JPH11219327A - Network interface and computer system using it - Google Patents

Abstract

(57) [Summary] [PROBLEMS] To provide a computer system in which a client program can immediately detect a stop of a server machine and can immediately proceed to the next response. A network interface (NIA) is provided.
A means for constantly supplying power to the server, a means for detecting whether or not the server on which it is mounted is stopped, and a means for creating a delivery failure notification to the client machine as a transmission destination. If a packet is sent from the network to the NIA while the server is stopped, a delivery failure notification indicating that the server is stopped is immediately sent back to the client.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a computer system using a network.

[0002]

2. Description of the Related Art As shown in FIG. 2, each computer included in the computer system has a network interface (hereinafter referred to as NIA) and is connected to a network. You. The NIA is a device that realizes communication between a computer on which the NIA is mounted and a network.

In the computer system shown in FIG. 2, a server 1 is one of the computers, and usually includes components 9, 9 'comprising one or more processors, main memory, input / output devices, and the like.
…have. The server machine 1 has an NIA 40 and is connected to the network 2 through the NIA 40 and the link 4 connected to the NIA 40.

Although not shown in FIG. 2, the components 9, 9 ',... And the NIA 40 are connected by signal lines and can transfer data to one another. As a result, on the server machine 1, the program stored in a part of the main memory and executed on the processor is NI
A40 sends a packet (also referred to as a datagram, but a packet is used in this specification) to the network 2 under the command of the A40, and receives a packet sent via the network 2 under the command of the NIA 40. Or you can.

The NIA 40 of FIG. 2 will be described in more detail. The NIA 40 has a communication control unit 41, and the communication control unit 41 has a transmission / reception unit 22 and a normal-time communication control unit 24. Here, the transmission / reception unit 22 is connected to the link 4 connected to the network 2,
To achieve a signal connection with In addition, the normal-time communication control unit 24 instructs the transmission / reception unit 22 to transmit the packet or instructs the transmission / reception of the packet from the network 2 to the main memory. , And controls the operation of the transmitting / receiving unit 22.

In the computer system shown in FIG. 2, the client machine 3 is also a computer having the same structure as that of the server machine 1 already described.
A, and connected to the network 2 via link 4. Thereby, in this computer system, the client machine 3 and the server machine 1 can transmit and receive the packet to and from each other.

The components 9, 9 ′,... And the NIA 40 of the server machine 1 shown in FIG.
1 and operates by being supplied with power from the power supply device 7. Here, the power supply device 7 may be outside the server machine 1 like a power line, or may be provided inside the server machine 1 like a battery. Further, a power supply control device 8 is provided between the power supply device 7 and the power supply line 11. The power supply control device 8 is a device for controlling the power supply of the server machine 1, and includes a power switch and the like. The above description regarding the power supply of the server machine 1 is the same for the client machine 3.

Although FIG. 2 shows only two computers, the same applies to a case where three or more computers are used. Each of the computers has an NIA, and the NIA and the link Each is connected to the network 2, and can transmit and receive the packet to and from another computer in the computer system 10.

More specifically, each NIA
Is assigned a unique network address (hereinafter, abbreviated as “address” in the present specification), and communication between a plurality of computers is realized as follows using this address. That is, when the transmitting computer creates a packet to be transmitted, it writes the address of the NIA mounted on the destination computer as a destination address in a specific location in the packet, and sends its own NIA. To send a packet to the network 2. Then
The packet is transferred in the network 2 based on the described transmission address, and the NIA having the address described as a destination in the packet receives the packet.

[0010] A unique address is assigned to each network interface, and the sending computer specifies the destination address, so that the data to be transmitted is delivered to the computer having the NIA assigned the specified address. For a more detailed description of the communication method that realizes the function "
s, "UNIX Network Programming", Prentice-Hal
1, pp. 171-196. In each of the computer systems in this specification, it is assumed that the communication method is established using the address, and further detailed description of the communication method is omitted.

In the computer system of FIG. 2 having the above configuration, a program (client program) 6 executed on the client machine 3 generally executes each step of FIG. The client program 6 is a part or all of an application program operating under the management of an operating system executed on the client machine 3, or a part of an operating system program.
Since these differences are not important in the present invention, they are collectively referred to as a client program in the present specification.

