JP3064319B2 - High reliability online system - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a highly reliable online system in a computer application system having a database.

[Prior Art] In the conventional art, the following measures were taken to improve the reliability of the online system (see FIG. 2).

The external storage devices (13 and 14) for storing database records and updating them online are doubled, and when it is necessary to update the database records, the database records (15 and 14) in each external storage device are required. And 16)
Is updated () from the central processing unit (12) at the same time. Always store the same database record in another external storage device.
Double external storage device (13 or 14)
Even in the event of a failure, a database record on another external storage device can be used.

In addition, backup the database to an external storage device (17) that is not connected online (),
By storing the data in the same calculation center (11) as the online system or in another warehouse or calculation center (18), the database can be used when a simultaneous failure of the external storage devices 13 and 14 or a simultaneous failure with the addition of the external storage device 17 occurs. Was trying to recover.

A related technique of this type is disclosed, for example, in JP-A-61-196347.

[Problem to be Solved by the Invention] The above-described conventional technology has the following problem in an online system requiring high reliability (see FIG. 3).

(1) When the calculation center (11) is in a disaster (A) All terminals (24a, 24b, 24c, 24d) using the online system become unavailable. In addition, recovery of the database at the time of this disaster becomes impossible.

(2) Wide area 21 including the calculation center (11) (area combining multiple neighboring prefectures and municipalities) In the event of a disaster (B), not only the terminals (24a, 24b) included in the calculation center 11 but also the disaster Terminals (24c, 24d) included in another wide area (25) that does not
Will also be unavailable online. In addition, recovery of the database at the time of this disaster becomes impossible.

(3) Between the wide area (21) including the calculation center and another wide area (25), a computer (23) and a terminal (24c,
When the transmission path (26b) connecting the 24d) fails (C), the terminals (24a, 24b) in the wide area 21 can use online,
The terminals (24c, 24d) in the wide area 25 become unavailable.

(4) When large-scale construction of a computer center is required to maintain hardware reliability or to take measures against processing capacity,
Need to suspend online service.

An object of the present invention is to solve these problems.

[Means for Solving the Problems] In order to achieve the above object, an online system according to the present invention comprises a first computer system having a primary database and a secondary computer system independent of the primary database and reflecting the contents of the primary database. A second computer system having a database and connected to the first computer via a transmission line, and updating related records of the primary database and the secondary database based on a transaction input from the terminal. It is configured as follows. As the transmission line, a multiplex transmission line capable of multiplexing and performing communication between the first computer system and the second computer system is used. In the first computer system, based on a transaction input from a terminal, update information including an update history serial number indicating an update order of a related record updated by the transaction is created, and the related record of the primary database is updated. Further, the update information is transmitted to the second computer system via the multiplex transmission line. The second computer system receives the update information, and updates the related record of the secondary database according to the update history serial number included in the received update information.

In a more preferred aspect, the terminal is connected to the first and second computer systems via a relay device. The relay device connects the terminal to the first computer system during normal operation. When the first computer system cannot operate, the relay device connects the terminal to the second computer system, and online processing is performed by the second computer system. When communication between the first computer system and the second computer system is not possible, the relay device connects some terminals to the first computer system and connects other terminals to the second computer system. Connected to the computer system, the first computer system and the second computer system process transactions input from the connected terminals and update their own databases.

[Operation] According to the present invention, the first computer system is positioned as a normal computer center and the second computer system is positioned as a backup computer center. Will continue to be available.

Further, since each terminal can be selectively connected to the computer center and the backup computer center by the relay device, the database can be used in each wide area even when the transmission path between the wide areas is interrupted. Furthermore, when the computer system is stopped due to construction work at one computer center, the terminal device can be connected to the other computer center to continue the online service, and the reliability of the computer system can be maintained without stopping the online service. Maintenance,
It will be possible to carry out construction for processing capacity measures.

Embodiment An embodiment of the present invention will be described below using a bank online system as an example.

