JPH02100718A - Program area determination system - Google Patents

Abstract

PURPOSE:To utilize the storage area of a main storage device effectively by obtaining the best storage capacity according to execution information in the case that a load module is executed actually. CONSTITUTION:An execution information gathering means 100 gathers the execution information indicating storage capacity occupied in a main storage device 2 when the load module LM is executed. An execution information storage means 200 stores and holds the execution information that the execution information gathering means 100 every time the load module LM is executed repeatedly. An execution information analyzing means 300 analyzes pieces of execution information held in the execution information storage means 200 to calculate the best storage capacity of a program area AL to be secured in the main storage device 2 so as to execute the load module M. Consequently, the best storage capacity is obtained according to the execution information when the load module is executed actually, so the necessary and sufficient program area is secured to utilize the storage area of the main device effectively.

Description

[Detailed Description of the Invention] [Summary] In an information processing system that stores and executes a load module generated by translating a source program into machine language in the main memory, the program area to be secured in the main memory is Regarding the program area determination method for optimization, the purpose is to secure the optimal program area according to the actual execution state. an execution information collection means for collecting information; an execution information storage means for accumulating and holding execution information collected by the execution information collection means each time a load module is repeatedly executed; and a plurality of execution information stored in the execution information storage means. In order to analyze the information and execute the load module,
Optimum storage capacity for program area that should be secured in main memory ■
and an execution information analysis means for calculating the execution information.

[Industrial application field]

The present invention relates to an information processing system, and in particular to optimization of a program area to be secured in a main memory when a load module generated by translating a source program into machine language is stored in the main memory and executed. This invention relates to a program area determination method that aims to achieve this.

A load module generated by an information processing system by translating a source program into machine language usually consists of multiple partial modules. The partial module is stored in a pre-secured program area of the main memory.

Since it is necessary to store various data in the main memory in addition to the load module, it is recommended that the storage capacity of the program area reserved for executing the load module be optimized as much as possible. This is necessary to effectively utilize the storage area of the device.

[Prior Art] As this type of program area determination method, for example, the invention “
``Divided Compilation Execution Data Management System'' (Japanese Patent Application No. 62-279504) was filed on November 5, 1988, and FIG. 3 shows an example of the conventional program area determination method. FIG. 2 is a diagram illustrating an example of a "divided compilation method for managing data for execution." FIG.

In FIG. 3, when the program translator icoM translates a source program SP consisting of partial programs SU 1.5tJ2 and SU3 into a machine language program MP, each partial program SU 1, SU2, and SU 3 is translated into a machine language program MP. The storage capacity required to store machine language program units MUI, MU2, and MU, 3, which are generated by individually translating (referred to as the stack size in ``2009'') are summed up by taking into account the mutual reference relationship when each machine language program unit MUI, MtJ2, and MU3 are executed, and the total machine language program M
The storage capacity required when executing P is statically calculated.

In FIG. 3, the partial program SUI is block A
(stack size XI) and B (stack size X2)
), and the partial program S [J2 is block C
Consists of only (stack size
block D refers to block C,
Furthermore, in plot C, when referring to blocks A and B, the storage capacity N (stack size) of the entire machine language program MP is calculated as (X4+X3+max(XI X2)).

[Problem to be solved by the invention]

As is clear from the above explanation, in the conventional program area determination method, when the program translator COM translates the source program SP into the machine language program MP,
It was calculated statically from the storage capacity and reference relationships of each partial program unit.

However, the mutual reference relationship between each partial program unit is falsified by the condition data input by the user during execution, and when calculating execution information, the partial program unit that is supposed to be referenced is not referenced by the condition data consisting of it. In some cases, depending on the condition data 2. However, there was a problem in that it was difficult to say that the storage capacity calculated conventionally was optimal.

An object of the present invention is to secure an optimal program area in accordance with the actual execution state.

[Means for solving problems] FIG. 1 is a diagram showing the principle of the present invention.

