JP2003316589A - Real memory using method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve the problem that delay due to the concentration of access is generated, or that the change of a user program is necessary for adopting a method for preventing the generation of the delay in a system where a shared data is disposed in an external storage device. <P>SOLUTION: A selected arbitrary computer in a virtual computer system is provided with a step for converting an address designated in an input/output request command transmitted from another computer to an external storage device by a communicating means into an address in a corresponding data region in a real storage device held in the selected logical block, a step for, when the command is an input request, returning the data stored in the data region of the converted address to a computer being the origin of transmission of the command, and a step for, when the command is an output request, storing the data transmitted with the command in the data region of the converted address. <P>COPYRIGHT: (C)2004,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of using a large-capacity real storage device, and more particularly to a method of using a real storage device in a virtual computer system capable of virtually executing a plurality of computers.

[0002]

2. Description of the Related Art In recent years, backbone server systems employ a cluster type system configuration composed of a plurality of computer images in order to flexibly cope with enormous amounts of data and processing. Is common. Further, in many computer systems, data that needs to be shared between computer images is arranged in a shared external storage device that can be accessed from each computer image.

However, since access requests are issued from a plurality of computer systems to shared data, it is necessary to control data access while executing exclusive processing in order to guarantee data consistency. In general, the control information used for such exclusive processing is also arranged in the shared external storage device. Therefore, in order to improve the response performance and throughput of the system, it is necessary to have an environment in which the data for exclusive processing arranged in the external storage device can be accessed at high speed.

However, in the external storage device, since the target area is accessed while rotating the disk,
When access is concentrated on the shared external storage device, waiting for the disk to rotate causes a delay in access.

As a means for solving this, Japanese Patent Laid-Open No. 9-163
There is -293055. In Japanese Patent Laid-Open No. 9-293055,
An extended memory is arranged in place of the shared external storage device, a shared file is arranged in this memory area, and a lock area used for exclusive processing is provided. By doing this,
While avoiding an access delay due to a rotation wait at the time of accessing shared data, the efficiency of exclusive processing is improved by speeding up access to control information.

[0006]

However, in order to use the above conventional technique, it becomes necessary to newly prepare an application program. In a system in which shared data is placed in a shared external storage device, a program is created to access the external storage device. But,
In order to utilize the above prior art, it needs to be modified to access the shared memory.

As a method for solving the above-mentioned problems, the operating system supports a function of accessing the above-mentioned shared memory by using an input / output interface to an external storage device prepared by the operating system. A method can be envisioned.
In this method, the operating system traps the input / output request issued by the application program to access the external storage device, and changes the access to the shared memory. Therefore, the operating system running on each computer node needs to support a function of accessing the shared memory by using an input / output interface to an external storage device. A huge amount of development is required for the operating system to support these functions.

A first object of the present invention is to share an application program with a relatively small operating system development amount while guaranteeing the operation of an interface for accessing an external storage device in the conventional manner. It is to provide a method for improving the response performance and throughput of a system by speeding up access to data.

A second object of the present invention is to improve the confidentiality of data by allocating such data in another logical partition area for accessing the data.

[0010]

In order to solve the above problems, in a system capable of constructing a plurality of computer images on one computer system by logical partitioning, the selected computer image is The address specified in the input / output request command to the external storage device sent from the other computer image to the external storage device is the address of the corresponding data area in the real storage device held by the selected logical partition. The step of converting to. When the input / output request command is an input request, the data stored in the data area of the address obtained in the converting step is returned to the computer image that is the source of the input / output request command, and the input / output request command is sent. When is an output request, a step of storing the data sent together with the input / output request command in the data area of the address obtained by the converting step is provided.

In addition to the selected computer image, there is provided a step of generating an input / output request command to the external storage device from the input / output request of the data issued by the operating program. Further, when this input / output request command is an input request for data stored in any of the selected computer images, the input / output request command is changed to the computer image in which the data to be accessed is stored. If the input / output request command is an output request to any of the selected computer images, a step of sending the input / output request command and data to the storage destination computer image by communication is provided.

This solution can also be applied to a cluster computer system composed of a plurality of computer nodes that can communicate with each other.