When it becomes necessary to communicate with the server 1 (called server access) in the client program 6, step 100 and subsequent steps in FIG. Is transmitted to the network 2 (101). Next, a timer is set so as to notify when a predetermined time (T) elapses (102). Here, even when the timer is one of the devices mounted on the client machine 3, the timer which is being executed on the client machine 3
In some cases, the function of the system is realized by software, but in any case, after a lapse of time T, the client program 6 is notified. After step 102, the execution of the program is stopped until a notification from the timer is received or a response from the server machine 1 is received in the form of a packet (10).
3).

It should be noted that the address of the server 1 is more precisely one of the NIAs installed in the server 1.
In the present specification, each computer is provided with one NIA, and the address of the NIA is the address of the computer on which the NIA is mounted. This assumption is not important in the present invention, and in an embodiment of the present invention described later, even if a single computer has a plurality of NIAs, it operates without any problem.

Thereafter, upon receiving the notification from the timer or the response from the server machine 1, the execution of the program is resumed, and it is checked whether or not the execution was resumed by the response from the server machine 1 (104). If so, execution of the client program 6 is continued based on the response (106). If not, that is, if the execution has been resumed by the notification from the timer, the client program 6
Is considered to be unable to communicate for some reason (called timeout).

Therefore, if the address of a server other than the server 1 (more specifically, the NIA address of the server) is registered as a server to be accessed, the address is transmitted to the destination. The process returns to step 101 as an address (105). Otherwise, the client program 6 determines that the server access has not been possible (107). The response when the server cannot be accessed depends on the client program 6, and outputs an error message and terminates the processing.
In some cases, the same server access may be repeated after a certain period of time, or the processing may be continued as if there was a response from the server machine.

[0016]

In the computer system shown in FIG. 2, when the server machine is stopped, for example, when the power of the server machine 1 is turned off, a client program executed on the client machine is executed. There is a problem that it takes a long time to detect the stop of the server machine, and as a result, the execution time of the client program becomes extremely long.

As an example, consider the case where the power supply of the server machine 1 is turned off (intentionally or due to a failure) in FIG. This corresponds to the case where the power supply control device 8 of the server machine 1 in FIG. 2 disconnects the power supply line 11 and the power supply device 7 (because the main body power switch is turned off or the like). I do. At this time, server machine 1
Since no power is supplied to the communication control unit 21 in the NIA 40, the normal-time communication control unit 24 and the transmission / reception unit 22 are not operating.

In the above situation, when the client program 6 of the client machine 3 attempts to access the server by the steps of FIG. 5 described above, the process proceeds as shown in FIG. In FIG. 7, the time elapses from the left to the right in the figure, and the operations of the client machine 3, the server machine 1 (however, the power is turned off), and another server machine 1 '(not shown in FIG. 2). , And communication between the computers.

At the time point 141 in FIG. 7, a packet having the destination address of the NIA 40 of the server machine 1 is transmitted from the client machine 3 at step 101 in FIG. N
It is sent to the IA 40 (142 in FIG. 7). However, since the transmission / reception unit 22 is not operating, the packet is normally discarded, and no response is returned to the client machine 3. Therefore, in the client machine 3, Step 102 in FIG.
7 of FIG. 7 by the notification from the timer set in FIG.
At the time point 43, the timeout occurs, and the client program 6 determines that the server machine 1 has stopped at this time point. The subsequent response varies depending on the client program 6 as described above.
For example, if the processing can be continued by accessing the server 1 ', the packet is transmitted to the server 1' again in step 105 in FIG. 5 (144 in FIG. 7). If power is supplied to the server machine 1 'and it is operating normally, a response is transmitted from the server machine 1' (145 in FIG. 7), and the client program 6 continues the processing after receiving the response. .

Here, as is apparent from FIG. 7, if the server machine 1 is stopped, the client program 6
Has to wait for the time T set in the timer from the start of the server access to the determination that the server machine 1 has stopped. Regardless of whether the client program 6 displays an error message or transmits a new packet to another server as shown in FIG. In general, the time T must be sufficiently longer than the time until a response is returned when the server machine 1 is operating normally, so that when the server machine 1 is stopped, the time of the client program 6 The execution time has to be very long.

