JPH0387956A - Computer network device - Google Patents

Abstract

PURPOSE:To disperse the load of work by using all resources included in a computer network including a computer in common between the nodes. CONSTITUTION:When a node A wants to use a file 13B of a node B not a file 13A of its own node, a designated command is inputted to a node controller 15A. The controller 15A is connected to a node controller 15B of the node B via a network cable 16, and the file 13B of the node B is assigned as a file device of the node A. Thus the node A can use the file device of the node B. As a result, the resources can be effectively used between the nodes.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to a computer network system in which a plurality of computers are connected via a transmission path, and is particularly a computer network device suitable for a system using personal computers (personal computers). Regarding.

[Conventional technology]

When using multiple computers, in order to improve their utilization efficiency, it has traditionally been possible to create a computer network system by connecting the computers through an appropriate transmission path and providing the computers as network nodes. A typical example is the system known as "MS-Networks" (trademark name of Microsoft Corporation, USA).

By the way, in such a conventional computer network system, the CPU of the node connected to the network
(Arithmetic processing unit) only performs local processing in that node, and the capability of this CPU itself cannot be used from other nodes.

For example, in the above-mentioned "MS-Networks" system, PCs are assigned as dedicated nodes called "servers" within the network (transmission system), and the resources (components) possessed by these servers are called shared resources. ,
This server's shared resources are made available to other nodes called workstations via the network, but even then, only the server's files, printers, and other resources can be used. The CPU capacity of this server itself cannot be used. In other words, it was not possible for the workstation to cause the server's CPU to perform compiling or the like.

[Problem to be solved by the invention]

The above-mentioned conventional technology does not give particular consideration to the mutual use of the capabilities of the CPUs existing in each node on the network, and even though they exist as multiple resources on the network, The problem was that they were not able to make full use of their abilities.

The purpose of the present invention is to
The U itself, as well as other resources such as keyboards, files, CRTs (image display devices), and printers,
It is an object of the present invention to provide a computer network device which can be used in common as a resource of the entire network and can fully utilize the capabilities of a plurality of CPUs.

[Means to solve the problem]

The above objective is achieved by providing a node controller whose function is to transmit data between each resource via the transmission path between each resource of the workstation that constitutes each node on the network and the transmission path. .

[Effect]

The node controller serves to form a single node consisting of any combination of all resources connected to this node controller, and
Since the functions of the PU can also be used as desired, efficient computer operation is possible.

〔Example〕

DESCRIPTION OF THE PREFERRED EMBODIMENTS A computer network device according to the present invention will be explained in detail below using illustrated embodiments.

FIG. 1 shows an embodiment of the present invention in which a plurality of nodes are electrically connected by network cables 16 to form a computer network device. Note that in this figure, only two nodes are shown as node A and node B, respectively.

As is clear from this figure, in this embodiment, these nodes A and B each have a node controller 15A.
, 15B are provided.

These node controllers 15A and 15B have keyboards 11A and 11B, CRT 12A, and
12B, files 13A, 13B, and CPU 14A
, 14B and the network cable 16, these node controllers 15A, 15B function as follows.

Now, for example, when the operator of node A wants to use the file 13B of node B instead of the file 13A of his own node, he first inputs a predetermined command to the node controller 15A. Then, the node controller 15A is connected to the node controller 15B of node B via the network cable 16, and the file 13B of node B is allocated as the file device of node A.

Therefore, as a result, node A can use the file device of node B.

The same thing happens when node A attempts to use the CPU 14B of node B. At this time, the operator of node A also
A command is sent from the keyboard 11A, which is the input device of the own node, to the node controller 15A.
B via a network cable 16 so that input from the keyboard 11A is input to the CPU 14B.

Therefore, at this time, node A uses CPU1 of node B.
I was able to perform the necessary processing using 4B, and in the end,
According to this embodiment, resources can be used effectively between nodes.

Next, regarding this example, according to FIGS. 2 to 7,
This will be explained in more detail.

First, FIG. 2 shows a data processing state using only the local resources of a node, for example, node A. At this time, the data flow inside the node controller 15A is in the state shown by the broken line. keyboard 11
The data input from A is taken into the CPU 14A,
The processing results are displayed on the CRT 12A. Also,
At this time, if necessary, the CPU 14A will open the file 13A.
access and update data. The flow of data and control at this time is the node controller 15.
Managed by A.

Furthermore, at this time, no data is input/output to/from the network cable 16.

Next, FIG. 3 shows a state in which resources are mutually used between f node A and node B among multiple nodes existing on the network. This shows a case where the CPU 14B of node B is used, and therefore, in this case, the data flow within the node controllers 15A and 15B of each node is as shown by broken lines.

Incidentally, in order to switch from the state shown in FIG. 2 to the state shown in FIG. 3, in this embodiment, the command shown in FIG. 4 is input from the keyboard IIA of node A. That is, when the operator of node A wants to use the CPU 14B of node B, first, from the keyboard 11A, as shown in FIG.
Enter rUsE ¥¥B¥CPUJ.

By manual operation of this command, node controller 1
5A disconnects the keyboard IIA and the CPU 14A, connects it to the node controller 15B on the node B side via the network cable 16, and then connects it to the node controller 15B on the node B side.
Allocate PU14B as the CPU on node A.

Therefore, from now on, the keys on the keyboard 11A will be connected to the CP via the node controller 15A, the network cable 16, and the node controller 15B of the node B.
It is taken into U14B.

Furthermore, as a result, the data that the CPU 14B attempts to display is now transmitted from the node controller 15B to the node A via the network cable 16, and then transmitted by the node controller 15A to the CRT 12A for display. , This allows the node A side to fully utilize the capability of the CPU 14B of the node B.