In FIG. 1, LM is a load module generated by translating a source program into machine language, 2 is a main memory that is stored when executing load module LM, and AL is a storage device for executing load module LM. This is a program storage area secured in the main storage device 2.

Reference numeral 100 denotes execution information collection means provided in the information processing system according to the present invention.

Reference numeral 200 denotes execution information storage means provided in the information processing system according to the present invention.

300 is an execution information analysis means provided in the information processing system according to the present invention.

[Effect]

The execution information collection means 100 collects execution information indicating the storage capacity occupied in the main storage device 2 when the load module LM is executed.

The execution information storage means 200 accumulates and holds the execution information collected by the execution information collection means 100 each time the load module LM is repeatedly executed.

The execution information analysis means 300 analyzes a plurality of pieces of execution information held in the execution information storage means 200, and determines an optimal storage container for the program area A to be secured in the main storage device 2 in order to execute the load module LM. Calculate.

Therefore, the optimal storage capacity is obtained based on the execution information when the load module is actually executed, so the necessary and sufficient program area is secured, the storage area of the main storage device is effectively used, and the relevant information Processing system reliability and economy are improved.

〔Example〕

An embodiment of the present invention will be described below with reference to the drawings.

FIG. 2 is a diagram showing an information processing system according to an embodiment of the present invention. Note that the same reference numerals indicate the same objects throughout the figures.

In addition to the central processing unit 1, main storage device 2, external storage devices 3 and 4, the information processing system shown in FIG. 2 is provided with an execution information collection mechanism 5 as the execution information collection means 100 in FIG. , an execution information storage mechanism 6 is provided as the execution information storage means 200 in FIG. has been done.

The external storage device 3 stores a load module LM that is made up of partial modules LU, ', LUz, and L TJ 3 to be executed, and the external storage device 4 stores a load module LM. n pieces of condition data Da to D1ゎ by the first user who executes, and n pieces of condition data Db+ to D1. ,
is stored.

In this state, the central processing unit 1 that executes the load module LM based on the first user is first given a storage capacity statically calculated by the program area determination method as shown in FIG. Program area A in device 2
, first extracts the condition data Da+ from the external storage device 4, stores it in the data area A in the main storage device 2 via the common bus 8, and also extracts the condition data Da+ from the external storage device 3. Each partial module LUI that makes up
r to u3 are sequentially extracted and stored in the program area AL in the main storage device 2 via the common bus 8, and the condition data D
Execute based on □.

When the execution based on the condition data D and l is completed, the central processing unit 1 retrieves the condition data D8□ to D from the external storage device 4.
An is sequentially extracted, and the load module M is sequentially extracted from the external storage device 3, and the execution is repeated based on the extracted condition data D, 2 to D3□.

On the other hand, when the central processing unit 1 executes the load module LM based on the condition data DI+ to D□7, the execution information collection mechanism 5 extracts the load module LM from the external fertilizer a device 3, By monitoring the address sent to the common bus 8, the storage capacity actually occupied by the load module LM, that is, the execution information S, l to S-, is collected and stored via the common bus 8. The execution information is stored in the execution information storage mechanism 6.

When the execution of the load module LM based on the condition data Da+ is completed, the execution information storage mechanism 6
Execution information S0 to S- collected by the execution information collection mechanism 5 each time is accumulated and held in the execution information storage section 1.

On the other hand, the execution information analysis mechanism 7 analyzes n sets of execution information regarding the first user held in the execution information storage mechanism 6,
The frequency distribution of execution information tta s is calculated based on S all, and all execution information Sat is calculated with a predetermined probability.
A value S, including S'an to S'an. is determined to be the optimal execution information, and the determined optimal execution information S1° is notified to the central processing unit 1 via the common node 8.

The central processing unit 1 that has been notified of the optimal execution information S au,
From now on, when executing the load module LM based on the condition data D regarding the first user, the optimal execution information S
Progera with a storage capacity equivalent to 1. Area At
is secured in the main storage device 2.