Heretofore, as a computer such as the server machine 1, a particularly reliable computer has been prepared, and has been always energized and hardly stopped. Therefore, the probability of the above-mentioned problem occurring is low. However, recently, with the progress of semiconductor technology, personal computers (PCs) placed on the desks of each user have become more sophisticated. As a result, as shown in FIG. , 1 ″,...
Computer systems that are used in such a way as to fulfill the role of the server machine 1 have increased.

In FIG. 4, one of the users of the computer system executes an appropriate client program on the PC 3 on his / her desk to access the servers, and obtains files 81 and 8 in the respective servers.
By obtaining the contents of 1 ', 81 "via the network 2 and browsing them on their own PC 3, each user can proceed with work efficiently while sharing documents and the like with each other. In a computer system such as PC4, for example, when the user of PC1
Is turned off at any time without notifying other users. Then, the client program (and the user) who wants to access the PC1 by using the PC3 has to wait for a long time until it is determined that the PC1 is stopped, as described above. Will take action. This greatly hinders the efficiency of the user's business performance.

An object of the present invention is to solve the above-mentioned problem and to provide a computer system which enables a client program to immediately take the following measures when a server accessing a server is stopped. That is.

[0024]

According to the present invention, a server machine 1 is provided.
The power is always supplied to the communication control unit 21 in the NIA (that is, even when the server machine 1 is stopped) to operate the communication control unit 21. If the server machine 1 is stopped, a packet is transmitted from the network 2. If sent to the NIA, a packet indicating that "the server machine 1 is stopped" (delivery failure notification)
Is immediately sent back to the client machine 3 to achieve the above object.

For this purpose, in the present invention, the NIA control method includes a means for constantly supplying power to the communication control unit 21 and a means for detecting whether or not the server 1 to which it is connected is stopped. The NIA is provided with a means for creating the delivery failure notification having the client machine 3 that has transmitted the packet as a destination.

With the above means, when the server machine 1 is stopped, the NIA immediately transmits a non-delivery notice to the client machine 3 upon receiving a packet sent from the client machine 3, and the client program 6 , After transmitting the packet to the server machine 1 after stopping the server machine 1,
By receiving the delivery failure notification, it is possible to know immediately and take the next response promptly.

[0027]

FIG. 1 shows the configuration of a computer system 10 according to an embodiment of the present invention. Each part of the computer system 10 of FIG. 1 is the NIA 5 of the server machine 1 of FIG.
2 is the same as the respective parts in FIG.

(Configuration of NIA) In FIG.
Are provided with a communication control unit 21, a NIA auxiliary power supply device 25, a main body stop detection unit 26, and a power supply control unit 27. The communication control unit 21 includes a transmission / reception unit 22, a communication control unit 23 when the main body is stopped, a communication control unit 24 during a normal operation,
Is provided.

The main body stop detecting section 26 and the power control section 27
Are connected to the power supply line 11 of the server machine 1. Further, the main body stop detection unit 26 and the power supply control unit 27
5 is also connected to the NIA auxiliary power supply 25. Further, the communication control unit 21 is connected to the power control unit 27 by a power line 35. Further, the selection unit 29 of the communication control unit 21 uses the communication line 28 to
Is connected to

Hereinafter, each part of the NIA 5 of FIG. 1 will be described. The NIA auxiliary power supply device 25 is a power supply device independent of the power supply device 7 and supplies power to each part of the NIA 5 even when the power supply line 11 and the power supply device 7 are cut off by the power supply control device 8. In particular,
The NIA auxiliary power supply 25 is, for example, a storage battery provided in the NIA 5, or a solar battery attached to the housing of the server 1.

The main unit stop detecting unit 26 detects whether the server machine 1 is stopped and transmits it to the selecting unit 29 of the communication control unit 21 via the signal line 28. More specifically, if power is not supplied to the power supply line 11, it is determined that the server machine 1 is stopped, and no signal is sent to the selection unit 29. Conversely, while the power is supplied to the power supply line 11, the signal is continuously sent to the selection unit 29. The main body stop detection unit 26 is constantly supplied with power from the NIA auxiliary power supply device 25.

