JPH09269912A - Information processing method and device - Google Patents

Abstract

(57) Abstract: An information processing method and an apparatus thereof are provided which can easily realize interactive processing between users when referring to shared data among a plurality of application software. An information processing method for controlling a plurality of applications to refer to shared data, the correction / correction of shared data based on a request from a first application.
Store additional data. Then, based on the commit request of the first application, the shared data is updated based on the stored correction / additional data (S30).
5). Then, the information regarding the updated shared data is notified to the application that is referring to the other shared data (S306).

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information processing method and an apparatus thereof, and more particularly to an information processing method and an apparatus thereof for improving parallelism between transactions.

[0002]

2. Description of the Related Art Conventionally, in transaction processing in a database system, parallel control using a lock mechanism has been used. As types of this lock, generally, there are an exclusive lock required when updating data and a shared lock required when referring to data.

Further, lock acquisition and release are usually performed according to a method called two-phase locking, that is, a method that prevents mutual interference between transactions and, as a result, guarantees consistent scheduling. .

[0004]

However, the conventional method has a problem that it is not suitable for an application having an interactive operation environment. For example, when a transaction is started for the purpose of updating data by an operation of a certain user, it is necessary to acquire an exclusive lock for the data. However, for that purpose, it is necessary that the exclusive lock or the shared lock is not secured by another transaction.

According to this, the following problems occur. That is, when the content of the above data is displayed in a window on a window system, for example, a shared lock must be acquired in order to guarantee the consistency of the content of the window. The exclusive lock cannot be acquired by another transaction, and the data cannot be updated while the window is displayed.

On the other hand, after the guarantee of consistency is abandoned, the data is read out after the shared lock is secured, and the content of the data is copied to the memory area in the application.
If the acquired shared lock is released, the data can be updated by another transaction. However, even in this case, while the exclusive lock is secured, the shared lock cannot be secured by the subsequent transaction. As a result, in an application that has finished reading before the exclusive lock is secured, data that is not always guaranteed to be consistent is held, while in an application that attempts to read after the exclusive lock is secured, There will be an unpredictable wait time until the exclusive lock is released. These behaviors are significantly unsuitable in a distributed environment where multiple users collaborate and share data. The present invention solves the problems described above, that is, it is difficult to realize the conventional transaction processing, in particular, the interactive processing using the data shared by a plurality of users by using the concurrency control mechanism. Solve.

The present invention has been made in view of the above-mentioned conventional example, and when referring to shared data among a plurality of application software, it is difficult for a conventional transaction processing mechanism to realize interactive processing. It is an object of the present invention to provide an information processing method and an apparatus therefor which solve the above problem.

[0008]

In order to achieve the above object, an information processing method and apparatus of the present invention have the following configurations. That is, an information processing method in which a plurality of applications perform control to refer to shared data, and a storage step of storing correction / addition data for shared data based on a request from the first application, and a commit of the first application. Based on the request, updating the shared data based on the correction / additional data stored in the storing step, and referring to the other shared data for information about the shared data updated in the updating step. And a notification step of notifying the application.

Another aspect of the present invention is an information processing apparatus for controlling a plurality of applications to refer to shared data, and a storage means for storing correction / addition data for the shared data based on a request from the first application. , Based on the commit request of the first application,
Update means for updating the shared data based on the correction / additional data stored in the storage means, and information about the shared data updated by the update means to other applications that are referencing the shared data. And notification means.

[0010]

BEST MODE FOR CARRYING OUT THE INVENTION First, after summarizing the points of an information processing method and an apparatus thereof according to an embodiment of the present invention, a detailed description thereof will be given. The point of the information processing method according to the embodiment of the present invention is that the application data (AP1) that accesses the desired data area (D1) of the database without the shared lock is different from the application software (AP1). When the software (AP2) commits the update contents in the middle of the transaction for updating the data area (D1), the update is notified. Then, when another application software (AP2) thereafter aborts, that is, invalidates the update and performs a process of restoring the original contents of the data area (D1). Then, the application software (AP1) is again notified that the processing that has returned to the original content has been performed. By providing such a processing configuration, it is possible to easily perform interactive processing using data shared by a plurality of users.

