JPS62217749A - Local area network equipment - Google Patents

Abstract

PURPOSE:To attain efficient operation by providing a means managing a transfer request remaining time and a means changing optionally the transfer priority or the like so as to realize the transfer request from other nodes while keeping the punctuality of an object node. CONSTITUTION:In exchanging information between plural nodes while managing centralizingly the transfer request information among the plural nodes, the 1st means 10 comprising a system timer 53 and a timer section 54 or the like and the 2nd means 11 comprising a CPU 51 and a main memory 56 or the like are provided. The means 10 manages the remaining time of the transfer request having the punctuality designation based on transfer request information including the transfer priority from each node and the punctuality designation as required. Then the means 11 changes optionally the transfer priority to an object node based on the result of management of the means 10. At the time of a data transfer request having the designation of arrival time but no urgency is given, it is possible to given the chance of transfer to other transfer requests while keeping the arrival time of the object node so as to attain the efficient operation.

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Industrial Application Field) The present invention relates to a local area network device, and particularly to a local area network device with an improved queue management method for each node.

(Prior Art) In response to the recent development of the information society, local area network devices suitable for transmitting large amounts of data at once have come into widespread use in various fields. For example, FIG. 5 shows a conventional example in which such a device is applied to mutual transmission of medical information in a hospital.

The apparatus shown in the figure includes nodes N1 to NrI, which are composed of a medical image acquisition device 1, a diagnostic terminal, a medical image database computer, a computer, etc., and each of the nodes N1 to Nn.
A low-speed network bus 2 for propagating and communicating control signals during data transmission between nodes, and a star coupler 4 for transmitting various data between each node between nodes N1 to Nn. The host computer NO centrally manages transmission requests from each node N1 to Nn. When the node is in a receivable state, transfer permission is given to a transfer request from a specific node according to its priority.

By the way, once permission is granted to a transfer request including image data of a large seedling, for example, a situation arises in which transfer requests from other nodes are forced to wait for a long time. Among such transfer requests, there are a considerable number of transfer requests that simply need to be completed by a specified time. In this case, if the priority of the transfer order for the transfer request is lowered, a situation will arise where some image data cannot be transferred until the specified time when the network bus is congested.

For this reason, the actual situation is that even if it is unavoidable to delay permission to other transfer requests, the specified time from a specific node must be respected.

(Problems to be Solved by the Invention) As described above, in conventional devices, management of transfer request priority and six-hour period is insufficient, resulting in long waiting times for transfer requests and failure to reach the destination node by the specified time. There is a problem that the data may not reach the destination.

Therefore, the present invention provides a local area that can reliably meet the arrival time to the destination node, provide more transfer opportunities for transfer requests from other nodes, and realize inefficient operation. The purpose is to provide network equipment.

[Structure of the Invention] (Means for Solving the Problems) The present invention provides a local area network device that centrally manages mutual transfer request information between a plurality of nodes and exchanges information between them. A first means for managing the remaining time of a transfer request with a specified arrival time based on transfer request information including a transfer priority order from each node and a specified arrival time as necessary, and a management result of the first means. The present invention is configured to include a second means that can arbitrarily change the transfer priority order for the base destination node.

(effect) The operation of the apparatus having the above configuration will be explained next.

Transfer request information from a plurality of nodes is first stored in the second means. Of these pieces of information, the required arrival time designation information is time-managed by the first means, and the remaining time from the reception time of the transfer request to the required arrival time at the destination node is counted. Then, the results of time management regarding each transfer request in the first means are sent to the second means, whereby the second means changes the priority order of the transfer request information in accordance with the specified arrival time, It is sent to the destination node so as to meet the specified arrival time.

This makes it possible to efficiently process transfer requests whose deadlines are close to each other and transfer requests with plenty of time, providing transfer opportunities for other transfer requests while meeting the deadline. It becomes possible.

(Examples) Hereinafter, the present invention will be specifically explained using examples. In this embodiment as well, it is assumed that the network between the nodes N1 and N1 is the same as that of the conventional device shown in FIG.

FIG. 1 shows a first means 10 and a second means 11 in an apparatus according to an embodiment of the present invention.

The second means 11 includes an interface unit 52 connected to the low-speed network bus 2, a CPU 51 that transmits and receives control information via this interface unit 52, various tables described later, and a preset limit value αH regarding transfer time. The main memory 56 has a built-in main memory 56.