The power control unit 27 always supplies power to the communication control unit 21. More specifically, the power control unit 27
When the power is supplied to the power supply line 11, the power supply from the power supply line 11 is used. Otherwise, the power from the NIA auxiliary power supply device 25 is used. Supply power to

The communication control unit 21 is a device provided in the NIA 5. As described above, the communication control unit 21 includes the transmission / reception unit 22, the communication control unit 23 when the main body is stopped, the communication control unit 24 during the normal operation, and the selection unit 29. Here, the transmission / reception unit 22 and the selection unit 29 are devices, but the communication control unit 23 at the time of stopping the main body and the communication control unit 24 at the normal time are each devices or provided in the communication control unit 21. They may be separate programs running on a common processor and main memory.

The transmission / reception unit 22 is the same as the transmission / reception unit 22 of the NIA 40 in FIG. 2, is connected to the link 4, and exchanges signals with the network 2.

The normal-time communication control unit 24 is the NIA 4 of FIG.
0, which is the same as the normal-time communication control unit 24,
Commanding the transmission of the packet, or commanding the transfer of the packet received from the network 2 to the main memory,
The operation of the transmission / reception unit 22 is controlled so that packets can be transmitted / received correctly. The normal-time communication control unit 24 operates only when the main unit is operating normally by the selecting unit 29 described later.

The communication control unit 23 at the time of stopping the main body controls the transmission / reception unit 22 in the same manner as the communication control unit 24 at the time of normal operation. Instruct the transmission / reception unit 22 to send it back to 2. How to create a non-delivery notice is described in the next section. The main body stop-time communication control section 23 operates only when the main body is stopped by the selecting section 29 described later.

The selection unit 29 enables only one of the communication control unit 23 at the time of main body stop and the communication control unit 24 at the normal time to control the transmission / reception unit 22 according to the presence or absence of a signal from the main body stop detection unit. . More specifically, when a signal is being sent to the signal line 28, the selection unit 29 connects the transmission / reception unit 22 to the normal communication control unit 24. Otherwise, the selection unit 29 connects to the main unit stop communication control unit 23. I do. Communication control unit 23 when the main body is stopped
When the communication control unit 24 is a device, the selection unit connects the signal line from the transmission / reception unit to one of the two. Also,
If both are programs, the processor executing both programs is allowed to refer to the state of the signal line 28, and the processor can call one of the two programs accordingly.

FIG. 9 shows a packet format according to the present embodiment. All the packets transmitted to the network 2 in FIG. 1 are in the form of the general form 17 shown in FIG. Here, the packet 17 has the protocol type 180 and the transmission source NIA.
, The transmission destination address 182 storing the address of the transmission destination NIA, and the data 183 storing communication contents.

FIG. 9B shows the format of a packet (hereinafter referred to as a "request packet") sent from the client machine 3 to the server machine 1 in FIG. The request packet conforms to the format of the packet 17, and a value corresponding to “normal communication” is stored in a portion corresponding to the protocol type 180 (184). The portion corresponding to the source address 181 stores the address 185 of the client machine, and the portion corresponding to the destination address 182 stores the address 186 of the server machine.

The data 187 of the request packet 18 (corresponding to the data 183 of the packet 17) includes an address 188 of the client program and a request content 189. The address 188 of the client program 6 is an identification number for identifying the client program 6 that has transmitted the request packet when a plurality of programs are executed on the client machine. Also, the request content 189 includes the content that the client program 6 requests from the server machine 1, and actually refers to a program that is being executed on the server machine 1 (if it is operating normally). The program executes a process according to the request of the client program.

In this embodiment, the format of a delivery failure notification created in the NIA 5 and returned to the network 2 when the server 1 receiving the request packet 18 is stopped is shown in FIG. The non-delivery notice 19 is also a packet 17
Format. First, the protocol type 180
Is stored in a portion corresponding to. This value is different from the value meaning the “normal communication”. Further, in the portions corresponding to the source address 181 and the destination address 182, an address 186 of the server and an address 185 of the client are stored, respectively. For these addresses, the addresses 185 and 186 included in the request packet 18 received by the server machine 1 may be copied. Further, data 194 (packet 1
7 corresponds to the data 183), a value meaning "packet delivery impossible" is stored as the type of the error message. Further, a part of the data 187 of the received request packet 18 is copied and stored in the data 194 (196). A portion 196 of the request packet includes at least the address 188 of the client program.

(Normal Operation) NI Having the Above Configuration
By using A5, the computer system of the present invention is realized. First, the operation of each unit when the server machine 1 is operating normally in FIG. 1 will be described.

