JP4228850B2 - Packet relay device - Google Patents

Description

The present invention relates to a packet relay device, in particular, it relates to packet relay device designed to reduce the loss of packets.

Conventionally, many packet relay methods have been proposed in which a specific packet is relayed preferentially over other packets so that important packets are transmitted preferentially.
For example, on the packet transmission side, an identifier indicating priority is added to the packet, and a plurality of transmission queues having priority are prepared in the transmission control unit of the packet relay apparatus, and the identifier indicating the priority added to the packet Thus, this packet is entered in the transmission queue of the corresponding priority among the transmission queues prepared in the transmission control unit, and the transmission control unit transmits the packets in order from the packet entered in the higher priority queue. A method of relaying preferentially is common.

In addition, an area for storing a packet from a transfer source that wants to prevent the packet from being discarded at the time of relaying is secured, and the packet from the transfer source is a dedicated area separately from other packets. A method has been proposed in which the packet is prevented from being discarded by storing the packet (for example, see Patent Document 1).
JP-A-12-106571

  However, as described above, when a specific packet is relayed preferentially, the specific packet can be relayed preferentially, but the network load on the communication line increases, and the packet is received from the communication line. When the amount of packets input to the relay device for relaying packets to the destination communication device exceeds the relay performance for the packet, the reception resources for receiving the packet on the relay device side are depleted. At some point, packet loss will occur.

Further, as described above, in order to relay a specific packet with priority, it is necessary to perform processing such as adding an identifier to the packet in consideration of the priority order on the transmission side of the packet. It is necessary to incorporate it into other devices. However, when the communication device on the packet transmission side is configured by a general-purpose device such as a personal computer, it is difficult to incorporate a process for adding an identifier indicating a priority order into the device.
Therefore, the present invention has been made paying attention to the above-mentioned conventional unsolved problems, and an object thereof is to provide a packet relay method and a packet relay device capable of easily reducing packet loss. .

To achieve the above object, a packet relay apparatus according to claim 1 of the present invention temporarily stores a packet from a communication line in a reception area, and then relays the packet in the reception area to a predetermined relay destination. In the packet relay apparatus, the reception load detection means for detecting the reception load of the packet from the communication line, the reception load of the packet detected by the reception load detection means exceeds a preset load threshold, and when receiving a packet other than the specific packet that has been set in advance, comprising: a packet discarding means for discarding the packet, the packet condition setting means for setting a packet condition for specifying the specific packet, wherein the packet Including a data body to be communicated and communication information necessary for communicating the data body via the communication line, The condition setting means sets a packet condition for each of the data body and the communication information, and the packet discarding means sets the specific condition when both the data body and the communication information satisfy the packet conditions. It is characterized by determining that it is a packet .
The packet relay apparatus according to claim 2 is characterized in that the data body includes numerical information, and the packet condition setting means sets a numerical range of the numerical information as the packet condition.

In the first aspect of the invention, the packet from the communication line is temporarily stored in the reception area, and then predetermined processing is performed on the packet in the reception area and relayed to a predetermined relay destination.
At this time, the reception load of the packet from the communication line is monitored by the reception load detection means, and if this reception load exceeds a preset threshold value, the packet received in this state is set in advance. If the packet is a packet other than the specific packet, the packet is discarded by the packet discarding means.

  Here, when the reception load of the packet from the communication line increases, there is no free area in the reception area, which results in packet loss. However, when the reception load increases, exceeds the threshold value, and a packet other than the specific packet is received, it is discarded. Therefore, an empty area can be secured in the reception area. For this reason, it is possible to avoid packet loss due to the absence of a free area in the reception area, and to discard packets other than the specific packet, so that the specific packet is reliably relayed. Is possible.

Also includes a packet condition setting means for setting a packet condition for specifying the specific packet, the packet discarding means, based on the set packet condition in the packet condition setting means, whether a particular packet Determine whether or not .
Therefore, any packet can be set as a specific packet by changing and setting the packet condition by the packet condition setting means.

As described above, according to the engagement Rupa packet relay apparatus in claim 1 of the present invention, the received load detected by the reception load detecting means packet exceeds a load threshold set in advance, was and preset When a packet other than a specific packet is received, this packet is discarded by the packet discarding means, so that it is possible to avoid the loss of the specific packet even when the reception load increases. .