[0013]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 shows the configuration of a computer system showing an embodiment of the present invention. In FIG. 1, 100
0 and 2000 are logical partition 1 and logical partition 2, respectively. Reference numeral 3000 is a hypervisor that allocates resources in the computer system to each logical partition and performs necessary communication between the logical partitions. Logical partition 1 to which resources in the computer system have been allocated by the hypervisor (3000)
By executing the programs stored in (1000) and the logical partition 2 (2000) by the control unit (CPU), the programs can be executed as a plurality of virtually independent computers.

Reference numeral 4000 is a physical disk device connected to the computer system, and in this embodiment, it is assigned to the logical partition 2 (2000) by the hypervisor (3000).

A virtual volume 1100 is created in the real storage device assigned to the logical partition 1 (1000). The virtual volume creation program (1200) creates this virtual volume (1100).

A virtual volume access execution program 1300 receives a virtual volume (1100) access request from the logical partition 2 (2000).
00) area, write data, read data, or calculate processing target data position.

A virtual volume management table 1400 stores various information necessary for managing the virtual volume (1100).

Reference numeral 1500 is a communication program for executing communication processing with an external computer via another logical partition or network. The communication program (1500) is
The process of transmitting communication from the logical partition 1 (1000) to another computer, and the process of transmitting communication from the other computer to the logical partition 1 (100)
0), and executes a predetermined program according to the communication content. When the communication content is an access request for the virtual volume (1100), the above-mentioned virtual volume access execution program (1200) is activated. Further, after the virtual volume access processing is completed, the response is transmitted to the requesting logical partition 1 (1000) by the communication program (15
00).

Next, the configuration contents of the logical partition 2 will be described. Reference numeral 2100 is an application program (hereinafter, AP) executed in the logical partition 2 (2000). AP
(2100) includes an execution request for input / output processing to the physical disk (4000) or virtual volume (1100). 2110 in the AP (2100) is positioning information indicating a processing target place in the physical disk (4000) or the virtual volume (1100) when executing the input / output processing. Reference numeral 2120 is an input / output buffer that stores data for input / output processing.

Reference numeral 2200 is a data management program that receives data input / output requests from the AP (2100) and executes actual input / output processing.

Reference numeral 2300 is a virtual volume access request program. The virtual volume access request program (2300) is called by the data management program (2200) when the data input / output processing target requested by the AP (2100) is a virtual volume. Then, the communication program (2500) is requested to execute communication with the logical partition 1 (1000) that is the management source of the virtual volume.

Reference numeral 2400 is a virtual volume access completion program. When access to the virtual volume (1100) is completed, the virtual volume (1100)
From the logical partition 1 (1000) that is the management source of the above, a response indicating the completion of processing is transmitted. When the communication program (2500) of the logical partition 2 (2000) receives this access completion communication, it executes the virtual volume access completion program (2400). When the input / output request from the AP (2100) is a data read request, the virtual volume access completion program (2400) copies the data read from the virtual volume (1100) to the input / output buffer area.

The communication program (2500), like the communication program (1500) in the logical partition 1 (1000), executes communication processing with an external computer via another logical partition or a network.

Next, the configuration contents of the hypervisor (3000) will be described. 3100 is a computer resource allocation program. The computer resource allocation program (3100)
Resources such as a real storage device and a processor that a computer has are allocated to the logical partition 1 (1000) and the logical partition 2 (200
0) statically or dynamically. This provides an environment in which the logical partition 1 (1000) and the logical partition 2 (2000) can operate as independent computer system images.

Reference numeral 3200 is a communication program between logical partitions in the hypervisor (3000). The inter-logical partition communication program (3200) transfers communication data to the requested logical partition when a communication request is made from the logical partition 1 (1000) to the logical partition 2 (2000) or vice versa.

FIG. 2 shows the flow of control when an input / output request is issued from the AP (2100) to the virtual volume (1100). Data management program (220
0) is a virtual volume access request program (23
00) and then the communication program (2500)
Causes an access request to be sent to the logical partition 1 (1000). This communication is passed to the communication program (1500) in the logical partition 1 (1000) via the inter-logical partition communication program (3200) of the hypervisor (3000).

Logical partition 1 (1000) that received the request
On the side, the virtual volume access execution program (12
00) and starts the virtual volume (11
After writing data to or reading data from (00), a response is returned to the requesting logical partition 2 (2000).

Logical partition 2 (200) that received the response information
The communication program (2500) in 0) activates the virtual volume access completion program (2400), and thereafter, via the data management program (2200), the AP
The process of (2100) is restarted.

