JPH05316150A - Packet copying machine - Google Patents

Abstract

PURPOSE:To maintain high speed communication of a general packet by discriminating whether a received packet is a multiple address communication packet or a general packet, implementing copying when the packet is the multiple address communication packet, stopping the copying when the packet is the general packet and sending the packet with priority so as to prevent increase in the hardware. CONSTITUTION:A reception packet inputted from an incoming line 21 (211-21M) is fed to a packet discrimination section 23 (231-23M) via a first-in first-out memory section 22 (221-22M), in which whether the packet is a multiple address communication packet or a general packet is discriminated. When the result of discrimination indicates the general packet, the received packet is fed to a selector 27 (271-27M) as an input A by a switch 25 (251-25M). When the result of discrimination indicates a multiple address communication packet, the packet is fed to a packet copy section 29 (291-29M), in which the packet is copied by number of paths specified by its identification number. When the general packet is inputted, the copying of the multiple address communication packet is stopped and the transmission of the general packet is sent with priority.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a packet copy device for copying a received packet for broadcast communication in a packet switch.

[0002]

2. Description of the Related Art Generally, in electronic mail or facsimile communication, broadcast communication is performed in which the same information is transmitted to a plurality of parties.

As a method of realizing this broadcast communication, there is a method of providing a transmission terminal with a sequential broadcast function for sequentially transmitting information to each destination.

However, this method has a problem that the operating time of the transmitting terminal becomes long and the load becomes heavy.

In order to solve this problem, conventionally, a broadcast communication system has been considered in which information output from a transmission terminal is automatically transmitted to each destination in a network.

If this broadcast communication system is provided, the transmitting terminal only has to output the information once, so that the operating time and the burden of the transmitting terminal can be shortened.

In order to realize this broadcast communication system, a copying apparatus is required for copying the received information by the number of routes defined by the destination.

As such a copying apparatus, Japanese Patent Application No.
There are three devices as described in JP-A-3-230339 (JP-A-2-76436).

2 to 4 are block diagrams showing the configurations of these three devices, respectively. Each of the illustrated devices is configured to copy a packet, which is a packet switching type information transmission block.

Here, the apparatus shown in FIG. 2 has a plurality of incoming lines 1
The received packets input from 1 _{1 to} 11 _M (M is an integer of 2 or more) are time-division multiplexed by the time-division multiplexing unit 12, the speed is converted by the speed conversion memory 15, and the converted output is filtered by the filtering unit 13. By filtering
The lines are appropriately output to the outgoing lines 14 _{1 to} 14 _N (N is an integer of 2 or more).

With such a configuration, it is possible to exchange packets at the same time as copying packets.

Similarly to the apparatus shown in FIG. 2, the apparatus shown in FIG. 3 time-division-multiplexes the received packets input from the plurality of incoming lines 11 _{1 to} 11 _M and then filters the received packets to appropriately output the outgoing line 14 _1. It outputs to ~ 14 _M.

However, in the apparatus shown in FIG. 3, the rate conversion memories 16 _{1 to} 16 _M are provided at the output stage of the filtering unit 13 so that the rate conversion of packets is performed after filtering. The configuration is different from the device.

The apparatus shown in FIG. 4 is provided with a filtering unit 17 _m for each incoming line 11 _m .
In other words, the apparatus shown in FIGS. 2 and 3 described above, by time-division multiplexing the received packets that are input from a plurality of incoming lines 11 ₁ to 11 _M, to filter for each of a plurality of incoming lines 11 ₁ to 11 _M In contrast to this, the device shown in FIG. 4 is configured to filter every incoming line 11 _m .

[0015]

As described above, the devices shown in FIGS. 2 to 4 output packets directly to the outgoing line by filtering so that the packets are copied and simultaneously exchanged. It has become.

However, in these devices,
Since a header conversion unit for converting the header information of the packet has to be provided on both the incoming line side and the outgoing line side, there is a problem that the amount of hardware increases.

This is because the header conversion unit needs a header conversion unit for converting the header information into the in-switch identification number for filtering and a header conversion unit for converting the header information into the outgoing line channel number after the exchange processing. Is.

