JP2003263366A - Swapping control method, its execution device and its processing program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technology capable of reducing the absolute quantity of data transfer when executing swapping, and thereby improving throughput of an actual work and a response time. <P>SOLUTION: This method has a step for setting the difference between a prescribed threshold and the number of blank pages in a main storage as the number of pages which are swap-out objects in the case where the number of blank pages in the main storage is below the prescribed threshold when swap-out is required, a step for stopping processing of a program of the swap- out object as the need arises to swap-out requirement, to thereby set the program in the state where a CPU is not allocated thereto, a step for saving the pages of the set number of pages in a swap-out region and releasing the saved pages in the main storage, a step for securing the pages in the main storage and recovering the pages saved in the swap-out region when swap-in is required, and a step for resuming processing of the program of a swap-in object, to thereby set the program in the state where the CPU is allocated thereto. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a computer system having a main memory device and an auxiliary memory device for controlling swapping in a virtual memory system, and more particularly to a computer system for efficiently performing swapping when an address width is expanded. It is related to the technology effectively applied to the system.

[0002]

2. Description of the Related Art In a conventional computer system, in order to improve the efficiency of business processing executed on the system, the address width is expanded, the virtual memory size that can be expanded on the system is expanded, and the actual computer that can be used on the system is expanded. We are trying to expand the memory size.

If the total capacity of the virtual memory expanded on the computer system is smaller than the capacity of the mounted real memory device, all virtual memory areas can be allocated on the real memory device. Has a larger virtual memory capacity.

Therefore, in a system in which the virtual memory capacity is larger than the actual memory capacity, the operating system expands the virtual memory by using the actual memory device and the auxiliary memory device, and allocates the virtual memory area necessary for executing each program. A virtual storage system is adopted in which an unnecessary virtual storage area is placed on an auxiliary storage device by allocating it on a real storage device. Such dynamic allocation of virtual storage areas is realized by operations such as paging and swapping. Here, paging is an operation of transferring data between the real storage device and the auxiliary storage device in units of the management size of the real storage device.
Swapping is an operation of dynamically deciding an arrangement location for each space unit in a computer system capable of arranging multiple virtual storage spaces.

In such a virtual memory system, a plurality of virtual memory spaces are swapped by the multiplicity control as disclosed in Japanese Patent Laid-Open No. 62-296237 to provide an environment in which a plurality of programs can be executed. is doing.

However, if the system scale is expanded while keeping the ratio of the total virtual memory capacity and the mounted actual memory capacity constant, the absolute amount of data transfer between the actual memory device and the auxiliary memory device such as the paging and the swapping described above will be increased. And the operating system processing overhead is enormous. In particular, if the paging or swapping amount increases abruptly, it may cause the throughput or response time of the actual work to deteriorate sharply, which must be avoided.

That is, in the computer system in which the address width is expanded and the scale is expanded as described above, regarding the data transfer between the real storage device and the auxiliary storage device by paging or swapping, the absolute amount of these data transfers is reduced. A method and a method of distributing the frequency of data transfer are required.

As a method for solving such a technical problem, there is a technique described in Japanese Patent Laid-Open No. 5-61774.
In Japanese Patent Laid-Open No. 5-61774, the virtual storage area of the virtual storage space to be swapped out is divided into a plurality of groups, and the data is transferred to the auxiliary storage device in units of the groups to distribute the swap-out load. The method of doing is disclosed.

Similarly, as a method for solving the above problem, there is a logical swapping method disclosed in Japanese Patent Laid-Open No. 5-94315. In Japanese Patent Laid-Open No. 5-94315, candidates for virtual storage spaces that can be swapped out are listed in advance, and when the utilization rate of a real storage device becomes high, the real storage actually used by the virtual storage space is increased. A method of transferring data in a region to an auxiliary storage device is disclosed.