Next, the configuration of the virtual volume management table (1400) will be described with reference to FIG. The virtual volume management table (1400) contains the virtual volume (1
100) using the same interface as the physical disk 4000, and information for managing the access location in the virtual volume 1100 at each time point. Here, when accessing the data in the physical disk (4000), the location of the data in the physical disk (4000) is a cylinder number, a header number, and an in-track record determined by the cylinder number and the header number. Identified by a number. The number of cylinders, the number of headers, and the track length of each volume are physically determined for each device.

In order to deal with such an access to the physical disk (4000) and a common interface, the virtual volume management table (1400) has a virtual cylinder number (1410), a virtual header number (1420), a virtual track length ( 1430) is stored. The values stored in these fields determine the size of the data area forming the virtual volume (1100).

The virtual volume management table (14
The record length (1440) in 00) is AP (210
This area stores the record length that represents the minimum unit of data accessed from 0). Further, the data head pointer (1450) is a field for storing the head address in the real storage device which secures the area of the virtual volume (1100). The access pointer (1460) indicates the virtual volume (1100) to be accessed at each time point.
This is a field for storing an address indicating a data position in the.

Virtual volume management table (1400)
The virtual volume generation program (1300) performs the initial setting of each field constituting the virtual volume (1100) and secures the data area of the virtual volume (1100). FIG. 4 is a diagram showing a state in which the virtual volume generation unit (1300) has initialized the virtual volume management table (1400). Here, the number of virtual cylinders (1410) is 1024
A virtual volume (1100) having 16 virtual headers (1420) and a virtual track length (1430) of 32768 bytes is generated. In addition, a record length (144) which is a unit of data input / output to be stored in each track.
As an initial value of 0), 4096 bytes are set.

In this embodiment, the virtual volume (11
00) is a data area that is logical partition 1 (100
It is continuously secured in the real storage device 0). The data arrangement order is ascending order of cylinder numbers and ascending order of track numbers in each cylinder. However, both the cylinder number and the track number start from 0. In the description of the virtual volume (1100) data area in FIG. 4, “cylinder i
"Track j" of "means a data area corresponding to" track having cylinder number i and header number j ".

In FIG. 4, the virtual volume management table (1
The value of the data start pointer (1450) in 400) is 4
It is set to 096. That is, in this embodiment, the head address that secures the data area for the virtual volume (1100) is the address 4096 on the real storage device of the logical partition 1 (1000).

In FIG. 4, the value of the access pointer (1460) is set to 528384. That is, when the access request to the virtual volume (1100) is generated next, the data located at the address 528384 on the real storage device of the logical partition 1 (1000) is processed. Here, the address 528384 coincides with “data head pointer (2450) + number of virtual headers (2420) × virtual track length (2430)”, and the virtual volume (1
100) corresponds to cylinder 1 and track 0.

FIG. 5 shows the structure of the volume identification table (2600) in the logical partition 2 (2000).
The volume identification table (2600) is the logical partition 2
The list of disk devices accessible by the AP (2100) executed in (2000) is shown as the volume ID (26
10) and the volume type (2620). This information is stored as the system configuration information, which is registered in advance by the user according to the parameters at the time of starting the system.

The volume ID (2610) is information used to identify the disk to be accessed.
In the present embodiment, the order of registration is 1, 2 ,. ． ． Each disk is identified by an ascending integer. It is assumed that the order of registration in the system is a physical disk (4000) and a virtual volume (1100). That is, volume ID
(2610) = 1 corresponds to the physical disk (4000), and volume ID (2610) = 2 corresponds to the virtual volume (1100).

The volume type (2620) is the type of the disk device indicated by the volume ID (2610).
This is a field for storing information for identifying whether it is a physical disk or a virtual volume. In this embodiment, the volume type (2620) = P indicates a physical disk, and the volume type (2620) = V indicates a virtual volume.

FIG. 6 shows the AP in the logical partition 2 (2000).
A channel command word (hereinafter CC) that stores information necessary to execute the input / output operation requested from (2100).
W) is shown. Reference numeral 5000 indicates the structure of a channel command word generated when requesting processing to the physical disk (4000). CCW (500
0) is generated by the data management program (2200). Reference numeral 5100 indicates the configuration of the communication CCW used for communication with the logical partition 1 (1000) when an access request is made to the virtual volume (1100). The communication CCW (5100) uses the data management program (22
00) to generate the virtual volume access request program (2300).