In this processing configuration, the above-mentioned application software (AP1) refers to the data regardless of whether or not the shared lock is applied to the data area that the application software (AP1) wants to refer to. Allow the operation. [Example 1] Hereinafter, an information processing method and an apparatus therefor according to an embodiment of the present invention will be described with reference to the drawings. A known workstation or personal computer may be used as an apparatus for implementing the present invention.

FIG. 1 is a flow chart showing a procedure for processing a data reference request from a predetermined application software. Here, the identifier of this application software is "a" and the identifier of data is "d".
FIG. 2 is a diagram describing the basic elements of the process shown in the flowchart of FIG. FIG. 2 shows an example in which the application software a (1000), the application software b (1001), ..., The application software m (1002) exist in the information processing apparatus. In addition, each application software commonly accesses the data area “d” in the DB (database) (3000).

Application software a (100
0) reference data identifier list D (a) (20
00) is the application software a (100
0) indicates which data in the DB (3000) is being accessed, and is a list storing the identifiers of the respective data being referred to. The data reference application list A (d) (2001) is a list in which the identifier of each application software that is accessing the data “d” in the DB (3000) is stored.

These reference data identifier lists D (a)
(2000) and data reference application list A
(D) (2001) is a memory 2 which will be described later with reference to FIG.
It is assigned to some areas of 02. The procedure for processing the data reference request from the application software will be described below with reference to FIG. 2 as an auxiliary diagram for facilitating the understanding of the processing procedure shown in the flowchart of FIG. In this example, a process of accessing a database without acquiring an exclusive lock or a shared lock will be described.
The process shown in the flowchart of FIG. 1 is executed by a predetermined transaction manager (TM) or the like.

First, in step S102, when the application software a requests the TM to refer to the data d in the database, the TM causes the list D (of data identifiers referred to by the application software a ( Take out a). In step S103,
It is determined whether D (a) includes d. If included, application software a
Assumes that the reference processing of the data d has already been started, and immediately ends the processing. On the contrary, if not included, the process proceeds to the next step S104.

In step S104, the list A (d) of application software that refers to d is taken out. In step S105, since a (d) does not include a, a is added to A (d). Step S1
At 06, the data content indicated by the identifier d is transferred to the application software a.

In step S107, d is added to D (a), and the process ends. The procedure for processing the data reference request from the application software a to the database has been described above. Note that, in FIG. 1, the consistency between the two lists D (a) and A (d) is always maintained by a known method.

The processing order of steps S102 to S104 and steps S105 to S107 depends on the implementation of the data sharing mechanism being implemented, and the present invention is limited to this processing order. is not. The reverse operation of the procedure shown in FIG. 1, that is, the cancellation of the data reference is exactly the same as in FIG. That is,
Regarding application software a and data d,
When d (a) includes d, it is sufficient to delete d from D (a) and delete a from A (d).

FIG. 3 explains the setting process of the shared lock for the data of the database accessed by each application after the TM starts the reference process for the shared data according to the procedure of FIG. FIG. 4 is a diagram showing a configuration example of the lists L and L ′ that manage the setting target of the shared lock.

The shared lock request identifier list L (4
000) is a list of data identifiers generated by the TM and requesting acquisition processing of a shared lock from each application software. This is a list of data that has been accessed by the database generated by the TM by monitoring access from each application to the database, that is, D (x); x = a, b ,. . . It is a list equivalent to c 1 or a list of data in D (x) for which a shared lock request has been made by the application software.