Therefore, according to the present invention, it is possible to prevent an increase in the amount of hardware due to the necessity of providing the header converting section at two places, and further impair the high-speed communication of packets which do not require copying. It is an object of the present invention to provide a packet copy device that can prevent the packet from being lost.

[0019]

To achieve the above object, the present invention determines whether a received packet is a first packet requiring copying or a second packet not requiring copying, If the packet is 1, the received packet is repeatedly copied by the number of routes (route to the destination terminal or route to the adjacent exchange) defined by the identification number (channel number or destination number), Second
If it is a packet of
The received packet is preferentially transmitted.

[0020]

According to the above construction, when the received packet is the first packet, the received packet is repeatedly copied by the number of routes defined by the identification number and then sent to the route. It

On the other hand, when the received packet is the second packet, this received packet is sent to the route without undergoing the copy process. In this case, since the copy operation of the first packet is forcibly stopped, even when the copy operation is actually being executed, the second packet is sent without waiting for the end of this copy operation. be able to. Accordingly, it is possible to prevent the high speed communication of the second packet from being impaired by the copy operation.

Further, since the method of copying the necessary amount of packets by shifting in time is used as the first packet copying method, the existing switching switch can be used as the packet switching switch. As a result, the header conversion can be performed in only one place, so that an increase in the amount of hardware can be prevented.

[0023]

Embodiments of the present invention will now be described in detail with reference to the drawings. FIG. 1 is a block diagram showing the overall configuration of the first embodiment of the present invention.

In this embodiment, a copy function, which is a feature of the present invention, is provided for each incoming line.

In the figure, 21 _{1 to} 21 _M (M is an integer of 2 or more) are incoming lines of received packets. Each incoming line 21
Received packets input from _m (m = 1, 2, ..., M) correspond to corresponding first-in first-out memory units (hereinafter, “FIFO”).
It is supplied to 22 _m of "memory section". The packets supplied to the respective FIFO memory units 22 _m are sequentially output from the one input first.

This output packet is sent to the packet decision unit 22 _m.
Is supplied to. The received packet supplied to the packet determination unit 22 _M is used to determine whether this packet is a “broadcast packet” or a “general packet”. Here, the "broadcast packet" is a packet that requires copying, and the "general packet" is a packet that does not require copying. The details of this determination processing will be described later.

The received packets subjected to the determination process of the packet determining section 23 _m is supplied to the switch 25 _m. Further, the determination result of the packet determination unit 23 _m is the control unit 2
Supplied for 6 _m .

The connection state of the switch 25 _m is controlled by the control unit 26 _m based on the determination result of the packet determination unit 23 _m . That is, the movable contact a of the switch 25 _m
When the determination result of the packet determination unit 23 _m is “general packet”, is connected to the fixed terminal b. Thus, in this case, the received packet is supplied as the input A to the selector 27 _m . On the other hand, when the determination result is “broadcast packet”, the movable contact a is connected to the fixed terminal c. Accordingly, in this case, the received packet is supplied to the FIFO memory unit 28 _m .

The received packets supplied to the FIFO memory unit 28 _m are sequentially output from the previously input packet.
This output packet is supplied to the packet copy unit 29 _m . The identification number (channel number or destination number) of the received packet supplied to the packet copy unit 29 _m
Are repeatedly copied by the number of routes defined by (the route to the destination terminal or the route to the adjacent exchange).

When the copy operation for one packet is completed, the copy operation as described above is executed again for the next received packet supplied from the FIFO memory unit 28 _m . For details on the copy process,
It will be described later.

By this copy operation, the packets sequentially output one by one from the packet copy unit 29 _m are supplied as the input B to the selector unit 27 _m . The selector unit 27 _m selects either the input A or the input B as the output Q.

This selecting operation is controlled by the control unit 26 _m based on the judgment result of the packet judgment unit 23 _m .
That is, the selector unit 27 _m selects the input A when the determination result is “general packet”, and selects the input B when the determination result is “broadcast packet”.

The selection output Q of the selector section 27 _m is supplied to a header converting unit 30 _m. The received packet supplied to the header conversion unit 30 _m undergoes processing for converting the header information into new header information (new identification number and switch outgoing line number). The received packet whose header information has been converted is supplied to the packet switching switch 31. The received packet supplied to the packet switching switch 31 is output to any one of the plurality of outgoing lines 32 _{1 to} 32 _N (N is an integer of 2 or more) based on the switch outgoing line number added to the header section. To be done.