Normally, the power supply controller 8 of the server machine 1
, The power supply device 7 and the power supply line 11 are connected, and power is supplied to the power supply line 11. Therefore, the power supply control unit 27 supplies power to the communication control unit 21 using the power supply of the power supply line 11. Here, if the NIA auxiliary power supply device is a rechargeable battery, the power supply control unit 27 can supply power to the NIA auxiliary power supply device 25 via the power supply line 35 to charge the storage battery.

Further, the main body stop detecting unit 26 determines that the server machine 1 is operating normally, and continues to send a signal to the signal line 28. The selection unit 29 detects the signal and connects the transmission / reception unit 22 to the normal communication control unit 24. Therefore, the control of the transmission / reception unit 22 is performed only by the normal communication control unit 24.

From the above, it is clear that the operation of the server 1 when power is supplied to the power supply line 11 is the same as the operation when the server 1 in FIG. 2 is operating normally. 1 operates in the same manner as the computer system 10 of FIG.

(Operation when Server Machine Stops) On the other hand, when the server machine 1 is stopped, the power supply device 7 and the power supply line 11 are cut off by the power supply controller 8 of the server machine 1, and the power supply line 11 Has no power. Therefore, the power supply control unit 27 supplies power to the communication control unit 21 using the power supply of the NIA auxiliary power supply device 25.

The main body stop detecting section 26 determines that the server machine 1 is stopped, and does not send a signal to the signal line 28.
The selection unit 29 detects that no signal is coming, and connects the transmission / reception unit 22 to the communication control unit 23 at the time of stopping of the main body.

The communication control section 21 of the NIA 5 connected as described above operates as follows, following step 120 in FIG. First, the communication control unit 21 waits until a packet is transmitted from the network 2 (12).
1). When the packet is received, the address of the client machine is extracted from the packet (122), and a non-delivery notice with the address as the destination is created by the method described above (123). Then, a delivery failure notification is transmitted to the network 2 (124), and the flow returns to step 121 to wait for the next packet.

When the server 1 is stopped by the above operation of the NIA 5, the client program 6
However, it is possible to immediately know that the server machine 1 has stopped by receiving a non-delivery notice after transmitting a packet to the server machine 1, and it is clear that the following measures can be taken promptly.

More specifically, the same processing as that of the client machine 3 in FIG.
, The time shown in FIG. 8 is traced. As in FIG. 7, the client program running on the client machine 3 starts server access at the time of 141 in FIG. 8, and transmits a packet to the server machine 1 (14
2). Then, the NIA 5 of the server machine 1 immediately returns a non-delivery notice to the client machine 3 by the method described above (160). Here, from the method of creating the delivery failure notification described above, it is clear that the destination address of the delivery failure notification is the address of the client machine 3, so that the delivery failure notification is sent to the client machine 3 via the network 2. is there. Further, since the delivery failure notification includes the address of the client program, even when a plurality of client programs are executed on the client machine 3, when the delivery failure notification is received, the fact that the delivery is impossible is not performed (the client machine). 3 to the client program 6 (in the form of a software interrupt or error notification) by the operating system running on it. Therefore, the client program 6 can confirm that the server machine 1 has stopped in the shortest time after starting the server access, and promptly take the next response (for example, to another server 1 'in FIG. 8). Send another request
(144), a response is obtained, and (145) the processing is continued).

(Modification) The NIA 5 in this embodiment is
The modification as shown in FIG. 10 is possible. These modifications can be made together or individually. Here, the embodiments described so far are referred to as “before deformation” for convenience,
The state after the deformation is referred to as “after the deformation”.

The first modification relates to a method for detecting a stop of the server machine 1. In the NIA 5 before the deformation, the main body stop detection unit 26 detects whether or not power is supplied to the power supply line 11 and determines whether or not the main body is stopped. 26, a system bus 201 provided for the server machine 1 (for mutual communication between the components 9, 9 ',... And the NIA 5) is connected by a signal line 202, and a signal on the system bus is connected. If an error occurs, it is determined that the server is stopped. Others are the same as before the deformation.

As a criterion for judging the abnormality of the signal of the system bus, for example, it is conceivable that the signal for stopping the processor has continued for a certain period of time or that only the same place in the program has been executed for a certain period of time. By performing this modification, it is possible to detect not only the shutdown of the power supply but also a broader stop state of the server machine.

