JP2006279188A - Transmission controller conducting priority control, communication controller, communication system, communication network and transmission method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve the problem that a packet is not transmitted until the transmission of the packet already started is completed even when the packet having higer priority reaches later if the transmission processing of the packet is started once because a packet transmission started once is not interrupted until its completion in a conventional QoS control, both the transmission side and the receiving side are required to be compatible with the new system when the transmission interruption of the packet being transmitted and the restart of the transmission of the packet are applied and an equipment investment is increased. <P>SOLUTION: In a communication transferring the packets, the transmission of the first packet is interrupted and the second packet is transmitted preferentially when the priority of the succeeding second packet is higher than that of the first packet being transmitted. In the communication, all data contained in the first packet are transmitted after the completion of the transmission of the second packet. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a transmission control device, a communication control device, a communication system, a communication network, and a transmission method, and more particularly to a transmission control device, a communication control device, a communication system, a communication network, and a transmission method capable of priority control processing.

Usually, communication on the Internet is performed by transferring a variable-length frame called a packet using the Internet protocol (IP). The Internet adopts a method called “best effort”, and each IP device accepts all requested packets, but does not guarantee the time required to complete the transmission of each packet. It is a feature.
FIG. 2 is a schematic diagram showing an example of a conventional IP network device (only the transmission part is shown). The transmission portion of this apparatus includes a processing-waiting packet storage unit 101, a packet control unit 120, a packet transmission processing unit 106, and a transmitting packet buffer 104. In the packet control unit 120, a program (Tx Thread) for controlling the packet transmission processing unit 106 is operating.

  A transmission packet arriving at the IP network device is first stored in the processing-waiting packet storage unit 101. The packet transmission processing unit 106 sends the packet extracted from the processing-waiting packet storage unit 101 to the transmitting packet buffer 104. The packet is sent out from the packet buffer 104 during transmission.

  Furthermore, in recent years, as the demand for transferring real-time information such as voice and moving images has increased, some degree of QoS (Quality of Service, communication quality) control has been required for the Internet. QoS control is a general term for control that guarantees a communication speed on a network. In order to improve QoS, the Internet has proposed several schemes based on the best-effort scheme, not handling all packets fairly, and giving some priority to the packets. In this case, among the transmission waiting packets stored in the queue (queue), the packet with the highest priority is processed first. For example, the Diff-serv method (Non-Patent Document 1), the method of Patent Document 1, and the like can be mentioned. These technologies ensure a certain level of QoS.

However, in any of these conventional techniques, QoS control is performed before transmitting a packet, and packet transmission that has started once is not interrupted until completion. Therefore, once the packet transmission process starts, even if a packet with a higher priority is stored in the process-waiting packet storage unit 101, the priority is kept until the transmission of the packet currently being processed is completed. High-frequency packets must wait for transmission. This phenomenon is particularly noticeable when the line connected to the network device at the next stage has a narrow band.
In order to solve this problem, Patent Document 2 discloses that even when a packet is being transmitted, if a high-priority packet arrives, the current packet transmission is interrupted and the high-priority packet is sent first. And a technique for resuming the interrupted packet transmission after the transmission is completed. What is characteristic of the method described in Patent Document 2 is that when resuming the transmission of the interrupted packet, the transmission is resumed from the interrupted point instead of retransmitting all the data included in the packet. This means that only the part after the part is retransmitted. As a result, it is possible to prevent a decrease in network utilization efficiency associated with retransmission of the entire interrupted packet.

JP 2004-015505 A JP 2002-185516 A Nikkei Communication, “Communication / Network Terminology Handbook”, 1st edition, Nikkei Business Publications, March 23, 2001, p. 96

  An IP packet is composed of an IP header and IP data. The length of the packet (total of IP header and IP data) is stored in a 16-bit field called Total Length in the IP header. Therefore, when the technique described in Patent Document 2 is simply implemented on an existing IP, for a packet that is cut off during transmission, a discrepancy occurs between the actual length of the packet and the value described in Total Length. . For this reason, if the receiving side does not support this method, a length error occurs on the receiving device side, and eventually, the transmitted part of the packet whose transmission is interrupted is lost. In the technology described in Patent Document 2, in order to avoid this, normal IP is extended or another protocol different from IP is introduced so that an error does not occur even if packet transmission is interrupted in the middle. It is necessary to do. Moreover, both the transmitting device side and the receiving device side must support this method. Therefore, the method described in Patent Document 2 has low compatibility with existing network devices, and a large amount of money is required to introduce this method. There is a problem that capital investment is necessary. For this reason, there is a concern that the technology described in Patent Document 2 is not easily introduced in reality.