FIG. 4 shows an outline of the system and a normal connection configuration.
The terminal (38a) on the original calculation center (32a, primary system) side is connected to two dedicated lines 49f and 49h, and the terminal 38a
Half of the transaction data from () passes through a dedicated line (49f) (), a packet switch (47a), and a communication controller (45c)
Via () to the central processing unit (43a). Similarly, the other half of the transaction data passes through the dedicated line 49h (), and passes through the packet switch 47c, the multiplex transmission device 46e, the high-speed line 49d, the multiplex transmission device 46c, and the communication control device 45c (,,,).
). The same applies to the terminal 38c. On the other hand, half of the data collected from the terminal 38b of the backup calculation center (32b, the secondary system) is dedicated line 49g, packet switch 47b,
Similarly, the other half of the data passes through the multiplex transmission device 46d, the high-speed line 49b, the multiplex transmission device 46c, and the communication control device 45c, and the dedicated half line 49i, the packet switch 47d, the multiplex transmission device 46f, the high-speed line 49
c, the multiplex transmission device 46e, the high-speed line 49d, the multiplex transmission device 46c, and the communication control device 45c. The same applies to the terminal 38d. The central ledger 43a updates the original ledger (44a) corresponding to the original database 34a by the processing of the picked-up data (). The update data of the main ledger 44a is transmitted from the central processing unit 43a to the communication control device 45a, the multiplex transmission device 46a,
9a, the multiplex transmission device 46b, and the communication control device 45b, and are sent to the central processing unit 43b on the backup side. Central processing unit 4
The subledger (44b) corresponding to the backup database 34b is updated by the database update information processing by 3b, and catches up with the main ledger. In FIG. 4, the database transmission line 35 in FIG. 3 is developed as high-speed lines 49a and 49b. Therefore, the high-speed lines 49a, 49b, and 49c constitute the group of wide-area trunk lines in FIG. The original transmission line 36a is a dedicated line 4
9f and high-speed line 49d.

FIG. 5 shows the state of the original calculation center 32a after the connection change at the time of disaster.

Transaction data from the terminal 38a passes through the packet switch 47c and the multiplex transmission device 46e in the relay center 42c (), passes through the high-speed line 49c (), and passes through the multiplex transmission device 46f and the high-speed line 49.
e, via the multiplex transmission device 46d and the communication control device 45d, reach the central processing unit 43b on the sub system side ((,,)). Secondary ledger (44b) by transaction data processing on the secondary system
Is updated (). Transactions from the terminal 38b are similarly processed by the secondary system and update the secondary ledger 44b.

Multiplexing device 4 in the event of a disaster at the original calculation center 32a
Change the connection line definition in 6e, 46f, 46d. FIG. 7 shows a method of changing the connection line definition in the multiplex transmission apparatus 46e, where 71, 72, and 73 are the addresses of the lines to be connected. In the multiplex transmission device 46e, an address correspondence table of a line to be relayed as shown in FIG. 7 is set. 73
Reference numerals a and 73b denote addresses 73 of the same line on the table. This table is for backup calculation center 32
In response to the instruction from b, the data is reloaded from the floppy disk connected to the multiplex transmission device 46e and changed as shown in FIG. The same applies to the tables in the multiplex transmission devices 46f and 46d.

FIG. 8 shows a method for changing the connection line of the packet switch 47b, where 81, 82 and 83 are the addresses of the lines to be connected. Although the table shown in FIG. 8 is provided in the packet switch 47b, the table is not changed when the line connection is changed. Instead, the header of the message sent from the terminal 38b is changed from 1 at normal time to 2 at abnormal time, the header is identified by the packet switch 47b, and the connection line is changed by referring to the table. The change of the header is instructed from the backup calculation center 32b to the terminal 38b when an error occurs.

Line switching and data flow when the wide area 31a including the original calculation center suffers a disaster are the same as those in FIG. However, the terminal 38a and the relay center 42c do not exist.

FIG. 6 shows the line switching and the flow of transaction data in the event of a high-speed digital line failure connecting the primary system and the secondary system.

On the primary system side, there is no need to change the line connection, etc., and the transaction data from the terminal 38a reaches the primary system as in FIG. 4 and updates the primary ledger 44a (-). On the secondary system side, a change in the connection line definition of the multiplex transmission devices 46d and 46f and a change in the route in the packet switch 47b due to a change in the header in the message between the central processing unit 43b and the terminal 38b occur. The change on the sub system side is the same as in the case of FIG.