Further, to set the packet condition for specifying a particular packet by packet condition setting means, the packet discarding unit based on the set packet condition in the packet condition setting means, so as to perform the determination of whether a particular packet Therefore, a packet suitable for this can be set as a specific packet in accordance with a change in the configuration of the network system.

Hereinafter, embodiments of the present invention will be described.
First, a first embodiment will be described.
FIG. 1 is an example of a network system to which the present invention is applied. Each communication device 5 is connected to the duplex line L via the relay device 10, and exchanges data with other communication devices 5 via the relay device 10 and the duplex line L.
The relay device 10 executes a known duplex communication process to exchange data with other communication devices through two lines, and has a relay function of the OSI reference model layer 3 and the network layer, and is duplexed. When a received packet from the line L is relayed to the communication device 5, only a packet that satisfies a preset relay condition is relayed.

  FIG. 2 is a schematic configuration diagram illustrating an example of the relay device 10, and includes a transmission relay unit 21 that relays data from the communication device 5 to the lines L 1 and L 2, and the communication device 5 from the line L 1 or L 2. Data transmission / reception processing between the line L1 reception relay unit 25a and the line L2 reception relay unit 25b, the transmission relay unit 21 and the reception relay unit 25a, and the line L1 for data relay. The network interface unit 27a for the line L1, the network interface unit 27b for the line L2 that performs data transmission / reception processing between the transmission relay unit 21 and the reception relay unit 25b, and the line L2, and the duplex communication processing unit 15. The duplex communication processing unit 15 transmits the reception data from the reception relay units 25a and 25b to the communication device 5 and receives the same data from the reception relay unit 25a and the reception relay unit 25b. For example, a known duplex communication process such as discarding data received later is performed.

The reception relay units 25a and 25b have the same configuration, and as shown in FIG. 2, received packets from the line L1 or L2 are stored in a line side reception buffer (hereinafter also referred to as a reception buffer) 31. The buffer management unit 32 manages the number of received packets stored in the line-side reception buffer 31 and the storage position thereof.
The reception packet stored in the line side reception buffer 31 is transferred to the communication apparatus side transmission buffer (hereinafter also referred to as a transmission buffer) 34 by the relay processing unit 33 in a known procedure. At this time, the relay processing unit At 33, in order to avoid packet loss at the relay device 10, only packets that satisfy the relay conditions set in advance and stored in the predetermined storage area 35 are transferred to the transmission buffer 34. In addition, it includes condition setting means 36 such as an input device for changing and setting the relay conditions stored in the storage area 35, and relays according to changes in the network system configuration and the contents of data exchanged between the communication devices. The conditions can be changed.

FIG. 3 is a flowchart illustrating an example of a processing procedure of the packet reception process executed when the relay processing unit 33 receives a packet from the line L1 or L2. Since the reception relay units 25a and 25b are configured identically, here, the packet reception processing in the relay processing unit 33 of the reception relay unit 25a will be described.
When the relay device 10 recognizes that the packet has been received from the line L1 based on the notification from the buffer management unit 32, etc., the relay device 10 proceeds from step S1 to step S2 and uses the reception buffer 31 managed by the buffer management unit 32. Get information. Specifically, the number of received packets stored in the reception buffer 31 (hereinafter referred to as packet storage number Np) is read.

Next, the process proceeds to step S3, and the reference value Nth of the packet stored in the reception buffer 31 that is preset and stored in the storage area 35 is read out. This reference value Nth is for determining the degree of the network load of the line L1. This reference value Nth is stored in the storage area 35 by operating the condition setting means 36.
The reference value Nth is the packet processing time when the time required to make a space in the reception buffer 31 by transferring the reception packet stored in the reception buffer 31 to the transmission buffer 34 is the packet processing time. The maximum number of packets stored in the reception buffer 31 that can store all packets received from the line L1 between them in the reception buffer 31, and the number of packets in the reception buffer 31 is the reference value Nth Is set to a value that allows it to be considered that all packets from the line L1 cannot be stored in the reception buffer 31.

  Next, the process proceeds to step S4, and the packet storage number Np acquired in step S2 is compared with the reference value Nth of the number of packets in the reception buffer 31 acquired in step S3. When the packet storage number Np is smaller than the reference value Nth, the network load on the line L1 is relatively small, and the number of packets stored in the reception buffer 31 from the line L1 can be predicted to be relatively small. It is determined that 31 will not be depleted, and the process proceeds to step S7 described later.