The shared lock request identifier list L (4000) is stored in the memory 20 which will be described later with reference to FIG.
It is assigned to some areas of 2. In step S202 of FIG. 3, the above-mentioned shared lock request identifier list L
A list L ′ (non-locked data identifier list) is created by removing the elements of (4000) that have already been secured with the shared lock or the exclusive lock.

In step S203, it is checked whether the list L'is empty. If it is not empty, the process proceeds to step S204. If it is empty, the process ends. In step S204, a shared lock is secured for each element of the list L ', and the process ends. The above processing is TM
By repeating the above, the shared lock can be appropriately applied to the data that needs the shared lock.

The TM also monitors the occurrence of data update in the access from each application to the database together with the above processing, and generates the data list G listing the identifiers of the data for which data update has occurred. next,
In Figure 5, each application updates data (commits)
5 is a flowchart showing a processing procedure of TM in a transaction for performing

First, in step S302, the data in the list G is fetched. In step S303, it is checked whether the list G is empty, and if it is empty, the process ends.
On the other hand, if it is not empty, the process proceeds to step S304. In step S304, one element is extracted from the list G and d1
And In step S305, it is checked whether the contents of d1 have already been saved. If the evacuation has been completed, the process returns to step S303. On the contrary, if the evacuation has not been completed, the process proceeds to step S306.

In step S306, the contents of d1 are temporarily saved. In step S307, the content of d1 is updated. In step S308, each application software of the list A (d1) that refers to d1 is notified of the change. Then, the process returns to step S303 and the same processing is repeated.

By the above processing, the information updated by the predetermined application can be notified to the other application being referred to. Incidentally, step S306
The save area of the contents of d1 in is finally released when the transaction is completed. Here, when the transaction finally ends is either a commit or an abort. Here, the commit is a process of confirming the updated contents. Further, the abort means a process of discarding the updated content without fixing it, in other words, a process of returning to the state before the update.

Therefore, in the case of commit, the save area is simply released. On the other hand, in the case of an abort, if the save area of d1 is not empty, the content of the update process performed temporarily is discarded, and the content of the save area is reset as the content of d1. At this time, if A (d1) is not empty, the change notification is given to each element of A (d1) as in step S308.

By using the processing procedure described above, for example, effective data sharing can be performed between certain application software a1 and a2. This will be described with reference to FIG. As shown in FIG. 6, assuming that the database has data dt1, dt2, dt3, dt4, the application software a1 and a2 refer to dt1, dt2, dt3, and dt1, dt2, dt4 according to the procedure of FIG. 1, respectively. Suppose you have started the procedure.

Here, the application software a
It is assumed that 1 and a2 start a transaction to update dt1 and dt2, respectively. At this time, FIG.
In the processing procedure shown in FIG. 3, the list of data identifiers for securing the shared lock is the application software a.
1 includes dt3, and a2 includes dt4.

If the above-mentioned shared lock is secured by the procedure shown in FIG. 3, at some point, a1 and a2 will secure exclusive locks for dt1 and dt2, respectively, but there is a lock conflict between a1 and a2. Since there is no, there are no other disturbance factors and they succeed together.
At this stage, if the contents are temporarily updated according to the procedure shown in FIG. 5, the application software a1 and a2
Each is immediately informed of the updates made by the other and can proceed accordingly.

In general, it is not preferable that a temporary update notification is issued by another transaction during the data update transaction. This is because the evaluation of the temporary update contents in the former transaction may not eventually guarantee the consistency at the commit stage. Therefore, at the start of a normal transaction, it is appropriate to apply the procedure shown in FIG. 3 to all the shared data whose reference is started in the procedure of FIG.

Further, as the transaction, there is a transaction of a type that does not update the data, but the procedure shown in FIG. 3 can be applied to this as well. The purpose of the transaction in this case is to prohibit other transactions from updating during the transaction. In the present embodiment, it is unnecessary to secure a shared lock in the data reference process, and it is appropriate to keep the read transaction to the minimum necessary in order to avoid conflict between transactions due to the unnecessary lock. [Embodiment 2] In the embodiment described above,
2 lock acquisition / release processing in transaction
It was supposed to be done in the phase.