  In addition, the technique described in Patent Document 2 always interrupts the current packet transmission when a packet having a higher priority than the currently transmitted packet arrives, and always places the high priority packet first. Configured to send. However, depending on the total length of the currently transmitted packet, the length of the already transmitted portion, and the length of the remaining untransmitted portion, the currently transmitted packet may be sent as it is even if the priority is low. Subsequently, it may be desirable to start transmission of a high-priority packet after completion of the transmission from the viewpoint of the utilization efficiency of the entire communication system.

  The present invention has been made to solve such problems of the prior art. In other words, the present invention is configured so that there is no need to expand or change the IP itself in the communication in which the delay time of the high priority queue is suppressed and the control is more faithful to the priority. The receiving device side may use the conventional IP device as it is. That is, an object of the present invention is to provide a communication system that enables control faithful to priority. Another object of the present invention is to construct the communication system so that it can be introduced with a small capital investment, and to promote the communication system. Furthermore, according to the present invention, not only the priority information of each packet but also other parameters such as the length of the packet are used to determine whether or not to stop packet transmission comprehensively. This also aims to improve the utilization efficiency of the entire communication system.

  The present invention stops the transmission of the first packet by stopping the transmission of the first packet when the priority of the subsequent second packet is higher than the first packet currently being transmitted in the communication for transferring the packet. A packet is transmitted first, and after the transmission of the second packet is completed, all data included in the first packet is transmitted. The present invention is further characterized in that the packet comprises a variable length frame. The present invention is further characterized in that the packet transfer is performed in a best effort manner. The present invention can be preferably used in communication in which packets are transferred using IP. In the present invention, it is further preferable that the first packet is sent back to the queue.

  The present invention is further characterized in that the length of the remaining portion of the first packet is used as one of the criteria for determining whether or not to stop transmission of the first packet. In particular, when the length of the remaining portion of the first packet is shorter than the first threshold, it is desirable to transmit the first packet to the end. Alternatively, when the length of the remaining portion of the first packet is equal to or less than the first threshold, it is desirable to transmit the first packet to the end. The first threshold value is preferably changeable.

  As the priority, a Diff-serv type Diff-serv Code Point (DSCP) can be preferably used. Alternatively, as the priority, for the first packet, an evaluation expression using the DSCP and the length of the untransmitted portion of the packet as arguments, and for the second packet, the DSCP and the total length of the packet as arguments It is preferable to express each in the evaluation formulas. In this case, for example, the priority is a positive real number x, the evaluation formula for the first packet is (DSCP) / (length of untransmitted portion of packet) ^ x, and the second The evaluation formula for a packet can be expressed as (DSCP) ÷ (total length of packet) ^ x. For example, x can be preferably 1 or 2.

  Alternatively, Precedence in the IP header can be used as the priority. Alternatively, the priority is an evaluation expression that uses the precedence and the length of the untransmitted portion of the packet as arguments for the first packet, and the precedence and the total length of the packet for the second packet as arguments. Can be represented by the evaluation formulas. In this case, for example, the priority is a positive real number x, the evaluation formula for the first packet is (Precedence) / (length of untransmitted portion of packet) ^ x, and the second The evaluation formula for a packet can be represented by (Precedence) / (total length of packet) ^ x. For example, x can be preferably 1 or 2.

  It is preferable that the present invention further has a function of dividing the first packet after transmission of the first packet is stopped. As the number of divisions, 2 can be preferably used.