A second modification is the NIA auxiliary power supply 25
Related to Before the deformation, the NIA5 has an independent power supply (N
Although the IA auxiliary power supply 25) was provided, if there is a function of supplying power to the network 2, the NIA auxiliary power supply 25 is removed and the power line 2
00 may be subtracted and connected to the power supply control unit 27. Others are the same as before the deformation. According to this modification, it is not necessary to provide a power supply device in the NIA 5, and an effect such as cost reduction by reducing parts is expected. In particular, this modification includes a link 4 connecting the network 2 and the transmission / reception unit 22 and a power line 200.
However, the effect is great for the network 2 that is bundled into one cable from the beginning.

[0055]

According to the NIA and the computer system using the same according to the present invention, when a server to access a server is stopped, the client program transmits the packet to the server to stop the server. After the transmission to the server machine, it is possible to know immediately by receiving the delivery failure notification, and it is possible to immediately proceed to the next response without having to wait long until the timeout.

Furthermore, according to the present invention, when the server is stopped, power is supplied only to the NIA. Therefore, compared to a case where the entire server is constantly energized, the server and, moreover, the entire computer system are powered on. Low power consumption.

[Brief description of the drawings]

FIG. 1 is a block diagram of a computer system according to an embodiment of the present invention.

FIG. 2 is a block diagram of a computer system according to the related art.

FIG. 3 is a block diagram showing a method of using a general computer system.

FIG. 4 is an explanatory diagram showing an example of a usage form of a general computer system.

FIG. 5 is a flowchart showing the operation of the conventional technique after the client program starts accessing the server.

FIG. 6 is a flowchart showing the operation of the NIA when the server is stopped according to one embodiment of the present invention.

FIG. 7 is an explanatory diagram showing a lapse of time of an operation in each computer according to the related art.

FIG. 8 is an explanatory diagram showing the time lapse of the operation of each computer in one embodiment of the present invention.

FIG. 9 is an explanatory diagram showing a data format in one embodiment of the present invention.

FIG. 10 is a block diagram showing an NIA according to another embodiment of the present invention.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Server machine, 2 ... Network, 3 ... Client machine, 5 ... Network interface (NIA), 1
DESCRIPTION OF SYMBOLS 1 ... Power supply line, 17 ... Packet (general type), 18 ... Request packet, 19 ... Notice of non-delivery, 21 ... Communication control unit,
22: transmission / reception unit, 23: communication control unit when the main body is stopped, 24
Normal communication control unit, 25 ... NIA auxiliary power unit, 26 ...
Body stop detector.

Continued on the front page (72) Inventor Kyoka Mina 1-280 Higashi Koigakubo, Kokubunji-shi, Tokyo Inside Central Research Laboratory, Hitachi, Ltd.

Claims (6)

[Claims] 1. A network interface (hereinafter referred to as NIA) for connecting a computer to a network, means for constantly supplying power to the NIA regardless of whether or not the computer is stopped. Means for detecting whether or not the computer connected to the NIA is stopped, and, when the NIA receives a packet via the network, a notification of non-delivery to the computer which has transmitted the packet as a destination. An NIA having means for creating. 2. The NIA according to claim 1, wherein said means for constantly supplying power to said NIA irrespective of whether or not said computer is stopped comprises a power supply from said network to which said NIA is connected. NI characterized by being
A. 3. The NIA according to claim 1, wherein power is supplied to the computer as means for detecting whether or not the computer to which the NIA is connected is stopped. NIA characterized in that it has means for detecting whether or not it is in use. 4. The NIA according to claim 1, wherein said means for detecting whether said computer to which said NIA is connected is stopped is provided on a system bus of said computer. An NIA having means for detecting a signal abnormality. 5. A computer system in which one or more computers are connected to a network, wherein the program executed on the first computer, which is one of the one or more computers, is the first computer. When transmitting the packet to a second computer different from the computer, if the second computer is stopped, the program is notified to the program that the second computer is stopped. A computer system characterized by being performed. 6. The computer system according to claim 5, wherein the NIA installed in at least one of said second computers is an NIA according to any one of claims 1 to 4.
A computer system characterized in that a computer is used.