[0010]

As described above, when the address width is expanded by the expansion of the architecture, the program size becomes large due to the increase of the virtual memory capacity, and the absolute amount of data transfer at the time of swapping increases. Further, when the system scale is expanded, the absolute amount of data transfer at the time of swapping is increased due to an increase in data arranged on the real storage device for processing by the program.
In the above-mentioned conventional techniques, no consideration is given to suppressing the increase in the swapping amount in these cases.

Further, in the system in which the address width is expanded,
In order to operate the program created before expansion without changing it, the area accessed by the program at the time of operation must be arranged in the area having the address width that can be expressed by the program. In particular, in case the virtual storage area used by the program is made resident on the real storage device and the area is accessed by an address on the real storage device, an address that can be accessed by the program when the resident storage is made The page must be placed in the width area.

As described above, even if the address widths of the real memory and the virtual memory are expanded, not all programs can use the expanded area without limitation. In particular, regarding the actual storage, in order to guarantee the operation of the program created before expanding the address width, the utilization rate of the first area that can be expressed with a small address width is It can be expected to be higher than that.

Despite the low utilization rate of the area expressed by a large address width, swapping occurs due to such a local load increase, and the input / output operation for this operation is performed as a whole system. Will increase.

An object of the present invention is to solve the above problems and to provide a technique capable of reducing the absolute amount of data transfer at the time of swapping and thereby improving the throughput and response time of actual work. is there.

Another object of the present invention is to reduce the data transfer time required for swapping a program which can operate only in a specific area expressed by a specific address width, thereby improving the throughput and response time of actual work. It is to provide the technology that can be done.

[0016]

According to the present invention, there is provided a computer system comprising a main memory device and an auxiliary memory device for controlling swapping in a virtual memory system, wherein a predetermined threshold value and a main memory are used when swapping is requested. Swapping is performed for pages of the number of pages set by the difference from the number of empty pages of the device.

In the computer system of the present invention, when swap-out is requested, it is checked whether the number of free pages in the main storage device is less than a predetermined threshold, and if the number of free pages is less than the predetermined threshold, Sets the difference between the predetermined threshold and the number of free pages in the main storage device as the number of pages to be swapped out.

Next, after the processing of the swap-out target program is stopped as necessary in response to the swap-out request to put the program in a state in which the CPU is not allocated, the page of the set number of pages is set. Is saved in the swap-out area, and the saved page in the main storage device is released.

When swap-in is requested, after securing a page in the main storage device, the page saved in the swap-out area is restored to the secured page in the main storage device and swapped. The process of the in-target program is restarted to bring the program into a state in which the CPU is assigned.

As described above, according to the present invention, the data for the number of pages necessary for resolving the shortage of empty pages in the main memory is saved in the swapping area, and some pages are retained in the main memory. Therefore, when swapping is performed, it is possible to reduce the absolute amount of data transfer, thereby improving the throughput and response time of actual work.

Further, according to the present invention, when swap-out is requested, the number of empty pages in the first area of the main memory represented by a specific address width is less than a predetermined threshold, and the first area is If the number of free pages in the second area located at a higher address is equal to or more than the set number of pages, the swap-out area is set as the second area, and the page targeted for swap-out is set as the second area. Perform intermediate swapping to move to a free page.

As described above, according to the present invention, when there is insufficient free space in the first area of the main memory represented by a specific address width, and there is free space in the other second area, the first area Since the intermediate swapping that moves the page of the second area to the second area is performed, the data transfer time required for swapping the program that can operate only in the first area expressed by a specific address width is shortened, and the actual business Throughput and response time can be improved.

As described above, according to the computer system of the present invention, when the swapping is requested, the swapping is performed for the page of the number of pages set by the difference between the predetermined threshold value and the number of empty pages of the main storage device. , It is possible to reduce the absolute amount of data transfer when performing swapping, thereby improving the throughput and response time of actual work.