If center construction is required, the configuration shown in FIG.
It becomes possible by systematically switching to the configuration shown in the figure. The same applies to the construction on the sub system side.

Next, the details of the process of reflecting the normal ledger update information in the subledger will be described (see FIG. 9).

In the main system, bank account related business processing is performed.
The transaction message input from the terminal (38) is analyzed by the business processing program (94a) in the central processing unit of the main system (), and updates the main ledger 44a (). At this time, DB (database) update information (95
a) Create In order to reflect the update information 95a of the primary ledger 44a in the secondary ledger 44b, the transmission processing program (98a) queues the DB update information 95a in the transmission queue (97a) (), reads this again, and reads the secondary system side. Send to ().

In the sub system, this is received by the receiving program (98b) and written into the receiving queue (97b) (). This is read again by the read program (98c), the DB reflection processing program (94b) is started, and the DB update information (95a)
give().

In the DB reflection processing program, the DB update information is reflected in the sub ledger 44b according to the DB update information without performing business processing ().

The primary ledger 44a and the secondary ledger 44b have update history serial numbers (99a, 99b) for each record (a unit constituting the database, which is a minimum unit having a single meaning, and is a unit in which update information is created). . The update history serial number starts from 1 and is incremented by one each time a record is updated. In the example of FIG. 9, the record of the primary ledger 44a has been updated from the content AAAA to BBBB, and the corresponding record of the secondary ledger 44b has the content AAA.
This shows the state immediately before updating from A to BBBB.

FIG. 10 is a flowchart showing details of the processing of the DB reflection processing program 94b. After the reception processing program 98b receives the data (step 101), the DB reflection processing program 94b determines whether the data is a database update record (102). If the record is a database update record, the record is read from the subledger 44b (103), and the database update serial number (N, 99a) in the received record and the database update serial number (M, 99b) in the record of the subledger 44b
Is compared with (+1). If N = M + 1, the corresponding record in the subledger is updated with the received record (105). If N> M + 1, the received record is temporarily stored in the storage area (106). Then, if a record satisfying N = M + 1 is already stored in the storage area (107 Yes), the record in the sub ledger is updated with this record (108). This process is repeated until the record of the serial number stored in the storage area runs out. If N <M + 1, the update record has been received twice, and the received information is discarded (109).
When receiving the information from the main system that the transmission of the update record has been completed (No in 102), the sub-ledger updates the record in the subledger until the consecutive serial number records satisfying N = M + 1 are exhausted in the records already stored in the storage area (102). 110, 1
11).

FIG. 11 shows an example of record update according to the processing flow of FIG. Since the serial number skip record was received in, the record is stored in the storage area, and the consecutive serial number record was received in, the corresponding record in the subledger 44b is updated, and the received record is stored in the temporary storage area, but is already stored. The serial number is the same as that of the existing record, so that it is reflected in the subledger 44b following the latter record. As described above, the update order for each record can be guaranteed by the subsystem.

As described above, the update order in units of records is guaranteed on the secondary system side, so that the DB update information on the primary system side can be multiplexed and transmitted on the line 49 toward the secondary system side. In other words, it is possible to physically multiplex transmission of DB update information by providing a plurality of rotations and logically setting a plurality of logical channels for these lines.

FIG. 12 shows an example of multiplex transmission of DB update information and parallel update of update information on the subsystem.

FIG. 13 shows an example of a transmission data format from the primary system to the secondary system. In the figure, the destination address (121) is the secondary system address as viewed from the primary system, and the inter-system transmission processing serial number (122) is that the transmission processing is multiplexed by a line or the like, so that the order is determined for each line. Add to give.
The update information management (123) particularly includes a plurality of update information (124a,
124b, 124c) manages each update information when the data is locked and transmitted. The update information includes an update history serial number 99a and update data 95a.