  The present invention is further characterized in that the entire length of the first packet is used as one of the criteria for determining whether or not to stop transmission of the first packet. In particular, when the overall length of the first packet is shorter than a second threshold, the first packet is independent of the priority of the first packet and the priority of the second packet. It is desirable to transmit to the end without stopping transmission. When the entire length of the first packet is equal to or smaller than a second threshold, transmission of the first packet is performed regardless of the priority of the first packet and the priority of the second packet. It is desirable to send to the end without stopping. The second threshold value may be changeable.

  Alternatively, the present invention is characterized in that the length of the already transmitted portion of the first packet is used as one of the criteria for determining whether or not to stop the transmission of the first packet. In particular, when the length of the already transmitted portion of the first packet is longer than a third threshold, the first packet can be transmitted to the end. The third threshold value is preferably changeable.

  Unlike the invention described in Patent Document 2, the present invention has an advantage that the transmitting side only needs to adopt the technology of the present invention, and the receiving side may be a conventional apparatus. For this reason, the capital investment for implementing the present invention is low, and the system of the present invention is expected to be easily introduced.

  Further, in the present invention, whether or not to cancel a packet that is currently being transmitted is determined not only by priority, but also by the total length of the currently transmitted packet, the length of the already transmitted portion, and the remaining untransmitted Judging comprehensively from the length of the part. Therefore, it is possible to improve the utilization efficiency of the entire communication system.

Embodiments of the present invention will be described with reference to the drawings.

Example 1
FIG. 1 is a schematic diagram showing an IP network device according to an embodiment of the present invention (only a transmission part is shown). The transmission part of this apparatus includes a packet determination unit 107 that compares packet lengths or priorities, a packet transmission processing unit 106 that performs packet transmission processing, a packet header information storage unit 108 that stores packet header information, A queue monitoring unit 111 that acquires a packet length, a packet control unit 120, a processing-waiting packet storage unit 101, a transmitting packet buffer 104, and a processing-waiting packet storage unit 101 from a transmitting packet buffer 104. And an enqueue processing unit 105 that sends it back to and a process waiting packet priority acquisition unit 110 that acquires the priority of the process waiting packet. In the packet control unit 120, a program (Tx Thread) that controls the packet determination unit 107, the packet transmission processing unit 106, the packet header information storage unit 108, and the queue monitoring unit 111 is operating.

  A transmission packet arriving at the IP network device of the present invention is first stored in the processing-waiting packet storage unit 101. The packet transmission processing unit 106 sends the packet taken out from the processing-waiting packet storage unit 101 by the packet determination unit 107 according to the priority of each packet to the transmitting packet buffer 104. The packet is sent out from the packet buffer 104 during transmission. The operation up to this point is the same as that of the conventional IP network apparatus shown in FIG. 2, but the IP network apparatus according to the first embodiment transmits a packet being transmitted in the transmitting packet buffer 104 as needed by an Enqueue. The difference is that it has a function of sending it back to the processing-waiting packet storage unit 101 via the processing unit 105. That is, in the IP network device of the present invention, when a packet with a high priority is stored in the processing-waiting packet storage unit 101 while a long packet with a low priority is being transmitted from the transmitting packet buffer 104, the Tx Thread Deletes the transmission process of a long packet with low priority during transmission and transmits the high priority packet first. The long packet with the low priority and the transmission process deleted is subjected to the transmission process again later.

  In the present invention, whether or not to enter these operations during the transmission of each packet is not only determined by the priority between both packets, but also by the overall length of the packet and the unsent packet being transmitted. Judgment is made after taking into account various parameters such as the length of the portion. In the first embodiment, while transmitting a packet in the transmitting packet buffer 104, a packet having a higher priority arrives at the processing-waiting packet storage unit 101, and the remaining unsent packets of the packet being transmitted When the length of the part is longer than a predetermined threshold A, the Tx Thread is configured to stop transmission. The threshold A is set when the IP network device is activated.