CCW (5000) is command (501)
0) and a data address (5020). The command (5010) is an area indicating what kind of processing is requested to the input / output device, and in the present embodiment, three types of positioning, writing and reading can be designated. this house,
The positioning command is a command for designating position information (positioning information) in the disk device and designating a data position to be processed next.

The data address (5020) is an area indicating an address on the real storage device in which the data to be processed exists. In the case of a positioning request, the address where the above-mentioned positioning information exists is the data address (502
0).

When the target is the virtual volume (1100), the virtual volume access request program (2300) uses the communication CCW (5) based on the CCW (5000).
100) is generated. Furthermore, the virtual volume (110
0) Even after the access is completed, when the response from the virtual volume access execution program (1200) is returned, this communication CCW
(5100) is generated. Communication CCW (5100)
Is the command (5110), data address (512
0) and data (5130). Of these, the command (5110) and data address (512
0) is the command (50) of CCW (5000).
10) and has the same meaning as the data address (5020). The data (5130) is an area for storing the data body when the data body also needs to be transferred according to the request type indicated by the command (5110).

Hereinafter, the CCW (5000) and the communication CCW (51) when accessing the virtual volume (1100)
00) will be described in the order of positioning request, write request, and read request.

FIG. 7 shows a state on the side of the logical partition 2 (2000) when the AP (2100) issues a positioning request to the virtual volume (1100). CCW (500
The address of the positioning information (2110) is stored in the data address (5020) of 0). This positioning information (2110) exists in the real storage device of the logical partition 2 (2000). Therefore, the contents cannot be read directly from the logical partition 1 (1000). Therefore, the positioning information (6000) indicated by the data address (5020) is stored in the data (5130) of the communication CCW (5100) for the logical partition 1 (1000).
(5100) is transmitted to the logical partition 1 (1000).

FIG. 8 shows the state on the side of the logical partition 1 (1000) that has received the positioning request communication. Logical partition 1 (1
000) side, data (5 in communication CCW (5100))
The access pointer (1460) in the virtual volume management table (1400) is updated based on the positioning information stored in 130). Since the original positioning information has a format that describes the placement location in the physical disk device, this information is the address in the real storage device where the virtual volume (1100) is placed in the logical partition 1 (1000). Need to be converted to. For details of this conversion processing, when the processing flow chart of the virtual volume access execution program (1200) is described later,
explain in detail.

After the positioning process is completed, the logical partition 2
As a response to (2000), the command (511
0), data address (5120), data (513
The communication CCW (5100) having the same content as that of 0) is generated and the logical partition 2 (2000) is communicated.

Next, with reference to FIGS. 9 and 10, a state in which the writing process to the virtual volume (1100) is performed will be described. FIG. 8 shows a state on the side of the logical partition 2 (2000) that requests writing of data. The data address (5020) in the CCW (5000) indicates the address of the input / output buffer (2120) in which the data to be written from the AP (2100) to the virtual volume is stored. The contents of this input / output buffer (2120) are changed to data (513
The communication CCW (5100) stored in 0) is generated and transmitted to the logical partition 1 (1000).

FIG. 10 shows the state on the side of the logical partition 1 (1000) that has received the write request. Communication CCW
The content stored in the data (5130) in the (5100) is used as the access pointer (1 in the virtual volume management table (1400) of the virtual volume (1100).
Copy to the data area (1110) indicated by 460). Communication C of response to logical partition 2 (2000)
The CW (5100) does not need to store information in the data (5130).

Next, with reference to FIG. 11 to FIG. 13, a state when a read request is made to the virtual volume (1100) will be described.

FIG. 11 shows the logical partition 2 requesting reading.
The state on the (2000) side is shown. Logical partition 1 (1000)
The contents of the command (5000) and the data address (5100) of the communication CCW (5100) transmitted to the CCW (5000) are the same as those of the CCW (5000). Data (513
It is not necessary to store information in 0) at the time of access request.

FIG. 12 shows the state on the side of the logical partition 1 (1000) that has received the read request communication. The virtual volume (110) indicated by the access pointer (1460)
0) from the data area (1110) in the communication CCW (5
Data is copied to the data (5130) in 100) and the communication CCW (5100) is responded to the logical partition 2 (2000).