The judgment operation of the packet judgment unit 23 _m is as follows.
This is done based on the copy number information registered in the copy number information registration table 33. Here, the copy number information is information indicating the copy number of the received packet. This copy number is an integer value of "2" or more when the received packet is a "broadcast packet", and is "1" when it is a "general packet".

The address of the copy number information registration table 33 is defined by the identification number of the received packet. Corresponding copy number information is registered in each address. This registration is performed by a network management device (not shown) in advance, for example, based on an instruction from the packet transmission source.

The copy number information read from the copy number information registration table 33 for determining the received packet is further supplied to the packet copy unit 29 _m via the FIFO memory unit 34 _m . As a result, in the packet copy unit 29 _m , the received packets are copied by the copy number represented by the copy number information.

When a "general packet" is received during this copy operation, this copy operation is stopped. The abort state, "General Packet" is continued during selected in the selector 27 _m. This stop control is performed by the control unit 2
_Made by 6 _m .

The above is the overall configuration of FIG. Next, a specific configuration of each unit in FIG. 1 will be described. FIG. 5 is a block diagram showing an example of a specific configuration of the packet determination unit 23 _m .

[0039] In Figure, the received packets supplied from the FIFO memory 22 _m is stored alternately in the buffer circuit 232 _m, 233 _m through the switch 231 _m.
The received packets accumulated in the buffer circuits 232 _m and 233 _m are alternately read out via the switch 233 _m ,
It is supplied to the switch 25 _m and the header information extraction circuit 235 _m .

The connection state of the switches 231 _m and 234 _m is controlled by the control unit 26 _m . In this case, when the switch 231 _m is connected to the buffer circuit 232 _m , the switch 234 _m is connected to the buffer circuit 233 _m .
As a result, in this case, the current reception packet output from the FIFO memory unit 22 _m is accumulated in the buffer circuit 232 _m , and the previous reception packet accumulated in the buffer circuit 233 _m is output via the switch 234 _m. To be done.

On the other hand, when the switch 231 _m is connected by the buffer circuit 233 _m , the switch 234 _m is connected to the buffer circuit 232 _m . As a result, in this case, the current reception packet output from the FIFO memory unit 22 _m is accumulated in the buffer circuit 233 _m , and the previous reception packet accumulated in the buffer circuit 232 _m is output via the switch 234 _m. To be done.

The header information extraction circuit 235 _m executes a process of extracting header information from the received packet. The header information extracted by this is supplied to the copy number information reading circuit 236 _m .

In the copy number information reading circuit 236 _m , a process of reading the copy number information from the header number information registration table 33 is executed based on the supplied header information. The read address in this case is defined by the identification number included in the header information. Copy number information read by the copy number information readout circuit 236 _m are provided to the decision circuit 237 _m.

The decision circuit 237 _m decides whether the received packet is a "general packet" or a "broadcast packet" based on the supplied copy number information. In this case, when the copy number represented by the copy number information is "1", it is determined as "general packet", and when it is an integer of "2" or more, it is determined as "broadcast packet". The result of this determination is supplied to the control unit 26 _m, and is used to control the switch 25 _m , the selector unit 27 _m , and the packet copy unit 29 _m .

FIG. 6 is a block diagram showing an example of a concrete configuration of the packet copy unit 29 _m .

In the figure, the received packet ("broadcast packet") supplied from the FIFO memory unit 28 _{m is} supplied to the packet writing circuit 291 _m , and the copy memory 29 _m is supplied.
_Written to 2 _m . The received packet written in the copy memory 292 _m is repeatedly read by the packet reading circuit 293 _m . In this case, the read count is FI
It is defined by the copy number information supplied from the FO memory unit 34 _m . This read output is supplied to the selector unit 27 _m .

[0047] Incidentally, during a read operation of the packet read circuit 293 _m, the "General Packet" is received, as described later, the read operation is forcibly stopped. Moreover, this stop control is executed regardless of whether or not the copy operation is actually performed.

FIG. 6 is a flowchart showing the control processing of the control unit 26 _m when the determination result is supplied from the packet determination unit 23 _m .