In the first embodiment, Diff-serv Code Point (DSCP) is adopted as an index indicating priority. Here, DSCP is a router that uses the Diff-serv method to identify the types of traffic including voice and video with various properties that flow on the Internet, and to perform forwarding processing according to each service. This is the code that determines the operation. In the Diff-serv method, the Type of Service (TOS) field in the IP packet is redefined as a DS (Diff-serv) field, and DSCP is described in 6 bits of the DS field. Therefore, the priority represented by DSCP is 2 6, that is, 64 levels.
FIG. 7 is a conceptual diagram showing where the DS field is located in the Diff-serve IP header. In FIG. 7, the Diff-serve IP header has a 4-bit version, a 4-bit header length, a 16-bit datagram length, a 16-bit identification number, and a 2-bit flag in order from the top. , 14-bit fragment offset, 8-bit Time to Live, 8-bit protocol number, 16-bit header checksum, 32-bit source IP address, 32-bit destination IP address, 32 And an optional area of bits. Thus, the DS field occupies a total of 16 bits from the 9th bit to the 16th bit from the top in the Diff-serve IP header.

  FIG. 8 is a conceptual diagram showing where the DSCP is located in the DS field. In FIG. 8, the DS field is composed of 6-bit DSCP and 2-bit unused area from the top. Thus, DSCP occupies a total of 6 bits from the first bit to the sixth bit in the DS field.

  FIG. 3 is a flowchart showing a specific processing procedure when the Tx Thread in the first embodiment performs priority control. In FIG. 3, the Tx Thread first stores packet information (IP header, payload, etc.) in the packet header information storage unit 108 (S303), and then instructs the transmitting packet buffer 104 to transmit a packet (S304). . Next, the Tx Thread checks whether or not the packet has been transmitted by monitoring a transmission port (not shown) using the transmitting packet buffer monitoring unit 114 (S305). When another packet is stored in the pending packet storage unit 101 during packet transmission (S306, S307), the Tx Thread uses the transmitting packet buffer monitoring unit 114 to transmit an untransmitted packet. It is determined whether the length of the remaining part is shorter than the threshold value A (S308). If the remainder is shorter than the threshold value A, the currently transmitted packet is sent as it is to the end (S311). If the remainder is longer than the threshold value A, the Tx Thread acquires the priority of the waiting packet using the waiting packet priority acquisition unit 110 (S309), and uses the packet determination unit 107 to wait for the waiting packet. And the priority of the packet being transmitted stored in the packet header information storage unit 108 are compared (S310). If the priority of the transmitting packet is higher than the priority of the waiting packet, the transmitting packet is sent to the end as it is (S311). When the priority of the packet being transmitted is lower than the priority of the waiting packet, the Tx Thread instructs the packet buffer 104 for transmission to stop transmitting the packet being transmitted using the packet buffer monitoring unit 114 for transmission. (S312). The already transmitted part is deleted at the transmission destination due to the invalid header (Length). In this case, the Tx Thread further enqueues the packet whose transmission has been canceled (S312) to the processing-waiting packet storage unit 101 via the Enqueue processing unit 105 (S314), and stores it in the packet header information storage unit 108. The information (IP header, payload, etc.) of the packet that has been canceled is deleted (S315), the waiting packet transmission process is started (S316), and the process returns to the first procedure (S317).

  In the first embodiment, when transmission of a packet is stopped, the packet is sent back from the transmitting packet buffer 104 to the processing-waiting packet storage unit 101 via the Enqueue processing unit 105. This eliminates the need for dedicated hardware for storing a packet whose transmission has been canceled. Further, the packet stored again in the processing-waiting packet storage unit 101 is again determined for priority based on the same criteria as other packets, and is eventually extracted by Tx Thread. For this reason, there is no need to treat the packet for which transmission has been stopped, and the Tx Thread configuration is not particularly complicated.

Example 2
As a second embodiment of the present invention, another IP network device is cited. The schematic diagram of the transmission part of the IP network device according to the second embodiment is shown in FIG. 1 as in the first embodiment. However, the processing of the program Tx_Thread operating in the packet control unit 120 is different, so the operation of the packet control unit 120 is Different. FIG. 4 is a flowchart showing a specific procedure for priority control of the IP network device according to the second embodiment. In FIG. 4, steps for performing the same processing as in FIG. 3 (Embodiment 1) are given the same numbers as in FIG.

  In the second embodiment, after the Tx Thread stops transmitting the packet being transmitted (S312), the packet whose transmission is stopped is divided into two and a header is added to each packet (S413). When a header is added to a packet divided into two, the datagram length (FIG. 7) is appropriately set so that a Length error does not occur on the receiving side. The Tx Thread further enqueues the divided packets to the processing-waiting packet storage unit 101 (S414). Steps before S312 and after S315 are the same as in the first embodiment.