On the other hand, when the packet storage number Np is equal to or greater than the reference value Nth, the network load is relatively large, and the number of packets stored in the reception buffer 31 from the line L1 is relatively large. It is determined that there is a possibility of missing, and the process proceeds to step S5.
In step S5, a relay / discard determination process, which will be described later, is performed to determine whether the newly received packet is to be relayed to the communication device 5 or discarded.
If it is determined in this relay / discard determination process that the relay is to be performed, the process proceeds to step S7, and the reception packet stored in the reception buffer 31 is transferred to the transmission buffer 34 by a known procedure. On the other hand, if it is determined not to relay, the process proceeds to step S8 where the received packet is discarded from the reception buffer 31.
FIG. 4 is a flowchart illustrating an example of a processing procedure of the relay / discard determination process.

In this relay / discard determination process, first, in step S11, a received packet to be determined is referred to and, for example, a protocol type such as TCP / IP or UDP / IP is extracted.
Next, the process proceeds to step S12, and the relay protocol type stored in the storage area 35 is read out by operating the condition setting means 36 in advance. This relay protocol type is a protocol type for determining whether or not to relay a packet to be determined. For example, the relay protocol type is set to a protocol type corresponding to a communication destination apparatus having a high priority.
Then, the process proceeds to step S13, the protocol type of the received packet is compared with the relay protocol type, and if they match, the process proceeds to step S14, and it is determined that the received packet to be determined is relayed. On the other hand, if they do not match, the process proceeds to step S15, and the received packet for determination handling is determined as a packet to be discarded.

  In other words, the loss of the packet occurs due to the reception resource for receiving the packet from the line L1 of the relay device 10, that is, the reception buffer 31 becoming empty due to an increase in the network load. When the number of stored packets Np in the reception buffer 31 is equal to or greater than the reference value Nth, that is, when the free space in the reception buffer 31 is reduced, a newly received packet is In the case of a packet with a low priority, the packet is discarded to secure a free space in the reception buffer 31, so that a packet can be received from the line L1, and communication is performed for a packet with a high priority. By relaying to the device 5, it is possible to avoid missing high priority packets and to reliably relay to the communication device 5. There.

Next, the operation of the first embodiment will be described.
In the network system of FIG. 1, the communication protocol type between the communication device 5a and the communication device 5b is protocol A, and the communication protocol type between the communication device 5a and the communication device 5c is protocol B. In addition, priority is given to packet communication between the communication device 5a and the communication device 5b.
Assuming that a packet is lost in the relay device 10a that relays the packet to the communication device 5a, first, the condition setting unit 36 is operated, and as a relay protocol type used for relay determination in the relay device 10a, Protocol A, which is a type of communication protocol with the communication device 5b that prioritizes packet communication, is set and stored in the storage area 35 in advance. Further, a reference value Nth of the number of packets stored in the reception buffer 31 for determining the network load is determined and stored in the storage area 35.

In this state, when a packet is transmitted from the communication device 5b to the communication device 5a, this packet is stored in the reception buffer 31 of the relay device 10a, and the buffer management unit 32 newly stores the packet in the reception buffer 31. Is recognized, this is notified to the relay processing unit 33.
When the relay processing unit 33 recognizes the reception of the packet, the process proceeds from step S1 in FIG. 3 to step S2, and reads the packet storage number Np of the reception buffer 31 as the usage status of the reception buffer. The stored reference value Nth is compared (steps S3 and S4).

Here, when the packet storage number Np stored in the reception buffer 31 is relatively small and falls below the reference value Nth, it is determined that the network load is small, and the process proceeds from step S4 to step S7. The packet is transferred to the transmission buffer 34 at a predetermined timing.
Since the above processing is performed in the same way in the reception relay units 25a and 25b, each transmission buffer 34 stores a reception packet consisting of the same information, and the duplex communication processing unit 15 stores this at a predetermined timing. Read and transmit to the communication device 5. At this time, if the packet read from the transmission buffer 34 of one of the reception relay units 25a and 25b is the same as the already transmitted packet from the other reception relay unit, this packet should be transmitted. In other words, measures are taken such as avoiding the same packet received by a different route from being discarded and transmitted to the communication device 5 again.