[0024]

BEST MODE FOR CARRYING OUT THE INVENTION A computer system of one embodiment, which comprises a main memory device and an auxiliary memory device and controls swapping in a virtual memory system, will be described below. In the present embodiment, the conventional swap-out that is output to the auxiliary storage at the time of swap-out is physical swap-out, and the swap-out that is newly created by the present invention to move from the first area to the second area on the main storage device is The explanation will be given assuming that it is called an intermediate swap-out. In addition, the pages allocated to the space due to swapping are saved / recovered in the auxiliary storage, and the pages in the main storage device are released / recovered as necessary.
The securing process is similar to the process performed in the conventional swapping, and the swapping request process is similar to the process shown in the related art.

FIG. 1 is a diagram showing a schematic configuration of a computer system of this embodiment. As shown in FIG. 1, the computer 1000 of this embodiment has a swapping management unit 600, a space deactivation unit 700, a main memory allocation unit 800, and a space activation unit 900.

When the swap-out request is made and the number of free pages in the main storage device is less than the predetermined threshold value, the swapping management unit 600 determines the difference between the predetermined threshold value and the number of free pages in the main storage device. This is a processing unit that is set as the number of pages to be swapped out.

The space deactivating unit 700 suspends the processing of the swap-out target program as necessary in response to the swap-out request, and executes the program concerned in the CPU 110.
This is a processing unit that sets a state in which 0 is not assigned.

When a swap-out is requested, the main memory allocation unit 800 saves the set number of pages to the swap-out area and releases the saved page in the main memory, This is a processing unit that secures a page on the main storage device and recovers the page saved in the swap-out area when swap-in is requested. The space activation unit 900 restarts the processing of the swap-in target program and executes the program to the CPU.
1100 is a processing unit that makes a state in which allocation is performed.

The swapping management unit 600 separates processing such as intermediate swap-out based on the input parameter 200. The space deactivating unit 700 receives a process from the swapping managing unit 600 when the request is an intermediate swap-out request, stops the input / output being processed in the target space, stops the job, and stops the job. Prevents allocation from happening and allows swapout.

The main memory allocation unit 800 receives the processing from the space deactivation unit 700 or the swapping management unit 600, and saves or restores the information of the page allocated to the target space to the auxiliary storage device 1600. The page is released or secured accordingly. In the case of a swap-in request, the space activation unit 900 is passed processing after the main memory allocation unit 800, restarts the job being executed in the target space, enables allocation of the CPU 1100, and space inactivation. The input / output for which the processing has been stopped by the conversion unit 700 is restarted.

The swapping management unit 60 is used for the computer 1000.
0, the program for functioning as the space deactivating unit 700, the main memory allocating unit 800, and the space activating unit 900 is recorded on a recording medium such as a CD-ROM and stored in a magnetic disk or the like, and then loaded into a memory. Shall be executed. The recording medium for recording the program is C
A recording medium other than the D-ROM may be used. The program may be installed in the information processing apparatus from the recording medium and used, or the recording medium may be accessed through the network to use the program.

FIG. 1 shows the configuration of a computer system according to this embodiment. A computer 1000 is composed of a CPU 1100 and a main storage device 1200. Also, calculator 1
000 has an auxiliary storage device 1600, and the main storage device 12
00 is stored in the auxiliary storage device 1600.

OS1 arranged on the main storage device 1200
300 is executed by the CPU 1100, and the virtual storage space 1
Jobs (user programs) on the server 500 are activated, and resources such as the virtual storage space 1500 are managed.
A plurality of virtual storage spaces 1500 can be arranged on the main storage device 1200.

Further, the computer 1000 has an auxiliary storage device 1601 which stores an initial setting parameter 1700, and sets an initial setting value necessary for the OS 1300 to operate such as a threshold value for determining the lack of the main storage device 1200. ,
It can be specified by the initial setting parameter 1700.