In this control process, it is first determined whether the determination result is "general packet" or "broadcast packet" (step S1). In the case of "general packet", the process of stopping the read operation of the packet read circuit 293 _m is executed (step S2). After this, switch 25
A process of connecting the movable contact a of _m to the fixed terminal b and setting the selection output Q of the selector unit 27 _m to the input A is executed (step S3). This process received packet output from the packet determining section 23 _m is, the switch 25 _m, or header converting section via the selector 27 _m 3
When it is supplied to 0 _m , processing for canceling the suspension of the read operation of the packet read circuit 293 _m is executed (step S4). As a result, the end state is reached.

On the other hand, when it is judged as the "broadcast packet" in step S1, the movable contact a of the switch 25 _m is
Is connected to the fixed terminal c, and the selector unit 27 _m executes the process of setting the selection output Q to the input B (step S5). In this case, this is the end state.

As is clear from FIG. 7, when the determination result is "general packet", the read operation is stopped regardless of whether the packet read circuit 293 _m is currently performing the read operation. The process is executed. This is to save the trouble of determining whether or not the reading operation is being performed, and there is no problem even if this determination is not performed. The same applies to the point that the current states of the switch 25 _m and the selector unit 27 _m are not determined.

FIG. 8 is a block diagram showing an example of a concrete configuration of the header converting section 30 _m .

[0053] In Figure, the received packet output from the selector unit 27 _m is supplied to the header information replacement circuit 301 _m and the header information extraction circuit 302 _m. In the header information extraction circuit 302 _m , a process of extracting header information from the received packet supplied is executed. The extracted header information is supplied to the header information reading circuit 303 _m .

In the header information read circuit 303 _m , the new header information registration table 304 _m is accessed based on the supplied header information to execute the process of reading the new header information from the table 304 _m. .. The read address in this case is defined by the identification number included in the header information.

The new header information read by the header information reading circuit 303 _m is the header information replacing circuit 301 _m.
Is supplied to. As a result, the header information of the received packet output from the selector unit 27 _m is replaced with the new header information. This replacement output is supplied to the packet switching switch 31. As a result, the received packet has the outgoing line 32 defined by the switch outgoing line number included in the header information.
output to _m .

The new header information registration table 304 _m
In the address corresponding to "general packet",
One piece of new header information is registered. In contrast,
New header information for a plurality of routes defined by the identification number is registered in the address corresponding to the "broadcast packet". The plurality of pieces of new header information are read one by one each time one of the copied “broadcast packets” arrives, and are used for replacement processing.

The above is the configuration of the first embodiment. Next, the operation of the above configuration will be described.

[0058] In FIG 1, the received packet input from the incoming line 21 _m is supplied to the packet determining section 23 _m via the FIFO memory 22 _m. As a result, it is determined whether the received packet is a "general packet" or a "broadcast packet".

[0059] If the received packet is "General Packet", the movable contact a of the switch 25 _m is connected to the fixed terminal b. Further, the selection output Q of the selector unit 27 _m is input A
Is set to. Thus, the received packet, the switch 25 _m, the header converting unit 30 through the selector unit 27 _{m m}
Is supplied to.

On the other hand, when the received packet is a "broadcast packet", the movable contact a of the switch 25 _m is fixed terminal b.
Connected to. As a result, the received packet is sent to the FIFO.
It is supplied to the packet copy unit 29 _m via the memory unit 28 _m . As a result, this received packet is copied by the number of routes defined by the identification number.

In this case, the selection output Q of the selector unit 27 _m is set to the input B. As a result, the packets output from the packet copy unit 29 _m are supplied to the header conversion unit 30 via the selector 27 _m one by one.

When a new packet is received in this state, the above-mentioned determination processing is executed again. When it is determined by this determination process that the received packet is a “general packet”, the connection position of the movable contact a of the switch 25 _m is switched from the fixed terminal c to the fixed terminal b. The selection output Q of the selector unit 27 _m is changed from the input B to the input A.
Can be switched to. As a result, the received packet is again supplied to the header conversion unit 30 _m via the switch 25 _m and the selector unit 27 _m .