At this time, packets are not discarded, but the packet storage number Np does not exceed the reference value Nth, and the reception buffer 31 can be regarded as having enough space to store packets from the network. Packets from the network are never lost.
From this state, the network load increases, and is stored in the reception buffer 31 because of the relationship between the number of packets stored in the reception buffer 31 from the network and the time required for processing of the reception packet of the reception buffer 31 in the relay device 10. When the number of stored packets increases and the packet storage number Np exceeds the reference value Nth, the process proceeds from step S4 to step S5, and the relay / discard determination process is executed for the received packet newly stored in the reception buffer 31. Is done.

At this time, since the protocol A is set as the relay protocol type in the storage area 35, if the protocol type of the received packet is the protocol A, it is determined to “relay”, and if it is not the protocol A, , “Discard”.
Therefore, for example, in the case of a protocol B packet, that is, a packet from the communication device 5 c is discarded from the reception buffer 31.
For this reason, the reception buffer 31 is freed up as much as the packet from the communication device 5c is discarded. After that, when a packet from the communication device 5b having a higher priority is received, this is stored in the reception buffer 31. Can be stored. Therefore, in the relay device 10, it is possible to avoid a packet having a high priority from being lost due to the reception buffer 31 being empty.

At this time, the packet of protocol A having a high priority is relayed without being discarded, so that the packet from the communication device 5b having a high priority can be reliably relayed to the communication device 5a. .
At this time, as a method of making a space in the reception buffer 31, a packet having a low priority is discarded. Here, a method of creating a space in the reception buffer 31 by transferring the reception packet stored in the reception buffer 31 to the transmission buffer 34 is also conceivable. However, since it is possible to perform processing in a shorter time when the packet is discarded from the reception buffer 31 than when the packet is transferred from the reception buffer 31 to the transmission buffer 34, the reception buffer 31 becomes empty at an earlier stage. Can be made. Therefore, it is possible to quickly secure a space in the reception buffer 31 accordingly, so that it is possible to more reliably avoid the loss of packets with high priority.

  At this time, the relay device 10 determines whether or not to relay based on the protocol type information originally included in the received packet. Therefore, since the transmission source communication device 5 may transmit the packet without being aware of anything, even if the communication device 5 is configured by an existing device, the communication device 5 does not do anything. It is not necessary to add a hand and can be realized easily. Therefore, even when an existing personal computer or the like is applied as the communication device 5, it can be easily realized without taking any measures against them.

At this time, the relay condition stored in the storage area 35, that is, the condition table and the reference value Nth of the number of stored packets can be changed by the condition setting means 36.
Therefore, even if the network system configuration is changed, the content of data exchanged between the communication devices 5 is changed, or the priority packet type is changed, the relay registered in the storage area 35 is used. By changing the conditions, appropriate relay conditions can be set according to changes in the system configuration of the network.

Next, a second embodiment of the present invention will be described.
The second embodiment is the same as the first embodiment except that the procedure of the relay / discard determination process in the packet reception process of FIG. 4 is different. And detailed description thereof is omitted.
In the relay / discard determination process in the second embodiment, it is determined whether to relay or discard a packet using the condition table shown in FIG.
As shown in FIG. 5, the condition table is composed of a packet relay table T1 including a plurality of entries and the number of registered entries. The entry registered in the packet relay table T1 includes a protocol type such as TCP / IP and UDP / IP, and a sublist pointer indicating a storage destination of the sublist table T2 in which more detailed determination conditions are stored. ing.

  The sublist table T2 includes a plurality of entries and the number of registered entries, and each entry includes offset information indicating a position from the head of the received packet, and a position specified by the offset information. Size information such as byte, word, long indicating the size of data, information stored in an area specified by the size information from a position specified by the offset information, for example, a source address, a destination address, Judgment conditions such as “=, ≠, <,>, ≦, ≧” set for the code information representing the specified value set in advance for the data value and the specified value specified by the code information It consists of and.

The condition table set as described above is set in advance based on, for example, the network configuration or specifications, and stored in the storage area 35.
FIG. 6 is a flowchart illustrating an example of a processing procedure of relay / discard determination processing in the second embodiment.
In the relay / discard determination process in the second embodiment, first, the variable i is set to i = 0 in step S21, and then the process proceeds to step S22. The packet relay table T1 in FIG. Read E1.