However, in a system in which the update processing and the reference processing according to the present embodiment coexist and which supports the temporary update processing, the so-called "seriali"
It is appropriate to allow data sharing at the expense of "zability". To achieve this, an explicit lock release instruction is provided to the application. FIG. 7 is a flowchart showing the processing procedure of the TM when the explicit lock release command is received.

In this flowchart, the procedure for releasing the exclusive lock and securing the shared lock is shown, but the same applies to the exclusive lock release processing and the shared lock release processing. However, regarding the shared lock release process,
The temporary update process described later is not performed. Step S4
In 01, the data identifier d for explicitly releasing the lock
Receive 2

In step S402, it is checked whether the exclusive lock for d2 has already been secured.
Then, if the exclusive lock is not secured, the process ends. Otherwise, it proceeds to step S403. In step S403, it is determined whether the content of d2 has been changed. If the change is made, the process proceeds to step S404. If there is no change, step S405
Proceed to.

In step S404, the updated contents are temporarily reflected according to the procedure shown in FIG. Step S4
In 05, the exclusive lock is changed to the shared lock, and the process ends. The processing in step S405 is assumed to be performed inseparably, and is performed with priority over other transactions waiting for the release of the lock.

FIG. 7 shows the processing procedure for individual transactions that do not follow the two-phase lock. However, in general, a nested transaction process may be performed, and although the upper transaction follows the two-phase lock, if the procedure of FIG. 7 is performed in the lower sub transaction, a contradiction occurs.

In order to prevent this, when any of the upper transactions follows the two-phase lock acquisition / release process, the lock acquisition / release process in the sub-transaction is performed in two phases regardless of the designation. In this case, the explicit lock release instruction is lazily evaluated, and finally evaluated when returning to the upper transaction that does not follow the two-phase lock.

Next, FIG. 8 is a diagram showing an example of the configuration of an information processing apparatus that executes the processes of the above-described flowcharts. In this figure, the CPU 200 controls the entire information processing apparatus by reading, interpreting, and executing various programs stored in the memory 202. The memory 202 stores programs corresponding to the above-described flowcharts and also stores a database.

The keyboard 203 and pointing device 204 input commands and data. The display monitor 201 displays processing results of the CPU 200 and commands and data input from the keyboard 203 and pointing device 204. In addition, the present invention
It may be applied to a system composed of a plurality of devices or an apparatus composed of one device.

Another object of the present invention is to supply a storage medium storing a program code of software for realizing the functions of the above-described embodiments to a system or an apparatus, and to provide a computer (or CPU) of the system or the apparatus.
And MPU) read and execute the program code stored in the storage medium.

In this case, the program code itself read from the storage medium implements the functions of the above-described embodiment, and the storage medium storing the program code constitutes the present invention. As a storage medium for supplying the program code, for example, a floppy disk, a hard disk, an optical disk, a magneto-optical disk, C
A D-ROM, a CD-R, a magnetic tape, a non-volatile memory card, a ROM, etc. can be used.

When the computer executes the readout program code, not only the functions of the above-described embodiment are realized, but also the OS (Operating System) running on the computer based on the instruction of the program code. ) May perform some or all of the actual processing, and the processing may realize the functions of the above-described embodiments.

Further, after the program code read from the storage medium is written in the memory provided in the function expansion board inserted into the computer or the function expansion unit connected to the computer, based on the instruction of the program code, It goes without saying that a case where the CPU or the like included in the function expansion board or the function expansion unit performs some or all of the actual processing and the processing realizes the functions of the above-described embodiments is also included.