The main memory 1200 includes a first area 1201 and a first area 120, each of which has a specific address width.
A second area 1202 located at an address higher than 1
Consists of. For example, when the 32-bit address width is expanded to 64 bits, the area represented by the 32-bit address width becomes the first area 1201, and the first area among the areas represented by the 64-bit address width is The area excluding 1201 becomes the second area 1202.

The virtual storage space 1500 includes the first area 12
01 first type page that must be placed on 01
1 and a second type page 1502 that can be arranged in any area on the main memory 1200. Also, a main memory allocation management table 500 storing information for managing pages allocated to the virtual memory space 1500 is prepared for each space. Here, the first type of page 1501 that must be arranged on the first area 1201 is the program created on the assumption of the address width of the first area 1201 to the real storage device of the virtual storage area. Page 1501 of the first type for guaranteeing the operation of the program, such as an I / O processing program in which an address is made resident. Arrangement on one area 1201 is performed.

On the OS 1300, a swapping activation unit 1400 for issuing a swapping request, and an input parameter 20 for the swapping unit 100 to receive the swapping request.
0, the number of empty pages information 300 holding the usage status information of the main storage device 1200, the threshold value 400 for judging the load status of the main storage device 1200, the main storage allocation management table 500, and the swapping unit 100. It is arranged. The swapping unit 100 of this embodiment includes a swapping management unit 600, a space deactivation unit 700, a main memory allocation unit 800, and a space activation unit 900.

FIG. 2 is a diagram showing the structure of the input parameter 200 of this embodiment. As shown in FIG. 2, the input parameter 200 stores information about swapping of the space that is the target of the swapping request.

The swapping space designation area 201 is an area in which a space identifier for identifying the space which is the object of the swapping request is stored. Swapping type 20
An area 2 stores information that specifies whether the request is swap-out or swap-in.

As the type 203 of the page to be moved, the page 1501 of the first type is moved or the page 1502 of the second type is used.
This is an area for storing information that specifies whether to move or to target both. The move destination 204 is an area for storing information specifying where to output the page specified by the type 203 of the page to be moved. When moving the first type page 1501 to the second area 1202,
Information indicating the second area 1202 is designated, and in other cases, information indicating the auxiliary storage device 1600 is designated. The number of pages to be moved 205 is an area for storing the number of pages to be moved and released.

The swapping space designation area 201 and the swapping type 202 are designated by the swapping starting unit 1400. For other information, swapping management section 6
Described with 00.

FIG. 3 shows the free page number information 30 of this embodiment.
It is a figure which shows the structure of 0. As shown in FIG. 3, the free page number information 300 is stored in the first area 120 of the main storage device 1200.
The number of empty pages of 1 is held in the number of empty pages 301 of the first area, and the number of empty pages of the second area 1202 of the main memory 1200 is held in the number of empty pages 302 of the second area.

FIG. 4 is a diagram showing the structure of the threshold value 400 of this embodiment. As shown in FIG. 4, the threshold value 400 holds the threshold value of the number of empty pages that considers the first area 1201 of the main memory 1200 as insufficient in the threshold value 401 of the first area, and the second area of the main memory 1200. The threshold value 402 of the second area holds the threshold value of the number of empty pages which considers the area 1202 to be insufficient.

FIG. 5 is a diagram showing the structure of the main memory allocation management table 500 of this embodiment. As shown in FIG. 5, the main memory allocation management table 500 is provided in space units and stores information for managing the storage resources used. The space identifier 501 in FIG. 5 indicates which space the management table corresponds to, and the swapping status information storage area 502 indicates whether the space is swapped in or swapped out. It shows whether it is in the state of being.

The page number 503 of the first type page stores the number of pages of the first area 1201 allocated as the first type page 1501 in the space. First holding area 504 of page management table of first type page
Is a region for storing a pointer that points to the beginning of the page management table of the page of the first region 1201 allocated as the first type page 1501 in that space, and a pointer to the next table is assigned to each table. By storing, it sequentially points to a plurality of page management tables. Further, it is assumed that the page management table stores information indicating to which page in the real memory the page in the virtual memory is allocated.