In this case, the control unit 26 _m executes control for stopping the copy operation of the packet copy unit 29 _m . As a result, the packet copy unit 29 _m is forced to stop the copy operation even if the copy operation is being continued. As a result, the received "general packet"
Without conflicting with the "broadcast packet" being copied,
Moreover, it is supplied to the header conversion unit 30 _m with priority over this “broadcast packet”.

When this supply is completed, the suspended state of the copy operation is released. As a result, the copy operation of the "broadcast packet" that was being copied is restarted.

In the received packet supplied to the header information converting section 30 _m , the header information is replaced with the new header information.
In this case, when the received packet is a "broadcast packet", a plurality of new header information assigned to this "broadcast packet" is added one by one each time a copy of this "broadcast packet" is given. It is read out and subjected to replacement processing.

The above is the operation of the first embodiment. Here, this operation will be described in more detail based on a specific communication example.

FIG. 9 shows terminals a 1 and b 2 accommodated in packet switch B from terminal a accommodated in packet switch A.
, B3 and the terminal c accommodated in the packet switch C are broadcast. In this case, each of the exchanges A, B, C has the above-mentioned packet copy device built therein.

In the following description, in order to simplify the explanation, a conceptual explanation will be given without following the actual format of the packet. Therefore, the identification number X of the packet P is represented by b1, b2, b3, c, and the switch output line number Y is represented by B, C, b1, b2, b3, c, *.
Here, * indicates that the switch outgoing line number Y has no meaning.

First, the registration contents of the copy number information registration table 33 and the header conversion information registration table 304 _{m in} the exchange A will be described.

In the case of the broadcast communication as described above, the exchange A
Needs to transmit the packet sent from the terminal a to the exchanges B and C. Therefore, the copy number of the received packet in the exchange A is "2". As a result, the copy number information registration table 3 of this exchange A
In 3, the information indicating the copy number “2” is registered. The storage address in this case is the identification number X (b1, b2, b3
, C).

In order for the exchange A to transmit the packet P to the exchange B, the identification number X is set to (b1, b2, b3,
It is necessary to change from c) to (b1, b2, b3).
Further, the switch output line number Y needs to be changed from (*) to (B).

Similarly, in order for the exchange A to transmit the packet to the exchange C, the identification number X is (b1, b2, b3,
It is necessary to change from c) to (c). Further, it is necessary to change the switch output line number Y from (*) to (C).

Therefore, in the header information registration table 304 _m of the exchange A, the above-mentioned two sets of information {(b1, b
2, b3), (B)}, {(c), (C)} are stored. The storage address in this case is also the identification number X (b1, b
2, b3, c).

Next, the registration contents of the copy number information registration table 33 and the header information registration table 304 _{m in} the exchange B will be described.

In the case of the broadcast communication as described above, the exchange B
Sends packets sent from the exchange A to the terminals b1 and
It is necessary to transmit to b2 and b3. Therefore, the copy number information registration table 33 of the exchange B stores information indicating the copy number "3". The storage address in this case is defined by the identification number X (b1, b2, b3).

Further, in order for the exchange B to transmit the packet to the terminal b1, it is necessary to change the identification number X from (b1, b2, b3) to (b1). Further, it is necessary to change the switch output line number Y from (B) to (b1).

Similarly, the exchange B selects the terminal b2 or b3.
In order to transmit the packet to the packet, the identification number X is (b1,
It is necessary to change from (b2, b3) to (b2) or (b3). Further, it is necessary to change the switch output line number Y from (B) to (b2) or (b3).

Therefore, in this case, in the header conversion information registration table 304 _m , the above three sets of information {(b1
), (B1)}, {(b2), (b2)}, {(b3
), (B3)} are stored. The storage address in this case is defined by the identification number X (b1, b2, b3).

Next, the registration contents of the copy number information registration table 33 and the header information registration table 304 _{m in} the exchange C will be described.

In the case of the broadcast communication as described above, the exchange C
Needs to transmit the packet sent from the exchange A to the terminal c. Therefore, the copy number information registration table 33 of the exchange C stores information indicating the copy number "1". The storage address in this case is the address corresponding to the identification number X (c).