CPtJ51 connects each node N1 to
Manage transfer requests from NN by priority class,
For transfer requests whose destination node is ready to receive, @
Transfer permission is granted in order of priority, and in the case of requests with the same priority, on a first-come, first-served basis.

FIG. 2 shows an example of a queue management method for transfer requests from each node N1 to NN, in which one of a plurality of priority classes is managed. V, that is,
The CPU 51 rearranges the contents of the transfer requests and stores them in a transfer request table 31 as shown in FIG. 2 provided in the main memory 56.

This transfer request table 31 includes an area 31a for storing the code of the originating node, an area 31b for storing the code for the destination node, and ID (certification) information.

It has an area 31G for storing image information, and also stores the numbers of counters 54a to 54R (these counter numbers are R1 to RR), which will be described later, and which are designated for managing transfer requests specifying arrival times. It has an area 31d. Incidentally, if any of these counters 54a to 54R is unused, a code indicating that counter is inserted into the area 31d.

Free area of the transfer request table 31 (table number T
1 to T[> are stack-managed. Specifically, the free areas of the transfer request table 31 are linked by pointers, and the number of the next free area is set to 1 as a pointer as shown in FIG.
Wait pointer table 3 provided in fff main memory 56
2.

Then, the number of the empty area at the beginning is stored in the empty head pointer column 33. This wait pointer table 3
2 is also used for queue management of transfer requests. That is, the first column of the waiting pointer table 32 forms a waiting head pointer column 33, in which a pointer indicating the order in which transfer requests are received or received is stored. Since this pointer corresponds to the table number of the transfer request table 31,
By sequentially following the pointer to the first column of the waiting pointer table 32, the transfer request contents can be obtained on a first-come, first-served basis.

Note that the example shown in FIG. 2 shows a configuration in which the second column is a pointer column for tracing transfer requests in reverse order.

That is, by sequentially following the pointer to the second column of this waiting pointer table 32, it is possible to know the transfer contents in the reverse order from the first-come-first-served basis.

The first means 10 includes a system timer 53 that provides the current time to the CPU 51, a timer part 54 for time management of transfer requests, and a clock generator 55 that supplies a clock to the timer part 54. It is composed of The timer portion 54 includes counters 54a to 54R.

The remaining time until the time when the transfer request must arrive is managed by the timer part 54 which includes the counters 54a to 54R, and also by the timer section 54 which includes the counters 54a to 54R.
The unused portion is managed using the same method as the method for managing the free area of the transfer request table 31 described above.

That is, unused counters are connected by pointers, and the pointers are stored in the free counter table 41 provided in the main memory 56, as shown in FIG. For example, the leading counter number is stored in the empty leading counter column 42 of the empty counter table 41.

Further, information regarding transfer requests managed by the counters 54a to 51 is stored in a counter information table 43 provided in the main memory 56, as shown in FIG.

This counter information table 43 has an area for storing table numbers @T1 to T[ of the transfer request table 31 in which the priority class and transfer request contents are occupied.

Next, an example of a packet for exchanging information between the nodes N1 to Nn will be described with reference to FIG. 4.

This packet has a packet mark (P), a receiving node (N
odel), the originating node (Node2) and the text (TE
XT).

The structure of this text varies depending on the type of request. Fourth
The figure shows an example of an image data transfer request.
In this case, the text is a code indicating a transfer request (Htrc
+), forwarding destination node (Node3), '/M precedence (
C) (CI, C2, . . . etc.), arrival time one hour (Th), arrival time - minutes (1'-m), ID information.

It consists of image information (image size, bit length of one pixel, number of two images), etc.

In addition, if you do not specify the required arrival time in the transfer request, the code indicating "No specification" will be changed to the required arrival time of 1 hour (Th>
You are entering the area.

Next, the operation of the apparatus having the above configuration will be explained.

When the CPU 51 receives a transfer request (Sl) specifying the required arrival time from each node N1 to NN, the CPU 51 first calculates the time difference ΔM between the current time (S2) obtained from the system timer 53 and the specified required arrival time in minutes. Correct and calculate. Note that, since all transfers end at night, the designation of the required arrival time is limited to that for the same day.