Next, the process proceeds to step S23, the protocol type is extracted from the received packet to be determined, and then the process proceeds to step S24, where the variable i and the entry number E1 are compared.
When the variable i is smaller than the number of entries E1, the process proceeds to step S25, and the protocol type is read from the i-th registered entry specified by the variable i in the packet relay table T1. Note that the entry of the packet relay table T1 is the “0” -th entry at the head entry.

In step S26, it is determined whether the protocol type of the i-th registered entry matches the protocol type of the received packet extracted in step S23. Then, the process proceeds to step S27, the variable i is incremented by “1”, and then the process returns to step S24.
Similarly, the protocol type is read from the i-th entry registered in the packet relay table T1, that is, the next entry, and it is determined whether or not this matches the protocol type of the received packet to be determined. (Steps S25 and S26).

  This process is repeated, and when the protocol type of the i-th entry registered in the packet relay table T1 matches the protocol type of the received packet to be determined, the process proceeds to step S31. On the other hand, when the variable i exceeds the number of entries E1 in step S24, the process proceeds to step S28, where it is determined that the received packet does not satisfy the determination condition registered in the condition table, that is, does not satisfy the relay condition. Determine that the packet is to be discarded.

On the other hand, in step S31, the variable j is set to j = 0, and then the process proceeds to step S32. The i-th entry of the packet relay table T1 specified by the variable i at this time is referred to, and its sublist pointer is referred to. Reference is made to the sublist table T2 specified by Then, the number of entries E2 is extracted.
Subsequently, the process proceeds to step S33, where the variable j is compared with the number of entries E2. If the variable j is smaller than the number of entries E2, the process proceeds to step S34, and j specified by the variable j in the sublist table T2 Refer to the second entry and extract its code information. The top entry of the sublist table T2 is the “0” th entry.

Next, the process proceeds to step S35, and the offset information and size information are extracted by referring to the jth entry specified by the variable j. Then, the code information is extracted from the position of the reception packet to be determined, which is specified by the offset information and the size information.
Then, the process proceeds to step S36, and the determination condition is extracted from the jth entry specified by the variable j. Then, the process proceeds to step S37, where the code information of the j-th entry extracted in step S34 is the left side, the code information of the received packet extracted in step S35 is the right side, and these are compared under the determination conditions extracted in step S36. An expression is formed, and it is determined whether or not the code information of the received packet satisfies this determination expression.

When the code information of the received packet satisfies the determination formula formed in step S37, the process proceeds to step S38, the variable j is incremented by “1”, and the process returns to step S33. If the variable j is smaller than the number of entries E2 in the sublist table T2, whether or not the condition specified by the jth entry is satisfied is determined in the same manner as described above.
Then, when the variable j becomes equal to or greater than the number of entries E2 in the process of step S39, it is determined that the determination conditions specified in all entries of the sublist table T2 are satisfied, and the process proceeds to step S39, where the received packet to be determined is determined. Determine to relay.
On the other hand, if the code information extracted from the sublist table T2 and the code information extracted from the received packet do not satisfy the prescribed determination condition in step S37, the process proceeds to step S27, and thereafter the same processing as described above is performed. I do.

Next, the operation of the second embodiment will be described.
Now, in the communication device 5a, in the packet from the communication device 5b, the numerical value of the information X consisting of numerical data is n (50), and the numerical value of the information Y consisting of numerical data is n (10) or more n ( 50) The following shall be relayed preferentially.
As shown in FIG. 7A, the information X is stored in the area specified by the offset “off1” and the size “s1” of the packet from the communication device 5b, and the information Y is offset “off2”. “And the size“ s1 ”. The communication protocol between the communication device 5b and the communication device 5a is assumed to be protocol A.

  First, a condition table shown in FIG. 8 is created in advance. As an entry of the packet relay table T1, an entry having the protocol type of protocol A is formed, and an entry t1 for defining numerical data of the information X is generated as an entry of the sublist table T2. Specifically, the offset is set to “off1”, the size is set to “s1”, the code defining the numerical data of the information X is set to “n (50)”, and the determination condition is set to “=”. Further, as an entry for defining the numerical data of the information Y, the offset is “off2”, the size is “S1”, the code for defining the lower limit value of the numerical data of the information Y is “n (10)”, and the determination condition is An entry t2 set as “≦”, an offset “off2”, a size “s1”, “n (50)” as a code that defines the upper limit value of numerical data of information Y, and a determination condition as “≧” The entry t3 to be set is set.