Next, when node A tries to access file 13B of node B, as shown in Figure 4,
This time, enter "rUSE ¥¥B¥file".

As a result, as shown in the figure, the file accessed by the CPU 14B is switched from the file 13A of node A to the file 13B of node B, and processing can be performed using this file 13B.

On the other hand, to cancel the connection state as described above, as shown in FIG.
All you have to do is input the command "J or rFREE\file", and by doing so, you can arbitrarily cancel a given resource.

Next, an embodiment of the node controller will be described with reference to FIGS. 5, 6, and 7.

As described above, in the embodiment of the present invention, there are a plurality of nodes on the computer network. Taking node A among them as an example, in this embodiment, each node is arranged as shown in FIG. It is roughly divided into three parts. That is, first 17A is a device redirect controller that manages where each resource is allocated, then 18A is a connection list that actually changes the connection status of resources, and 19A is a network controller that interacts with the network. be.

The device redirect controller 17A issues a predetermined command to the connection list 18A in order to set or change the allocation of each resource to a node in response to a command input from the keyboard 11A. At this time, when communication within the network is required, a predetermined command is also issued to the network controller 19A to request connection with the other node.

As a result, first, for example, in a local operating state in which all resources within the own node are used, the connection state according to the connection list 18A is set as shown in FIG. 5, and the key input from the keyboard IIA is Inputs are input to the CPU 13A of the own node from the contacts a2 and al of the connection list A18 via the contacts b1 and b2, and on the other hand, the display output from this CPU 13A is input to the contacts c2. The signal is supplied from cl to the CRT 12A of the own node via contacts di and d2.

In addition, the file 14A and the CPU 13A are also connected by a connection path via the contacts e2, el, fl, and f2 of the connection list 18, and in the end, all processing at this node A is performed using only local resources. In this case, no data is exchanged via the network cable 16.

Next, if the device redirect controller 17A determines that the data input by the operator of node A from the keyboard 11A requests a change to the current connection list 18A, then the data is changed to the connection list 18A. A command is issued. For example, if the operator inputs rUSE ¥¥B1ICPUJ, the device redirect controller 17A determines that connection to the CPU of node B is requested. Furthermore, if rUSE\\B\CRTJ is input, it is determined that the output data to the CRT of node B is requested to be displayed on the CRT 12A of the own node.

Therefore, the connection state according to the connection list 18A at this time is as shown in FIG. That is, at this time, according to the command from the device redirect controller 17A, the connection list 18A connects the output of the keyboard IIA to the contact a2.
, a3 to be input to the network controller 19A and sent out onto the network cable 16, and also set to allocate the input from the network controller 19A to the CRTI 2A via the contacts d3 and d2.

Furthermore, the device redirect controller 17A
A predetermined connection request is sent onto the network cable 16 via the network controller 19A. In this case, since it is a connection request to node B, this request is sent to the node controller 1 of node B shown in FIG.
Detected by network controller 19B in 5B,
The device redirect controller 17B is notified that there is a connection request from another node.

Therefore, this device redirect controller 17B
In response to the connection request from node A, if it is acceptable, it changes the settings of the connection list 18B as shown in FIG. That is, this node B
The CRT output of the CPU 13B is connected to the network controller 19B through the path of contacts C2 and c3, so that it can be output onto the network cable 16.

In node A, the connection list 18A sets the connection path to the CRT 12A, as explained in FIG. That will happen.

Also, at this time, in the connection list 18B on the node B side, the key power of the CPU 13B is extended to the node A via the path of contacts b2 and b3, so in the end, data input using the keyboard IIA on the node A is input on the node B. CPU 13
It will be taken into B.

Therefore, at this time, the operator of node A can control the CP in node B by operating the keyboard IIA there.
The CPU 13B can be made to perform processing, and the output of the CPU 13B at this time can also be grasped from the display on the CRT 12A on the node A side.In the end, according to this embodiment, each node in the computer network All resources, including the CPU, can be fully utilized from any node, and the capabilities of multiple CPUs can be fully utilized.

〔Effect of the invention〕

According to the present invention, all the resources existing in a computer network, including the CPU, can be used in common by each node, making it possible for multiple nodes to work efficiently and distributing the work load through the network. It is possible to easily provide multi-user and multi-job support.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an embodiment of a computer network device according to the present invention, FIGS. 2 and 3 are block diagrams for explaining operation, and FIG. 4 is an explanation of commands used in an embodiment of the present invention. , FIG. 5, FIG. 6, and FIG. 7 are respectively explanatory diagrams of the operation of one embodiment of the node controller. 11A, IIB...Keyboard, 12A, 12
B...CRT, 13A, 13B...File, 14A, 14B...CPU, 15A,
15B...Node controller, 16...
...Network cable.

Claims (1)

[Scope of Claims] 1. In a computer network device consisting of at least two computer systems connected to each other via a transmission path, there is a connection between each component in each of the computer systems and between these components and the transmission path. A node controller for controlling the connection state of the network is provided between each of the computer systems and the transmission line, and each component in the computer system is configured to behave equally as a component in the network without having a master-slave relationship. A computer network device characterized by: 2. The device according to claim 1, wherein the components within the computer system are an arithmetic processing unit, an input device such as a keyboard, an output device such as a printer, and an input/output device for files. 3. In the invention of claim 1, the node controller includes a device redirect controller that manages allocation of each of the components, a connection list that changes the connection state of each of the components, and information via the transmission path. A computer network device comprising a network controller that performs sending and receiving.