On the other hand, the maximum waiting time for a transfer request given the highest priority allowed for normal image data transfer can be predicted to some extent once the network configuration and usage conditions are determined. Therefore,
As the limit value (threshold value) αH for determining whether there is time remaining for a transfer request with a specified arrival time, a value obtained by adding a margin time for safety to the predicted value of the maximum waiting time is selected, and this value is stored in advance in the main memory. Store it in 56.

Then, the CPU 51 calculates a limit value αH from the time difference ΔM.
The initial value RH of the counters 54a to 54R is calculated by subtracting . Next, the CPU 51 secures an empty area (table number @) T1 in the transfer request table 31 based on the priority class C1 specified by the packet, organizes and writes the contents of the transfer request. Further, a counter number R1 for managing the remaining time is secured, and this counter number R1 is stored in the counter number area 31d of the transfer request table 31. Furthermore, the priority class C1 and the table number T1 of the transfer request table 31 are written in the counter information table 43.

Finally, the calculated initial value RH (
S3) is sent to initialize the transfer request and complete the registration of the transfer request.

The counter with the initialized counter number R1 immediately starts operating, and the clock generator 55 counts the clock (S4) generated every minute.

When image data transfer requests to the network bus 2 become crowded, transfer requests occur such that the remaining time until the arrival time reaches the limit value αH.

In this case, it is assumed that the timer part 54 including the counter with the counter number R2 finishes counting the specified number of pulses (S4), generates an interrupt signal, and stops its operation. This interrupt signal and counter number R2 (S5) are C
It is input to PU51. The CPU 51 returns the used counter number R2 to the free pointer column 42 of the free counter table 41.

Next, the CPU 51 reads the transfer request priority C2 from the counter information table 43 and the table lT2 of the transfer request table 31 in which the transfer request contents are written, and
This transfer request is removed from the queue of priority class C3 and sent to the low-speed network 2 via the interface section 52, and is re-registered in the queue of priority class C3 instead.

That is, an empty area (table number) is secured in the transfer request table 31 of priority C3, and the request contents are written anew.

On the other hand, if there is a request to transfer image data to the network bus 2, transfer permission is granted to the transfer request for which the remaining time is being managed. The operation of the CPU 51 in this case will be explained below.

Refer to the counter number area 31d of the transfer request table 31 to determine whether the transfer request for which transfer is to be permitted specifies the arrival time, that is, whether the remaining time is being managed by the timer section 54. You can tell by doing this. When granting permission to a transfer request whose remaining time is being managed by the timer section 54 that includes counters 54a to 54R,
It is sufficient to stop the operation of the counter with counter number R4, return this counter number R4 to the free pointer column 42 of the free counter table 41, and then grant permission to the transfer request as in the case described above.

It goes without saying that the present invention is not limited to the embodiments described above, and that various modifications can be made within the scope of the invention.

[Effects of the Invention] As explained above, according to the present invention, when there is a request to transfer data that has a specified arrival time but does not need to be transferred in a hurry, other transfers can be performed at the same time while meeting the required arrival time. It is possible to provide a local area network device that can provide as many transfer opportunities as possible to requests and can achieve efficient operation.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of an embodiment device of the present invention;
FIG. 2 is an explanatory diagram showing the contents of the transfer request table and waiting pointer table in the second means of the same device, and FIG. 3 is an explanatory diagram showing the contents of the counter information table and free counter table in the second means of the same device. FIG. 4 is an explanatory diagram showing an example of a packet used for data transmission in the device, and FIG. 5 is an explanatory diagram without showing the network of the local area network device. 10...first means, 11...second means, 51.
...CPU, 54...part of timer, N1~NN...
node. Agent Patent Attorney Nori Chika Megumi Kanshu
Ogo Qufu M Figure 5

Claims (2)

[Claims] (1) In a local area network device that centrally manages mutual transfer request information between multiple nodes and exchanges information between them, the transfer priority from each node and the required arrival time are specified. a first means for managing the remaining time of a transfer request with a must-arrival time specification based on the included transfer request information; and a transfer priority order for the destination node can be arbitrarily changed based on the management result of the first means. A local area network device comprising: second means. (2) The second means compares the remaining time of the transfer request with a must-arrival time specified managed by the first means with a preset limit value, and if the remaining time is less than the limit value, The local area network device according to claim 1, wherein the priority of this transfer request is forcibly raised.