If there are other packets to be relayed, they are set in the same manner. The same setting is made for packets of other communication protocols.
As shown in FIG. 7A, the code stored in the area specified by the size “s1” from the position where the protocol type is protocol A and the offset is “off1” is “n (50)”, and the offset “ Assume that a packet A having a code “n (30)” stored in an area of size “s1” of “off2” is received. Note that n (30) is a value satisfying n (10) <n (30) <n (50).

At this time, if the packet storage number Np of the reception buffer 31 exceeds the reference value Nth, the relay / discard determination process of FIG. 6 is executed, and the entry number E1 is extracted from the packet relay table T1 (step S21, Also, the protocol type of the received packet, in this case “protocol A”, is extracted (step S23).
At this time, the protocol type is extracted from the i-th or first entry of the packet relay table T1, and in this case, it is “protocol A”, which matches the protocol type of the received packet, so that steps S26 to S31 are performed. The process proceeds to S32, and the number of entries in the sublist table T2, in this case, E2 = 3 is extracted.

  The first entry t1 of the sublist table T2 has an offset “off1”, a size “s1”, a code “n (50)”, and a determination condition “=”, so that the received packet, that is, FIG. The code information of the area specified by the offset “off1” and the size “s1” of the packet A is extracted (step S35). In this case, since it is “n (50)”, the code of the packet A and the code of the entry t1 are the same and satisfy the determination condition “=”, the process proceeds from step S37 to step S38, and the variable j Is incremented by "1", and then it is determined whether or not the condition of entry t2 is satisfied.

  The entry t2 has an offset “off2”, a size “s1”, a code “n (10)”, and a determination condition “≦”, and the code of the area specified by the offset “off2” and the size “s1” of the packet A is Since “n (30)” and the code “n (30)” of the packet A satisfies the determination formula “n (10) ≦ n (30)”, the variable j is set through steps S37 to S38. It is incremented by “1”, and then it is determined whether or not the condition of entry t3 is satisfied.

  The entry t3 has an offset “off2”, a size “s1”, a code “n (50)”, and a determination condition “≧”, and is an area specified by the offset “off2” and the size “s1” of the packet A. Since the code is “n (30)” and the code “n (30)” of the packet A satisfies the determination formula “n (50) ≧ n (30)”, the variable is changed from step S37 to step S38. j is incremented by “1”. At this time, the variable j is j = 3, and j (= 3) <E2 (= 3) is not satisfied in the process of step S33. Since all three conditions of the table T2 are satisfied, it is determined that the packet A is relayed.

As a result, the numerical value of the information X consisting of numerical data set as a packet to be relayed preferentially is n (50), and the numerical value of the information Y consisting of numerical data is n (10) or more and n (50) The following packet A is relayed to the communication device 5 side.
Subsequently, for example, it is assumed that a packet B whose protocol type is protocol A as shown in FIG.
Also in this case, the sublist table T2 is derived from the head entry of the packet relay table T1, and the determination is performed according to the sublist table T2, but the packet B has an offset “off1”, a size “s1”, a code “N (50), offset“ off2 ”, size“ s1 ”, code“ n (60) ”, and code“ n (60) ”is n (50) <n (60).

For this reason, the conditions of the entries t1 and t2 in the sublist table T2 are satisfied, but in the entry t3, the offset “off2”, the size “s1”, the code “n (50)”, and the determination condition “≧” The code “n (60)” in the area specified by the offset “off2” and the size “s1” of the packet B does not satisfy the determination formula “n (50) ≧ n (60)”.
Therefore, the process proceeds from step S37 to step S27, and the variable i is incremented by “1”.

This time, the second entry registered in the packet relay table T1 and the packet B are compared with each other. However, since they do not match, the process proceeds from step S26 to step S27, and the variable i Is incremented, and the protocol type is compared for the next entry in the packet relay table T1.
Then, as a result of searching for all entries in the packet relay table T1, if there is no entry that matches the protocol type with the packet B, it is considered that the packet B does not satisfy the relay condition set in the condition table. The process proceeds from step S24 to step S28, and packet B is determined to be discarded.

