analysis and other data processing using the process history data is connected by network communication,
The two history servers have a dual master / slave configuration, and the two history servers independently collect process data and store the process history of the process data, respectively, and the database statistical information is asynchronous with each other. It is supposed to be updated to
Normally, the client acquires the process history from the master history server, and when the master history server updates the statistical information, the process history acquisition destination so that the process history is acquired from the slave history server. Process history database system, characterized by switching.   The two history servers are shared by nodes connected to a network by a personal computer, and are adjusted to each other so that update of statistical information does not overlap between the master history server and the slave history server. The described process history database system.   The client is configured such that the client program switches the acquisition destination of the process history to the slave history server by looking at the update status of the statistical information of the master history server, or the client program executes the environment management program. 3. The process history database system according to claim 1, wherein the environment management program is called to switch between acquisition destinations of the process history.   The process history database according to any one of claims 1 to 3, wherein the two history servers can selectively switch a role as a master history server and a role as a slave history server. system.

Images