Next, when the central processing unit 1 executes the load module LM based on the second user, the program area AL having the statically calculated storage capacity is first stored in the main storage device 2 in the same manner as described above. After securing, the condition data Dbl to Dbn are sequentially extracted from the external storage device 4, the load module LM is extracted from the external storage device 3, and the load module L is extracted based on the extracted condition data Dbl to Db, l.
Execute M repeatedly.

Furthermore, as described in 117I, when the central processing unit l executes the load module LM based on the condition data Db+ to Dbn, the execution information collection mechanism 5 also uses the execution information actually occupied by the load module LM in the main storage device 2. fW S b
l to Sbn are collected and stored in the execution information storage mechanism 6 via the common lotus 8.

Furthermore, in the same manner as described above, the execution information analysis mechanism 7 also calculates the frequency distribution of execution information (execution information S based on the execution information Sbl to Sbn of ni regarding the second user held in the K storage mechanism 6), All execution information S with predetermined probability
Determine the value Sbo including bl to S as the optimal execution information,
The determined optimal execution information S,0 is notified to the central processing unit l via the common lotus 8.

Optimal execution situation! :Isb. When the central processing unit 1 that has been notified of the following executes the load module LM based on the condition data Db regarding the first user, the optimum execution information IIs. A program area AL having a storage capacity equivalent to 1 is secured in the main storage device 2.

As is clear from the above description, according to the present embodiment, the central processing unit l performs the load module L with respect to the first user.
When executing M, condition data D s l~,
, , the program area AL corresponding to the optimal execution information S8° determined based on the execution information Is, . . . When executing the load module LM for the second user, the condition data D
Optimal execution information S, determined based on execution information Sbl to Sbn collected when the load module LM is executed based on bl to Db. In order to secure the program area A corresponding to , the minimum necessary program area AL that can be stored by the load module LM with a predetermined probability is secured, and the storage area of the main storage device 2 is effectively used. used.

Note that FIG. 2 is only one embodiment of the present invention, and the types of users, the configuration of the load module LM, and the number of condition data D1 and D1 are not limited to those shown in the figure. (Although many other modifications may be considered, the effects of the present invention will not change in any case. Also, the configuration of the information processing system to which the present invention is applied is not limited to that shown in the drawings. Needless to say.

〔Effect of the invention〕

As described above, according to the present invention, in the information processing system, the optimum storage capacity can be obtained based on the execution information when the load module is actually executed, so that the necessary and sufficient program area is secured, and the main memory The storage area of the device is effectively used, and the reliability and economic efficiency of the information processing system are improved.

[Brief explanation of drawings]

FIG. 1 is a diagram showing the principle of the present invention, FIG. 2 is a diagram showing an information processing system according to an embodiment of the present invention, and FIG. 3 is a diagram showing an example of a conventional program area determination method. In the figure, 1 is a central processing unit, 2 is a main storage device, 3 and 4 are external storage devices, 5 is an execution information collection mechanism, 6 is an execution information storage mechanism, 7 is an execution information analysis mechanism, 8 is a common lotus, 100 1 is an execution information collection means, 200 is an execution information storage means, and 300 is an execution information tree 3.

Claims (1)

[Scope of Claims] In an information processing system in which a load module (LM) generated by translating a source program into machine language is stored in a main storage device (2) and executed, the load module (LM) is executed. execution information collection means (100) for collecting execution information indicating the storage capacity occupied in the main storage device (2) when the load module (LM) is executed repeatedly; Execution information storage means (200) that accumulates and holds the execution information collected by the information collection means (100); and Execution information storage means (200) that analyzes the plurality of pieces of execution information held in the execution information storage means (200), and analyzes the execution information stored in the execution information storage means (200). A program area determination method characterized by comprising an execution information analysis means (300) for calculating the optimal storage capacity of the program area (A_L) to be secured in the main storage device (2) in order to execute the program area (LM). .