Therefore, the numerical value of the information X consisting of numerical data set as a packet to be relayed preferentially is n (50), and the numerical value of the information Y consisting of numerical data is n (10) or more and n (50) or less. Packet B that does not satisfy this condition is discarded.
In this way, when the condition table is set and it is determined whether the packet is to be relayed in the information unit stored in a specific area of one packet, when determining based on the protocol type It is possible to determine whether to relay a packet in a more detailed unit by comparison.

Therefore, it is possible to select the packet to be relayed and the packet to be discarded in more detail, and it is possible to more accurately discard the low priority packet. Can be formed. Therefore, it is possible to more accurately avoid the loss of a high priority packet.
In the second embodiment, the case where the number of entries in the sublist table T2 is “3” has been described. However, the present invention is not limited to this, and an arbitrary number of conditions can be set. Similarly, the number of entries in the relay determination table T1 can be arbitrarily set, and can be configured to include entries in which the same protocol type is set according to conditions. It is possible to set a condition necessary for specifying.

  In the second embodiment, in the condition table, first, based on the packet relay table T1, it is determined whether or not the protocol type of the received packet matches the condition, and then another condition is determined based on the sublist table T2. Although the case where the conditions are also determined has been described, the present invention is not limited to this. The protocol type condition of the packet relay table T1 is incorporated in the sublist table, and one packet relay table T1 and one sublist table T2 are included. And can be arbitrarily set according to the condition for specifying the priority packet.

In each of the above-described embodiments, the case where the storage area 35 and the condition setting unit 36 are provided in each of the reception relay units 25a and 25b has been described. However, the present invention is not limited to this. An area 35 and condition setting means 36 may be provided, and the storage area 35 may be shared by the relay processing sections 33 of the reception relay sections 25a and 25b.
Further, in each of the above embodiments, the case where each communication device 5 is connected by the duplex line L has been described. However, the present invention is not limited to this, and may be connected by a single line. In this case, the processing function of the packet reception process may be incorporated in the communication device 5 and the communication device 5 may prevent a packet with a high priority from being lost by discarding a packet with a low priority. .

In each of the above-described embodiments, the relay device that relays between each communication device 5 and the duplex line L has been described. However, the present invention is not limited to this, and relay that relays packets between communication lines. It is also possible to apply to an apparatus. In short, any device that relays packets can be applied.
Note that the line receiving buffer 31 in FIG. 2 corresponds to the reception area, the buffer management unit 32 corresponds to the reception load detecting means, it corresponds to the condition setting means 36 Gapa packet condition setting means. Further, in the packet reception process of FIG. 3, the process proceeds from step S1 to step S6 to step S8, and the packet discarding process corresponds to the packet discarding means.

It is a schematic block diagram which shows an example of the network system to which this invention is applied. It is a block diagram which shows an example of the relay apparatus of FIG. 3 is a flowchart illustrating an example of a processing procedure of packet reception processing executed by a relay processing unit in FIG. 2. It is a flowchart which shows an example of the process sequence of the relay / discard determination process in 1st Embodiment. It is an example of a condition table. It is a flowchart which shows an example of the process sequence of the relay / discard determination process in 2nd Embodiment. It is an example of the packet for demonstrating operation | movement of this invention. It is an example of a condition table for explaining the operation of the present invention.

Explanation of symbols

5, 5a to 5c Communication device 10 Relay device 25a, 25b Reception relay unit 31 Line side reception buffer 32 Buffer management unit 33 Relay processing unit 34 Communication device side transmission buffer 35 Storage area 36 Condition setting means

Claims (2)

In a packet relay device in which a packet from a communication line is temporarily stored in a reception area, and then the packet in the reception area is relayed to a predetermined relay destination.
  A receiving load detecting means for detecting a receiving load of a packet from the communication line;
  A packet discarding unit for discarding the packet when the reception load of the packet detected by the reception load detection unit exceeds a preset load threshold and a packet excluding a preset specific packet is received;
  Packet condition setting means for setting a packet condition for specifying the specific packet,
  The packet includes a data body to be communicated and communication information necessary for communicating the data body via the communication line,
  The packet condition setting means sets a packet condition for each of the data body and the communication information,
  The packet discarding device, wherein the packet discarding unit determines that the packet is the specific packet when both the data body and the communication information satisfy the packet conditions. The data body includes numerical information,
  2. The packet relay apparatus according to claim 1, wherein the packet condition setting means sets a numerical range of the numerical information as the packet condition.

Images