When the present invention is applied to the above-mentioned storage medium, the storage medium stores the program code corresponding to the above-described flowcharts. Briefly, it corresponds to each of the above-mentioned flowcharts. Each program module will be stored in the storage medium. As described above, in the information processing method according to the present embodiment, the application software is allowed to perform the data reference operation regardless of whether the shared lock is acquired. In addition, when the application software performing the data reference operation starts a transaction for the purpose of data update processing, prior to the activation of the transaction, a series of shared data for which the data reference operation is activated Specify whether to secure a shared lock. Then, when the update contents are temporarily committed in the middle of the transaction for the data update process, a change notification notifying that the data has been changed in the commit process is activated for the data reference operation. Notify a series of application software.

When the transaction is finally aborted, the reverse operation corresponding to the commit process is performed to recover the original data. Then, the change notification in the reverse operation is notified to the series of application software that has started the data reference operation. The transaction for the data update process can specify whether the lock acquisition / release process is performed according to the two-phase lock.

With the above basic processing configuration, interactive processing using data shared by a plurality of users can be easily realized.

[0048]

As described above, according to the present invention, it is difficult for a conventional transaction processing mechanism to realize interactive processing when referring to common data among a plurality of application software. Can solve problems.

[Brief description of drawings]

FIG. 1 is a flowchart showing a procedure of processing a data reference request for predetermined data in a database from each application according to an embodiment of the present invention.

FIG. 2 is a diagram for explaining the process of FIG.

FIG. 3 is a flowchart showing a procedure for acquiring a shared lock.

FIG. 4 is a diagram showing a configuration example of a shared lock request data identifier list and a non-lock data identifier list.

FIG. 5 is a flowchart showing a processing procedure for temporarily notifying a predetermined application of data update contents during a transaction.

FIG. 6 is a diagram for explaining a processing example of the first embodiment.

FIG. 7 is a flowchart showing a processing procedure for explicitly releasing a lock.

FIG. 8 is a diagram showing an example of a hardware configuration for executing information processing according to the present embodiment.

Claims (8)

[Claims] 1. An information processing method for controlling a plurality of applications to refer to shared data, comprising a storage step of storing correction / additional data for the shared data based on a request of the first application, An updating step of updating the shared data based on the correction / additional data stored in the storing step based on an application commit request, and information regarding the shared data updated in the updating step to the other shared data. And a notification step for notifying an application that is referring to the information processing method. 2. The method further comprises a restoration step of restoring the shared data updated in the updating step to the data before being stored in the storing step on the basis of an aboard request of the first application. The information processing method according to claim 1. 3. A detection step of detecting an application accessing the shared data, and when the application does not secure a shared lock or an exclusive lock for the data accessed by the application detected in the detection step, The information processing method according to claim 1, further comprising: a setting step of setting a shared lock prior to the application performing the updating step. 4. The restoration step further includes information indicating that the shared data updated in the updating step is restored to the data before being stored in the storing step, and another application using the shared data. The information processing method according to claim 2, further comprising a return notifying step of notifying the user. 5. An information processing apparatus for controlling a plurality of applications to refer to shared data, comprising: storage means for storing correction / addition data for the shared data based on a request from the first application; An update unit that updates the shared data based on the correction / additional data stored in the storage unit based on an application commit request, and information about the shared data updated by the update unit to the other shared data. An information processing apparatus, comprising: a notification unit for notifying an application that is referring to. 6. A return means for returning the shared data updated by the update means to the data before being stored in the storage means, based on an aboard request of the first application. The information processing device according to claim 5. 7. A detection unit that detects an application that accesses the shared data, and a case where the application does not secure a shared lock or an exclusive lock for the data that the application detected by the detection unit accesses. The information processing apparatus according to claim 5, further comprising: a setting unit that sets a shared lock before the application performs the updating step. 8. The restoring means further includes information indicating that the shared data updated by the updating means is restored to the data before being stored in the storage means, and another application that is using the shared data. The information processing method according to claim 7, further comprising return notification means for notifying the user.