FIG. 13 shows the state on the side of the logical partition 2 (2000) that has received the read response communication. Communication CC
The content stored in the data (5130) in the W (5100), that is, the data read from the virtual volume (1100) indicates the AP indicated by the data address (5120).
Copy to the input / output buffer (2120) in (2100).

FIG. 14 is a processing flowchart of the virtual volume access execution program (1200) in the logical partition 1 (1000) which is executed when the access request to the virtual volume (1100) is received.

First, as a preparation for response communication, the command (5110) and data address (5120) in the requested communication CCW (5100) are set to the communication CCW for response communication.
The command (5110) of (5100) and the data address (5120) are stored (step 1210).

Next, the requested command type is judged (step 1220). Here, there are three types of commands: positioning, writing, and reading. If the command type is a positioning request, the communication CCW (510
0) The positioning information stored in the internal data (5200) is converted into the address of the real storage device in the logical partition 1 (1000) storing the virtual volume (1100) (step 1230). The designation designated as the positioning information is common with the information for accessing the physical disk (4000), and is indicated in the format of "CCHHRR". Here, CC is a cylinder number, HH is a header number, and R is a record number. For such a specification,
As the data allocation destination in the virtual volume (1100),
"Data start pointer (2450) + (CC x number of virtual headers (2420) + HH) x virtual track length (243
0) + R × record length (2440) ”is converted to the address of the data area in the virtual volume (1100). Furthermore, the converted data is
Stored in access pointer (1460) (step 124)
0)

If the command type is a write request,
The content of the data (5130) in the communication CCW (5100) is copied to the data area in the virtual volume (1100) indicated by the access pointer (1460) (step 125).
0) If the command type is a read request, the content is copied from the data area in the virtual volume (1100) indicated by the access pointer (1460) to the data (5130) in the communication CCW (5100) for response communication (step). 1260).

Finally, the communication CCW (5100) for response communication created in the above steps is set to the requesting logical partition 2
The communication program (1500) is requested (step 1270) to transmit to (2000).

FIG. 15 is a processing flowchart of the data management program (2200) executed when an input / output request is received from the AP (2100) in the logical partition 2 (2000).

First, a CCW for executing a request to an input / output device is created (step 2210). Next, the volume ID of the device that executes the input / output request is acquired (step 2220). Subsequently, it is determined whether the I / O device to be accessed is a virtual volume or a physical disk (step 2230). This determination process is
This can be realized by referring to the volume identification table (2600) and checking the volume type (2620) corresponding to the volume ID (2610) acquired in step 2220.

When the access target is a physical disk, an input / output activation command is executed for the corresponding input / output device (step 2240). The process of accessing this physical disk device is the same as that of the computer system that does not create the virtual volume (1100) in the logical partition 1 (1000).

If it is determined in step 2230 that the access target is the virtual volume (1100), the logical partition 1 (100) in which the virtual volume (1100) exists.
0), the virtual volume access request program (2300) is executed (step 225).
0)

FIG. 16 is a processing flowchart of the virtual volume access request program (2300).

CCs created by the data management program (2200) for the command (5110) and data address (5120) that make up the communication CCW (5100), respectively.
The same information as the command (5010) and data address (5020) in W (5000) is stored (step 231).
0) Next, the command type is determined (step 232
0) If the result of determination is that the command type is positioning or writing, the contents of the area indicated by the data address (5120) are stored in the data (5130) of the communication CCW (5100) (step 2330). This information is positioning information (6000) in the case of a positioning request, and a virtual volume (1100) in the case of a writing request.
The data in the input / output buffer (7000) to be written to.

Request the communication program (2500) to send the communication CCW (5100) created by the above processing steps to the logical partition 1 (1000) in which the virtual volume (1100) exists (step 23).
40)

FIG. 17 shows that when the access to the virtual volume (1100) is completed, the communication program (2500) which receives the communication CCW (5100) responded from the logical partition 1 (1000) is started and the virtual volume access is completed. It is a processing flowchart of a program (2400).

First, from the contents of the communication CCW (5100),
The command type is determined (step 2410). In the same judgment, if the command type is read, communication CC
The content of the data (5130) in the W (5100) is the area in the real storage device indicated by the data address (5120), that is, the input / output buffer (70) designated by the AP (2100).
Copy (2420) to 00). If it is determined in step 2410 that the command type is positioning or writing, data copying from the communication CCW (5100) is not necessary.