FIG. 14 shows the flow of database records and transmission records in the original calculation center 32a and the backup calculation center 32b, and the range of physical addresses handled by programs in each layer. 94a and 94b are business processing programs, 145a and 145b are online control programs and 146a,
146b indicates the range of the operating system (OS). 141 indicates the absolute address in the disk where the record is stored, and 142 indicates the relative address in the disk where the record is stored. 147 is logical management information for identifying a database record. 143 denotes an absolute address of the line 49 allocated for data transmission, and 144 denotes a relative address of the line 49. The OS 146 needs to be aware of the absolute address 141, the relative address 142, the absolute address 143 and the relative address 144 of the disk in which the database record is stored. However, the online control program 145 only needs to be aware of the relative address 142 in the disk and the relative address 144 of the line. The business processing program 94 does not need to be aware of these addresses. Therefore, the DB update record passed from the business processing program 94a to the business processing program 94b via the line 49 does not have these physical addresses. The conclusion obtained from the above considerations is that the primary ledger 44a and the secondary ledger 44b can be laid out as databases (groups) that are physically independent of each other.

FIG. 15 is an example showing a state in which the primary ledger 44a and the secondary ledger 44b are respectively assigned to physical recording media independently. 151
Is a small-capacity recording medium, and 152 is a large-capacity recording medium.
A, B, C, and D indicate each database constituting the database group. Initially, data A, B, C, and D are stored in media 151-1, 15
1-2, 151-3 and 151-4. Next, since the capacities of the databases A and C have expanded, the storage layout of the database is changed in the original ledger 44a by adding a medium 151-5 as shown. On the other hand, in the subledger 44b, large-capacity media 152-1 and 152-2 are introduced to change the storage layout of the database as shown. This example shows that the system is highly scalable.

FIG. 16 shows an example of a state in which discrete empty areas are generated as a result of deleting / adding records from the databases stored in the media 151-1 and 151-3. The shaded area is a used record area, and the blank area is an empty record area. In the example of Fig. 16, the ledger
The database is reorganized for 44a to sort out the free areas, but the subledger 44b remains the same. This example shows that the database is a highly operable system.

Finally, it is added that the backup calculation center 32b according to the present invention becomes an online system smaller than the original calculation center 32a, and a low-cost backup calculation center 32b can be constructed.

FIG. 17 is a diagram schematically illustrating this state. 17th
FIG. 1A shows a case in which the same transaction as that input to the original calculation center is transferred to the backup calculation center, and the same business processing as the original calculation center is performed in the backup calculation center. The scale is the same as the system scale of the original calculation center. FIG. 17 (2) shows a program that exclusively performs database update processing and DB update information reception processing without performing various determinations based on input transaction information or performing processing for transmitting processing results to a terminal in the backup calculation center. By operating, the backup calculation center can be downsized. However, in this case, the backup calculation center is limited to the update processing of the backup database, and does not back up the same business processing of the original calculation center.

[Effects of the Invention] According to the present invention, even in the event of a computer center disaster or a wide area disaster involving a computer center, online services can be provided continuously using the sub-database. In addition, since the transmission of the update information between the computer center and the backup computer center is performed in a multiplexed manner, the transmission of the update information can be speeded up. At this time, even when a plurality of pieces of update information for the same record are multiplexed and transmitted, it is possible to maintain the record change order between the primary database and the secondary database.

[Brief description of the drawings]

1 is a system configuration diagram showing the concept of a highly reliable online system, FIG. 2 is a diagram for explaining a conventional system, FIG. 3 is a diagram for explaining problems of the conventional system, and FIG. 4 is a highly reliable system. FIG. 5 is a diagram showing a configuration change at the time of a computer center disaster, FIG. 6 is a diagram showing a configuration change at a wide area transmission line disaster, and FIG. 7 is a line switching of a multiplex transmission device. FIG. 8 is a diagram for explaining a method, FIG. 8 is a diagram for explaining a route change of a packet switch, FIG. 9 is a diagram showing a processing method for reflecting primary ledger update information in a secondary ledger, and FIG. 10 is a process of a secondary ledger reflection program Flowchart, FIG. 11 is a diagram showing a specific example of the update order of the subledger reflection, FIG. 12 is a diagram showing an example of multiplex transmission of database update information and parallel update of update information on the sub system side, FIG. Transmission data format from main system to sub system Fig. 14 shows an example, Fig. 14 illustrates the flow of database records and transmission records, and the range of physical addresses handled by programs in each layer. Fig. 15 shows different layouts of databases in the primary system and the secondary system separately. FIG. 16 is a diagram for explaining an example in which databases are separately organized in a primary system and a secondary system, and FIG. 17 is a diagram for explaining a system scale of an original calculation center and a backup calculation center. . 31a, 31b Wide area 32a Original calculation center 32b Backup calculation center 34a Original database 34b Backup database 38a, 38b, 38c, 38d Terminal 44a Primary ledger 44b Secondary ledger 49 … Line 94b …… DB reflection processing program 95a, 95b …… DB update information 99a, 99b …… Update history serial number