The page number 505 of the second type page stores the number of pages allocated as the second type page 1502 in the space. The head management area 506 of the page management table of the second type page is an area for storing an address indicating the head of the page management table of the page allocated as the second type page 1502 in the space.

The number of pages of the moved first type page 507
Stores the number of pages of pages that are allocated as the first type of page 1501 in the space and are intermediately swapped out to the second area 1202. The head holding area 508 of the page management table of the moved first type page is allocated as the first type page 1501 in the space,
This is an area for storing the address indicating the beginning of the page management table of the page swapped out to the second area 1202.

FIG. 6 is a flow chart showing the processing procedure of the swapping management processing of this embodiment. As shown in FIG. 6, when the swap-out is requested, the swapping management unit 600 of the computer 1000 indicates that the number of free pages in the first area 1201 of the main storage device expressed by a specific address width is the first area. If the number of empty pages in the second area 1202 is equal to or larger than the number of pages targeted for swap-out, the swap-out area is set to the second area 1202.

The swapping starting unit 14 of the computer 1000
00 requests swapping to the swapping unit 100 when the I / O wait time of the program loaded in the main storage device 1200 reaches a predetermined value or when the resource usage rate reaches a predetermined ratio.

In step 601, the swapping management unit 600 of the swapping unit 100 causes the swapping starting unit 14 to
When the swapping request is received from 00, it is determined from the swapping type 202 of the input parameter 200 whether or not the request is a swap-out. If the result of the above determination is true, the process proceeds to step 602.

In step 602, the threshold value of the first region is set to 4
01 and the number of empty pages 301 in the first area and the threshold value 402 of the second area and the number of empty pages 30 in the second area
2 is compared to determine whether only the first area 1201 is out of memory, and if the result of the above determination is true, the process proceeds to step 603.

In step 603, the information 203 indicating the first area 1201 is set in the type 203 of the page to be moved. Next, in step 604, it is necessary to set the number of free pages 301 in the first area to the threshold 401 of the first area or more from the difference between the threshold 401 of the first area and the number of free pages 301 in the first area. The number of pages to be moved is calculated and set to the number of pages to be moved 205.

At step 607, the number of empty pages 302 in the second area is compared with the number of pages 205 to be moved, and the empty pages necessary for moving are found in the second area 12.
It is determined whether or not the result is 02, and if the result of the above determination is true, the process proceeds to step 608. In step 608, information indicating that the area is the second area 1202 is set in the movement destination 204, and thereafter, the processing of step 610 and thereafter is continued.

On the other hand, if the result of the determination in step 602 is false, the process proceeds to step 605, in which the type 203 of the page to be moved is the first area 1201 and the second area 1
Information indicating that the number is 202 is set.

In step 606, the threshold value of the first region is set to 4
01 and the free page number 301 of the first area, the free page number 301 of the first area is set to the threshold value 40 of the first area.
The number of pages required to be 1 or more is calculated, and the number of empty pages 302 in the second area is calculated from the difference between the threshold value 402 of the second area and the number of empty pages 302 in the second area. The number of pages required to be equal to or greater than the threshold value 402 is calculated and set as the number of pages to be moved 205.

After step 606, or step 60
When the result of the determination in 7 is false, the process proceeds to step 609, and information indicating that the auxiliary storage device 1600 is set in the movement destination 204.

In step 610, referring to the swapping status information storage area 502 in the main memory allocation management table 500 of the swapping target space indicated by the swapping space designation area 201, the swapped-in status of the space If the result of the above determination is true, the process proceeds to step 611, and the space deactivation process is called. If the result of the determination in step 610 is false, the process proceeds to step 612 and the main memory allocation process is called. This completes the swapping management process.