Subsequently, the AP (2100) of the input / output request source
Data management program (220
0) is executed (step 2430).

Next, a second embodiment using the present invention will be described with reference to FIG.

In the second embodiment, the logical partition 1 (10
00), logical partition 2 (2000), and logical partition 3
(6000) and a physical disk (7000) assigned to the logical partition 3 (6000). Logical partition 3
The configuration content of (6000) is the same as that of the logical partition 2 (2000), and the description thereof is omitted. With such a configuration, the AP executed in the logical partition 2 (2000)
(2100) and the AP (6100) executed in the logical partition 3 (6000) can share the data in the virtual volume (1100).

Next, a third embodiment using the present invention will be described with reference to FIG.

The computer system according to the third embodiment is not a system in which one computer is divided into logical partitions and executed, but a cluster type computer system composed of a plurality of computer nodes. Here, the computer system includes a computer node 1 (8000) and a computer node 2 (900).
0) and. Further, 10000 is a physical disk device connected to the computer node 2 (9000), and 11000 is a communication path for executing data communication between the computer node 1 (8000) and the computer node 2 (9000). is there.

The configuration contents in the computer node 1 (8000) are the same as those of the logical partition 1 (1000) in the first embodiment.
Is the same as. Similarly, the computer node 2 (900
The configuration contents in 0) are the same as those of the logical partition 2 (2000) in the first embodiment. Virtual volume (81
00), the format of the data communicated between the computer nodes is also the communication C described in the first embodiment.
It has the same configuration as the CW (5100). As a result, even in the cluster computer system, the computer node 1 (8
Virtual volume (8
100) to the AP (91 of the computer node 2 (9000)
It is possible to use it from 00). In the configuration in which the number of computer nodes is further increased, the virtual volume (8100)
The data within can be shared among multiple computer nodes.

[0073]

According to the present invention, it is possible to speed up access to shared data without changing the program operating in the system in which the shared data is arranged in the external storage device. , System response performance and throughput can be improved. Also, according to the present invention, even for data that is not shared, by utilizing the memory, it is possible to speed up access and improve the response performance and throughput of the system. Further, according to the present invention, the confidentiality of the data can be improved by arranging such data in another logical partition area for access to the data.

[Brief description of drawings]

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

FIG. 2 is a flow of control in the first embodiment.

FIG. 3 is a configuration diagram of a virtual volume management table according to the first embodiment.

[Fig. 4] State of virtual volume initialization in the first embodiment

FIG. 5 is a configuration diagram of a volume identification table according to the first embodiment.

FIG. 6 is a CCW and a communication CC according to the first embodiment.
Diagram of W

FIG. 7 shows a state of the logical partition 2 when a positioning request is made in the first embodiment.

FIG. 8 is a state of the logical partition 1 when a positioning request is made in the first embodiment.

FIG. 9 is a state of the logical partition 2 when a write request is made in the first embodiment.

FIG. 10 is a state of the logical partition 1 when a write request is made in the first embodiment.

FIG. 11 is a state of the logical partition 2 when a read request is made in the first embodiment.

FIG. 12 is a state of the logical partition 1 when a read request is made in the first embodiment.

FIG. 13 is a state of the logical partition 2 at the time of a read request according to the first embodiment

FIG. 14 is a processing flowchart of a virtual volume access execution program according to the first embodiment.

FIG. 15 is a processing flowchart of a data management program according to the first embodiment.

FIG. 16 is a processing flowchart of a virtual volume access request program according to the first embodiment.

FIG. 17 is a processing flowchart of a virtual volume access completion program according to the first embodiment.

FIG. 18 is a configuration diagram of a computer system showing the second embodiment.

FIG. 19 is a configuration diagram of a computer system showing a third embodiment.

[Explanation of symbols]

1000 ... Logical partition 1, 1100 ... Virtual volume 1200 ... Virtual volume access execution program, 1300 ... Virtual volume generation program, 14
00 ... Virtual volume control table, 1500 ... Communication program, 2000 ... Logical partition 2, 2100 ...
Application program, 2110 ... Positioning information, 2120 ... Input / output buffer, 2200 ... Data management program, 2300 ... Virtual volume access request program, 2400 ... Virtual volume access completion program, 2500 ... Communication program, 300
0 ... Hypervisor, 3100 ... Computer resource allocation program, 3200 ... Logical partition communication program, 40
00 …… Physical disk