Further, in order for the exchange C to transmit the packet P to the terminal c, it is necessary to change the identification number X from (c) to (c) (substantially no change). Further, it is necessary to change the switch output line number Y from (C) to (c).
Therefore, in this case, the header information registration table 241
_A set of information {(c), (c)} described above is stored in _m . The storage address in this case is the identification number X (c).
Stipulated by

When the registration process as described above is completed, the actual communication becomes possible. In this communication, first, a packet is transmitted from the terminal a to the exchange A. The identification number X of this packet is set to (b1, b2, b3, c), and the switch outgoing line number Y is set to (*).

The packet supplied to the exchange A is as follows.
The packet determination unit 23 _m determines whether the packet is a “general packet” or a “broadcast packet”. in this case,
It is determined to be a "broadcast packet". This is because "2" is registered as the copy number at the address of the copy number information registration table 33 defined by the identification number X (b1, b2, b3, c) of this packet.

As a result, this packet is supplied to the packet copy unit 29 _m and copied twice. The two packets obtained by this copy operation are sequentially supplied to the header conversion unit 30 _m , and the header information is converted.

As a result, the identification number X of the first packet
From (b1, b2, b3, c) to (b1, b2, b3
), And the switch output line number Y is converted from (*) to (B). On the other hand, the second packet P
The identification number X of (b1, b2, b3, c) to (c)
The switch output line number Y is converted from (*) to (C)
Is converted to.

The first packet which has been subjected to the header conversion process is supplied to the exchange switch 25, and is supplied to the exchange B based on the switch outgoing line number Y (B). Similarly,
The second packet is also supplied to the exchange C based on the switch outgoing line number Y (C).

For the packet supplied to the exchange B, first,
It is determined whether the packet is a "general packet" or a "broadcast packet". In this case also, since the registered copy number is "3", it is determined to be a "broadcast packet".

As a result, after this packet is copied three times, the header information is converted. By this header conversion, the identification number X of the first packet is (b1, b2, b
3) is converted to (b1), and the switch output line number Y is converted from (B) to (b1). The identification number X of the second packet P is from (b1, b2, b3) to (b
2), and the switch output line number Y is converted from (B) to (b2). The identification number X of the third packet P is converted from (b1, b2, b3) to (b3), and the switch outgoing line number Y is converted from (B) to (b3). As a result, each packet is transmitted to the terminal b.
It is supplied to 1, b2 and b3.

The packets supplied to the exchange C are also
It is judged whether it is a "general packet" or a "broadcast packet". In this case, since the registered copy number is "1",
It is determined to be a "general packet".

As a result, this packet is transmitted to the switch 2
5 _m is supplied to the header conversion unit 30 _m via the selector unit 27 _m and the header information is converted. As a result, the identification number X of this packet is converted from (C) to (c), and the switch outgoing line number Y is converted from (C) to (c). As a result, this packet is supplied to the terminal c.

As described above, when the terminal a sends the packet once, this packet is transmitted to the plurality of terminals b1 and b.
2, b3, c.

According to this embodiment described in detail above, the following effects can be obtained.

(1) First, the header converters 30 ₁ to 30 _M
Need only be provided on the side of the incoming lines 21 _{1 to} 21 _M, it is possible to prevent an increase in the amount of hardware due to having to provide the header conversion section at two locations. This is because, as the packet copying method, a method of copying a necessary amount of packets with a temporal shift is used, and therefore an existing packet switching switch can be used as it is as the packet switching switch 31.

(2) Further, although the method of sequentially repeating copying is used as the method of copying the received packet, it is possible to prevent the transmission of the "general packet" from being delayed due to this copying operation. it can. This is because the type of the received packet is determined, and if it is a “general packet”, the copy operation of the “broadcast packet” is stopped and the “general packet” is preferentially transmitted. As a result, high-speed communication of "general packets" having a higher generation rate than "broadcast packets" can be maintained.

(3) Further, it can be designed more easily than the device shown in FIG. 2 or FIG. This is each incoming line 21
_{Since the} packet copy unit 29 _m is provided for each _m , the speed conversion memories 13 and 16 which must be accessed at a high speed required by the apparatus of FIG. 2 or FIG.
_{This is because 1 to} 16 _N are unnecessary.

(4) Further, the memory capacity can be reduced as compared with the device shown in FIG. In the case of the apparatus shown in FIG. 4, this requires a table and memory having a capacity proportional to M × N, whereas in the case of this embodiment, a copy number information registration table 33 having a capacity proportional to M, This is because the copy memory 292 _m may be provided.