FIG. 7 is a flow chart showing the processing procedure of the space deactivation processing of this embodiment. As shown in FIG. 7, the space deactivating unit 700 of the computer 1000 stops the processing of the swap-out target program in response to the swap-out request as necessary, and executes the program in the CPU 11
The process of making the state of no allocation of 00 is performed.

In step 701, the space deactivating section 700 of the computer 1000 sets, in the swapping state information storage area 502, information indicating that the space is deactivated and swapped out.

Next, at step 702, the input / output processing being processed issued from the space to be deactivated is stopped. In step 703, the execution of the job operating in the space to be deactivated is stopped, and the C
Prevents PU resource allocation. Then, in step 704, the main memory allocation processing is called. This completes the space deactivation process.

FIG. 8 is a flow chart showing the processing procedure of the main memory allocation processing of this embodiment. As shown in FIG. 8, the main memory allocation unit 800 of the computer 1000 moves the number of pages 205 to be moved when swap-out is requested.
When the swap-in is requested, the pages in the main storage device are released after the pages of the number of pages set in the Is secured and the pages saved in the swap-out area are recovered.

In step 801, the main memory allocation unit 800 of the computer 1000 determines from the swapping type 202 of the input parameter 200 whether or not the request is swap out, and if the result of the above determination is true, Proceeds to step 802.

In step 802, it is determined whether the type 203 of the page to be moved is the first type page 1501 and the destination 204 is the second area 1202, and the result of the above determination is true. In this case, the following processing is performed.

That is, in step 803, the second area 1202 of the number specified by the number of pages to be moved 205 is set.
Page of the main memory allocation management table 500
The value of the page number 507 of the moved first type page in
Increment by the number of secured pages. Then, the page management table of the secured page is connected to the end of the page management table sequence obtained by tracing the pointer of the head holding area 508 of the page management table of the moved first type page.

Next, at step 804, the first type of pages 15 having the number of pages designated by the number of pages to be moved 205 are set.
The contents of 01 are copied to the page secured in step 803. Then, information indicating to which page of the second area 1202 on the real storage the first type page 1501 on the copied virtual storage has been intermediately swapped out is stored in the secured page management table.

Next, in step 805, step 804
In the page management table of the first type page, the number of pages of the first type page 503 is subtracted after releasing the page in the main memory allocated to the first type page 1501 whose contents are copied in The page management table of the released page is removed from the page management table sequence obtained by tracing the pointer in the head holding area 504 of the.

On the other hand, if the result of the determination in step 802 is false, the process proceeds to step 806, and the page specified by the type 203 of the page to be moved is supplemented by the page number specified by the page number 205 to be moved. Storage device 16
Output to 00 and release those pages.

If the result of the determination in step 801 is false, the process proceeds to step 807 to perform the following processing. That is, in step 807, the second area 1202
First Page 150
1 is returned to the first area 1201. That is, the head holding area 50 of the page management table of the moved first type page
By referring to the page management table sequence obtained by tracing the pointer of No. 8, the storage location of the first type page 1501 that is intermediate swapped out to the second area 1202 is read, and the page of the first type page that has been moved The first type page 1501 of the equation 507 is copied back from the second area 1202 to the first area 1201. Also the first type of page 1501
Is physically swapped out to the auxiliary storage device 1600, it is returned to the first area 1201 from there.

Next, in step 808, the auxiliary storage device 1
The second type page 1502 physically swapped out to 600 is returned to the second area 1202, and finally, in step 809, the space activation processing is called. With this, the main memory allocation processing ends.

FIG. 9 is a flow chart showing the processing procedure of the space activation processing of this embodiment. As shown in FIG. 9, the space activating unit 900 of the computer 1000 restarts the processing of the swap-in target program to execute CP processing of the program.
U1100 is allocated.

At step 901, the space activation unit 900 of the computer 1000 restarts the execution of the job operating in the space to be activated so that the CPU resource is allocated to the space.