   ─────────────────────────────────────────────────── ─── Continued front page    (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 (72) Inventor Yuri Hiraiwa             1099 Ozenji, Aso-ku, Kawasaki City, Kanagawa Prefecture             Ceremony company Hitachi Systems Development Laboratory F term (reference) 5B065 BA01 CA02 CA13 CC02 CC08                       CH04 PA12                 5B098 AA03 GA01 GD03 GD15 HH04                       HH07

Claims (8)

[Claims] 1. A virtual computer system capable of virtually executing a plurality of computers by executing a program stored for each logical partition obtained by logically dividing a physical storage device. And a selection means for selecting execution of a program stored in at least one first logical partition, and at least the first logical partition is capable of communicating with a second logical partition. Receiving means for receiving a data input request command for the physical storage device; and conversion for converting a data address specified in the input request command into a logical data address assigned to the first logical partition. And storage means for storing the data in the physical storage device based on the logical data address. Virtual computer system with butterflies. 2. The virtual computer system according to claim 1, wherein at least the first logical partition is further executed by executing a program stored in the first logical partition selected by the selecting means. Generating means for generating an input request command for the physical storage device based on the issued input request for data to the second logical partition, wherein the communication means includes the input request command and the input Data to be stored in the second logical partition by using the request command,
A virtual computer system characterized by transmitting to a logical partition of. 3. A virtual computer system capable of virtually executing a plurality of computers by executing a program stored for each logical partition obtained by logically dividing a physical storage device. And a selection unit for selecting execution of a program stored in at least one first logical partition, wherein at least the first logical partition is capable of communicating with the second logical partition. Means, receiving means for receiving a data output request command for the physical storage device, and converting a data address specified in the output request command into a logical data address assigned to the first logical partition. A conversion unit, the communication unit, based on the logical data address, the data stored in the physical storage device, A virtual computer system characterized by transmitting to a second logical partition that has issued an input request command. 4. The virtual computer system according to claim 3, wherein at least the first logical partition further executes a program stored in the first logical partition selected by the selecting means. Generating means for generating an output request command for the physical storage device based on the issued data output request for the second logical partition, wherein the communication means outputs the output request command for the second storage area. The virtual computer system is characterized by transmitting to the communication means of the logical partition. 5. A method for utilizing real storage by a virtual computer system, which is capable of virtually executing a plurality of computers by executing a program stored in each logical partition obtained by logically dividing a physical storage device. A selection step of selecting execution of a program stored in at least one first logical partition of the logical partitions, and communication capable of communicating with a second logical partition in at least the first logical partition A reception step of receiving a data input request command for the physical storage device; and converting a data address specified in the input request command into a logical data address assigned to the first logical partition. A step of converting the data to the physical storage device based on the logical data address. Real storage method of use; and a storage step of storing. 6. A method for utilizing real storage by a virtual computer system capable of virtually executing a plurality of computers by executing a program stored in each logical partition obtained by logically dividing a physical storage device. And a selecting step of selecting execution of a program stored in at least one first logical partition of the logical partitions, at least in the first logical partition between the second logical partition and Communication step capable of communicating with the physical storage device, a receiving step of receiving a data output request command for the physical storage device, and a data address specified in the output request command by a logical address assigned to the first logical partition. A conversion step of converting into a data address, the communication step comprising: A method for utilizing actual storage, characterized in that the data stored in the physical storage device is transmitted to the second logical partition that issued the input request command. 7. A computer system having a plurality of computer nodes and a physical storage device shared by the plurality of computer nodes, comprising: selecting means for selecting at least one first computer node among the above. At least the first computer node, communication means capable of communicating with the second computer node, receiving means for receiving a data input request command for the physical storage device, and designated in the input request command Conversion means for converting the data address to be converted into a logical data address assigned to the first logical partition, and a storage for storing the data in the physical storage device based on the logical data address. And a computer system. 8. A computer system having a plurality of computer nodes and a physical storage device shared by the plurality of computer nodes, comprising: selecting means for selecting at least one first computer node among the above. At least the first computer node, communication means capable of communicating with the second computer node, receiving means for receiving a data output request command for the physical storage device, and designated in the output request command A data address to be converted into a logical data address assigned to the first computer node, and the communication means stores the physical data in the physical storage device based on the logical data address. A computer system characterized by transmitting the stored data to the second computer node that issued the input request command. Stem.