──────────────────────────────────────────────────の Continuing on the front page (72) Inventor Yukio Takaguchi 890 Kashimada, Sachi-ku, Kawasaki-shi, Kanagawa Prefecture Inside Hitachi, Ltd. Information Systems Factory (72) Inventor Yoshikuni Watanabe 890 Kashimada, Sachi-ku, Kawasaki-shi, Kanagawa No. 12 Hitachi Information Systems Factory Co., Ltd. (72) Inventor Shimizu Takanori 890 Kashimada, Kosoku-ku, Kawasaki-shi, Kanagawa Prefecture 12 Hitachi Systems Information Systems Factory Co., Ltd. (72) Inventor Kazuto Yamazaki, Kawasaki-shi, Kanagawa Prefecture 890 Kashimada 12 Hitachi, Ltd. Information Systems Factory (72) Koji Ogino 3-2-16 Omorikita, Ota-ku, Tokyo Hitachi System Engineering Co., Ltd. (72) Kazuaki Ogawa Osaka, Osaka, Japan 3-56-29 Kitahama, Chuo-ku, Hitachi Hitachi West Software Co., Ltd. (56) Documents JP-A-64-28737 (JP, A) JP-A-58-121465 (JP, A) Hitachi Review, 70 [3] (1988.3) “Trends of the Toshi Bank Third Online System- Japanese Bank Third SANBAC System- "p. 225−232 (58) Field surveyed (Int. Cl. ⁷ , DB name) G06F 12/00 G06F 13/00

Claims (2)

(57) [Claims] 1. A first computer system having a main database, and a sub-database provided separately from the main database and reflecting the contents of the main database, and connected to the first computer via a transmission line. An online system for updating related records of the primary database and the secondary database based on a transaction input from a terminal, wherein the transmission line is connected to the first computer system. Computer system and the second
And a multiplex transmission line capable of multiplexing communication with the computer system of the first computer system. In the first computer system, a related record updated by the transaction based on the transaction input from the terminal is provided. Creating update information including an update history serial number indicating an update order, updating a related record of the primary database, transmitting the update information to the second computer system via the multiplex transmission line, The computer system receives the update information, and when the update history serial number included in the received update information indicates that it is the next update of the record last updated in the second computer system, Updating a related record of the sub-database based on update information, and the update history included in the received update information If the number indicates that the record is updated after the record last updated in the second computer system, the update information is stored in a storage area, and the sub-database is stored based on the received update information. As a result of updating the related record of, the update history serial number included in the update information stored in the storage area is the next update of the last updated record in the second computer system,
An online system for updating a related record of the sub-database based on update information stored in the storage area. 2. A first computer system having a plurality of terminals for inputting transactions, a first computer system having a main database, and a second computer having a sub-database provided separately from the main database and reflecting the contents of the main database. Multiplex transmission that connects the first computer system and the second computer system and multiplexes data transmission between the first computer system and the second computer system A line, and a relay device for selectively connecting the plurality of terminals to the first computer system or the second computer system. In a normal operation, the relay device connects the plurality of terminals to the plurality of terminals. Connected to a first computer system, wherein the first computer system receives the transaction based on transactions input from the plurality of terminals. Creating update information including an update history serial number indicating the update order of the related records updated by the transaction, updating the related records in the primary database, and transmitting the update information to the second computer via the multiplex transmission line The second computer system receives the update information, and the update history serial number included in the received update information is the next update of the record last updated in the second computer system. If it indicates that, the related record of the sub-database is updated based on the update information, and the update history serial number included in the received update information is higher than the last record updated in the second computer system. If the update information indicates a later update, the update information is stored in the storage area, and the secondary information is stored based on the received update information. As a result of updating the related record of the database, the update history serial number included in the update information stored in the storage area is the next update of the last updated record in the second computer system,
An online system for updating a related record of the sub-database based on update information stored in the storage area.