[0097] (5) In addition, since as in front of each packet copy unit 29 _m provided FIFO memory 28 _m, 34 _m, while a configuration of sequentially copying the packet, a packet copying unit 29 _m at a high speed The advantage is that no access is required.

(6) Since the type of packet is determined by the number of copies, the copy number information registration table 33 is stored in the packet determination unit 23 _m and the packet copy unit 2
The advantage is that it can be shared at 9 _m .

(7) Further, when controlling the states of the switch 25 _m , the selector unit 27 _m , and the packet copying unit 29 _m , the control is performed regardless of the present state. It can be omitted.

FIG. 10 is a block diagram showing the configuration of the second embodiment of the present invention.

In the first embodiment, the copy number information read from the copy number information registration table 33 is stored in the FIFO.
The case of supplying the packet copy unit 29 _m via the O memory unit 34 _m has been described.

On the other hand, in this embodiment, this copy number information is added to the received packet and supplied by the adder circuit 35 _m .

With such a configuration, the copy number information can be supplied to the packet copy unit 29 _m by using the adder circuit 35 _m having a simpler structure than the FIFO memory unit 34 _m. As a result, the circuit scale can be reduced.

FIG. 11 is a block diagram showing the configuration of the third embodiment of the present invention.

In the first and second embodiments described above,
The case has been described in which the copy number information read from the copy number information registration table 33 is supplied to the packet copy unit 29 _m when determining the received packet.

On the other hand, in this embodiment, the packet copy section 29 _m is provided with a copy number information reading function.

This is, for example, the packet copy unit 29 _m.
First, a header information extraction circuit for extracting header information from the received packet and a copy number information registration table 33 based on the header information extracted by the header information extraction circuit.
This can be realized by providing a copy number information reading circuit for reading the copy number information from the.

Even with such a structure, the same effect as that of the first embodiment can be obtained.

FIG. 12 is a block diagram showing the structure of the fourth embodiment of the present invention.

In the above first to third embodiments, the case where the type of this packet is determined based on the copy number information of the received packet has been described.

On the other hand, in this embodiment, type identification information indicating the type of packet is registered in advance.

That is, in FIG. 12, reference numeral 36 is a type identification information registration table in which the type identification information indicating the type of packet is registered in advance. The address of the type identification information registration table 36 is defined by the identification number of the received packet. The type identification information of the packet is registered in each address.

With such a configuration, it is possible to omit the determination process for determining the type of packet from the registration information. Therefore, it is possible to determine the received packet more than in the first to third embodiments. The time required can be shortened.

FIG. 13 is a block diagram showing the structure of the fifth embodiment of the present invention.

In the above first to fourth embodiments, the case where the header converter 30 _m is provided in the output stage of the selector 27 _m has been described.

On the other hand, in this embodiment, the selector unit 2
Header converters 37 _m and 38 _m are provided at the 7 _m input stage. In other words, for each of the "general packet" transmission route and the "broadcast packet" transmission route,
Header conversion units 37 _m and 38 _m are provided.

In FIG. 13, the case where the above-mentioned structure is applied to the structure of the fourth embodiment is shown as a representative.

FIG. 14 is a block diagram showing the structure of the sixth embodiment of the present invention.

In the first to fifth embodiments, each incoming line 21
_The case where a packet copy device is provided for each _m has been described.

On the other hand, in this embodiment, received packets input via a plurality of input lines 21 _{1 to} 21 _M are time-division multiplexed, and a packet copy device is provided on the multiple output transmission line. Is.

That is, in the figure, reference numeral 41 is a time division multiplexing circuit 41 for time division multiplexing the reception packets input from the plurality of incoming lines 21 _{1 to} 21 _M. A distribution switch 42 outputs the time-division-multiplexed received packet to the outgoing line 32 _m defined by the switch outgoing line number.

FIG. 14 shows a typical case where the packet copying apparatus having the structure as described in the first embodiment is used as the packet copying apparatus. Therefore, in FIG. 14, the portions having the same functions as those in FIG. 1 are denoted by the reference numerals of FIG.