Next, in step 902, the input / output process being processed in the space to be activated, which was stopped in step 702 of the space deactivating process, is restarted, and step 903 is executed.
Then, in the swapping state information storage area 502, information indicating that the space is activated and is in the swap-in state is set. With this, the space activation process is completed.

As described above, according to the computer system of this embodiment, when swapping is requested, swapping is performed on pages having the number of pages set by the difference between the predetermined threshold value and the number of empty pages in the main storage device. Since it does, it reduces the absolute amount of data transfer when swapping,
As a result, it is possible to improve the throughput and response time of actual work.

Further, according to the computer system of the present embodiment, when the free space in the first area of the main storage device expressed by the specific address width is insufficient, the page in the first area is Intermediate swapping to move to the area reduces the data transfer time required for swapping a program that can operate only in a specific area expressed by a specific address width, thereby improving the throughput and response time of actual work It is possible to

[0075]

According to the present invention, when the swapping is requested, the swapping is performed for the page of the number of pages set by the difference between the predetermined threshold value and the number of empty pages of the main storage device. It is possible to reduce the absolute amount of data transfer, thereby improving the throughput and response time of actual work. Further, according to the present invention, when there is insufficient free space in the first area of the main memory represented by a specific address width, intermediate swapping for moving a page in the first area to the second area is performed. Since this is done, it is possible to shorten the data transfer time required for swapping a program that can operate only in a specific area expressed by a specific address width, and thereby improve the throughput and response time of actual work.

[Brief description of drawings]

FIG. 1 is a diagram showing a schematic configuration of a computer system of this embodiment.

FIG. 2 is a diagram showing a configuration of an input parameter 200 of this embodiment.

FIG. 3 is a diagram showing a configuration of empty page number information 300 according to the present embodiment.

FIG. 4 is a diagram showing a configuration of a threshold value 400 of this embodiment.

FIG. 5 is a main memory allocation management table 50 of this embodiment.
It is a figure which shows the structure of 0.

FIG. 6 is a flowchart showing a processing procedure of swapping management processing according to the present embodiment.

FIG. 7 is a flowchart showing a processing procedure of space deactivation processing according to the present embodiment.

FIG. 8 is a flowchart showing a processing procedure of main memory allocation processing of the present embodiment.

FIG. 9 is a flowchart showing a processing procedure of space activation processing of the present embodiment.

[Explanation of symbols]

100 ... Swapping unit, 200 ... Input parameters, 30
0 ... Free page number information, 400 ... Threshold value, 500 ... Main memory allocation management table, 1000 ... Computer, 1100 ...
CPU, 1200 ... Main storage device, 1201 ... First area, 1202 ... Second area, 1300 ... OS, 1400
... Swapping start-up unit, 1500 ... Virtual storage space, 15
01 ... first type page, 1502 ... second type page, 1
600 ... Auxiliary storage device, 1601 ... Auxiliary storage device, 17
00 ... initial setting parameter, 600 ... swapping management section, 700 ... space deactivating section, 800 ... main memory allocating section, 900 ... space activating section, 201 ... swapping space designation area, 202 ... swapping type, 203 ... move target Type of page, 204 ... Destination, 205 ... Number of pages to be moved, 301 ... Number of empty pages in first area, 302
... number of empty pages in second area, 401 ... threshold value of first area, 402 ... threshold value of second area, 501 ... space identifier,
502 ... Swapping state information storage area, 503 ... Page number of first type page, 504 ... Head holding area of page management table of first type page, 505 ... Page number of second type page, 506 ... Second Top holding area of the page management table of the first type page, 507 ... Number of pages of the first type page that has moved, 508 ... First holding area of the page management table of the first type page that has moved.