Even with such a structure, the same effects as (1), (3) to (7) of the first embodiment can be obtained. In addition, one device is provided with a plurality of incoming lines 21 _{1 to} 21.
Since it can be shared by _M , the circuit scale of the device can be reduced as compared with the previous embodiment.

Although the first and second embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments.

(1) For example, in the above first to fifth embodiments, the case where the copy number information registration table 33 and the type identification information registration table 36 are shared by all the incoming lines 21 _{1 to} 21 _M has been described. However, in the present invention, these may be provided for each incoming line 21 _m .

(2) Further, in the above-mentioned first to sixth embodiments, the case where the switches 25 _m and 25 are provided has been described. However, in the present invention, these may be omitted.

(3) In addition to this, the present invention has a function of sequentially copying "broadcast packets", and when a "general packet" is received, this copy operation is stopped,
Any configuration may be used as long as it is a configuration for transmitting a "general packet".

[0128]

As described in detail above, according to the present invention,
Since the header conversion unit may be provided in one place, it is possible to prevent an increase in the amount of hardware due to having to provide the header conversion unit in two places.

When the "general packet" is received, the copy operation of the "broadcast packet" is stopped, so that the high speed communication of the "general packet" can be prevented.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a first exemplary embodiment of the present invention.

FIG. 2 is a block diagram showing a configuration of a first example of a conventional packet copy device.

FIG. 3 is a block diagram showing a configuration of a second example of a conventional packet copy device.

FIG. 4 is a block diagram showing a configuration of a third example of a conventional packet copy device.

FIG. 5 is a block diagram showing an example of a specific configuration of a packet determination unit.

FIG. 6 is a block diagram showing an example of a specific configuration of a packet copy unit.

FIG. 7 is a flowchart showing processing of a control unit.

FIG. 8 is a block diagram showing an example of a specific configuration of a header conversion unit.

FIG. 9 is a block diagram for explaining the operation of the first embodiment.

FIG. 10 is a block diagram showing a configuration of a second exemplary embodiment of the present invention.

FIG. 11 is a block diagram showing a configuration of a third exemplary embodiment of the present invention.

FIG. 12 is a block diagram showing a configuration of a fourth exemplary embodiment of the present invention.

FIG. 13 is a block diagram showing the configuration of a fifth embodiment of the present invention.

FIG. 14 is a block diagram showing a configuration of a sixth exemplary embodiment of the present invention.

[Explanation of symbols]

21 _{1 to} 21 _M ... Incoming line, 22 _{1 to} 22 _M , 28 _{1 to} 28
_{M, 34 1 ~34 M, 28,34} ... FIFO memory unit,
23 ₁ to 23 _M , 23 ... Packet determination unit, 25 _{1 to} 25
_M, 25 ... switch, 26 ₁ ~26 _M, 26 ... control unit,
27 _{1 to} 27 _M , 27 ... Selector unit, 29 _{1 to} 29 _M ,
29 ... Packet copy unit, 30 ₁ to 30 _M , 30 ... Header conversion unit, 31 ... Exchange switch, 32 _{1 to} 32 _N ... Outgoing line, 33 ... Copy number information registration table, 35 _{1 to} 35 _M
... Addition circuit, 36 ... Type identification information registration table, 41 ...
Time division multiplexing unit 42 ... Distribution switch unit.

Front Page Continuation (72) Inventor Fumatoshi Masaki 1-7-12 Toranomon, Minato-ku, Tokyo Oki Electric Industry Co., Ltd.

Claims (1)

[Claims] 1. A packet copy device provided in a packet switch for copying a received packet for broadcast communication, wherein the received packet is a first packet requiring copying or a second packet not requiring copying. If the received packet is determined to be the first packet by the packet determination unit that determines whether the packet is a packet, the received packet is repeatedly copied by the number of routes defined by the identification number. A packet copy means, and a packet priority means for stopping the copy operation of the packet copy means and outputting the received packet with priority when the received packet is judged to be the second packet by the packet judgment means. , The packets output from the packet copy means and the packet priority means are identified by Header conversion means for converting the header information of this packet for transmission to the route specified by the number, and when the first packet is newly received during the copy operation of the packet copying means, A packet copy device, comprising: a packet storage unit that stores received packets until the copying operation is completed.