   ─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Masahiro Kiyoi             1099 Ozenji, Aso-ku, Kawasaki City, Kanagawa Prefecture             Ceremony company Hitachi Systems Development Laboratory (72) Inventor Akira Otsuji             5030 Totsuka Town, Totsuka Ward, Yokohama City, Kanagawa Prefecture             Ceremony company Hitachi Ltd. software division (72) Inventor Naoko Ikegaya             1099 Ozenji, Aso-ku, Kawasaki City, Kanagawa Prefecture             Ceremony company Hitachi Systems Development Laboratory F-term (reference) 5B005 JJ12 MM31 VV02

Claims (6)

[Claims] 1. A main storage device and an auxiliary storage device are provided,
In the swapping control method that controls swapping in the virtual memory system, when the number of free pages in the main storage device is less than the predetermined threshold when swap-out is requested, the predetermined threshold and the number of free pages in the main storage device And a step of setting the difference as the number of pages to be swapped out, and for the swapout request, the processing of the program to be swapped out is stopped so that the CPU is not allocated to the program. A step of saving the set number of pages in the swap-out area and releasing the saved page in the main storage device; and a step in the main storage device when swap-in is requested. Securing the page and recovering the page saved in the swap-out area, and Swapping control method characterized in that by resuming the processing of the in-target program and a step of the state of the program allocation of CPU is performed. 2. When a swap-out request is made and the number of free pages in the first area of the main storage device represented by a specific address width is less than a predetermined threshold, the predetermined threshold and the first threshold are set. The difference from the number of free pages in one area is set as the number of pages to be swapped out, and the number of free pages in the second area located at an address higher than the first area is greater than or equal to the set number of pages. The swapping control method according to claim 1, wherein the swap-out area of the set number of pages is set as the second area. 3. When the number of free pages in the second area is less than a predetermined threshold, the difference between the predetermined threshold and the number of free pages in the second area is set as the number of swap-out target pages. 3. The swapping control method according to claim 2, wherein the swap-out area of the set number of pages is used as an area on the auxiliary storage device. 4. The CPU is referred to by referring to an area holding information indicating whether or not a step of setting a state in which the CPU is not allocated to the program has been executed.
When the information of the area indicates that the step of bringing the CPU into the non-allocated state is not executed, the step of putting the CPU into the non-allocated state is performed for the program. Claim 1
To the swapping control method according to claim 3. 5. A main storage device and an auxiliary storage device are provided,
In a computer system that controls swapping in the virtual memory system, if the number of free pages in the main memory is less than the predetermined threshold when swap-out is requested, the specified threshold and the number of free pages in the main memory The swapping management unit that sets the difference as the number of pages to be swapped out, and the processing of the program to be swapped out is stopped as necessary in response to the swapout request so that the program is not allocated to the CPU. When the swap-out is requested, the space deactivation unit that performs the swap-out is performed to save the set number of pages to the swap-out area and to release the saved page in the main storage device. Is requested, a page on main memory is secured and a page saved to the swap-out area is saved. Computer system comprising a main storage allocation unit that performs di recovery, that by resuming the processing of the swap-target program and a space activation unit to a state in which the program allocation of CPU is performed. 6. A main storage device and an auxiliary storage device are provided,
In a program that causes a computer to function as a computer system that controls swapping in the virtual memory system, if the number of free pages in the main storage device is less than a specified threshold when swap-out is requested, the specified threshold A swapping management unit that sets the difference from the number of free pages in the main storage device as the number of pages to be swapped out, and the processing of the program to be swapped out is stopped as necessary in response to a swapout request, and the program is executed by the CPU. Space deactivating unit that puts the specified number of pages in the swap-out area and swap-out is requested on the main storage device when swap-out is requested. When a page is released and swap-in is requested, the page on main storage is A computer as a main memory allocation unit that performs storage and recovery of the pages saved in the swap-out area, and a space activation unit that restarts the processing of the program that is the swap-in target and puts the program into a state where the CPU is allocated. A program characterized by making a function of.