  In the priority control procedure of the first embodiment (FIG. 3), the low-priority packet whose transmission has been stopped (S312) is enqueued in the processing-waiting packet storage unit 101 as it is (S314). Therefore, even if transmission of the low priority packet is started again, if packets with high priority arrive one after another, the low priority packet is enqueued again in the processing-waiting packet storage unit 101 and transmitted without limit. There is a fear of having to wait. In the priority control procedure according to the second embodiment (FIG. 4), in order to solve this problem, the Tx Thread uses the Enqueue processing unit 105 to divide a packet whose transmission has been stopped into two and add a header to each of the packets. (S413). When a header is added to a packet divided into two, the datagram length (FIG. 7) is appropriately set so that a Length error does not occur on the receiving side. The Tx Thread further enqueues each divided packet to the processing-waiting packet storage unit 101 (S414). As a result, the low-priority packet is divided into two (S413) each time transmission is stopped and enqueued to the processing-waiting packet storage unit 101 (S414), so that the length of each fragment eventually falls below the threshold A, Even if a high-priority packet arrives during transmission, transmission will not be interrupted (S308, S311). Therefore, in the second embodiment, it is possible to prevent the low priority packet from waiting indefinitely.

Example 3
As a third embodiment of the present invention, another IP network device is cited. The schematic diagram of the transmission part of the IP network device of the third embodiment is shown in FIG. 1 as in the first embodiment. However, the processing of the program Tx_Thread operating in the packet control unit 120 is different, so Different. . A specific procedure for priority control of the IP network device according to the third embodiment is shown in FIG. In FIG. 5, steps that perform the same processing as in FIG. 3 (Example 1) are denoted by the same numbers as in FIG. 3.

  In the third embodiment, at the start of transmission processing, the Tx Thread checks the packet length of the packet to be transmitted using the packet determination unit 107 (S501). Otherwise, normal transmission processing is performed (S502). S303 and subsequent steps are the same as those in the second embodiment. The threshold B is set when the IP network device is activated. In the third embodiment, this priority control procedure (FIG. 5) prevents a packet having a length of a certain length or less from entering S303 and later regardless of the priority.

  In the third embodiment, when the packet length is short to some extent, even if the priority of the packet is low, the influence on the waiting time of the subsequent high-priority packet is small. Is preferable from the viewpoint of the utilization efficiency of the entire network.

  Furthermore, in the priority control procedure (FIG. 5) of the third embodiment, in addition to the priority control procedure (FIG. 4) of the second embodiment, Tx Thread cancels transmission using the Enqueue processing unit 105. The packet is divided into two and a header is added to each of them (S413). When a header is added to a packet divided into two, the datagram length (FIG. 7) is appropriately set so that a Length error does not occur on the receiving side. The Tx Thread further enqueues each divided packet to the processing-waiting packet storage unit 101 (S414). As a result, a packet whose transmission has been stopped many times and divided into two (S413) each time will eventually fall below the threshold value B and will not enter S303 and thereafter. Therefore, transmission is not interrupted regardless of whether or not a high priority packet has arrived (S310) and whether or not there is a length of a portion that has not been transmitted yet (S308). According to such control, in the third embodiment, it is possible to more reliably prevent the low priority packet from waiting for transmission for a longer time than in the second embodiment.

Example 4
As a fourth embodiment of the present invention, another IP network device is cited. A schematic diagram of the transmission part of the IP network device of the fourth embodiment is shown in FIG. The specific procedure for priority control of the IP network device of the fourth embodiment is shown in FIG. 3 as in the first embodiment. However, in the fourth embodiment, since the processing of the program Tx_Thread operating in the packet control unit 120 is different from that in the first embodiment, the operation of the packet control unit 120 is different from that in the first embodiment. That is, in the fourth embodiment, (DSCP) ÷ (packet length) for untransmitted packets, and (DSCP) ÷ (length of remaining untransmitted portions) for packets being transmitted, respectively. The difference from the first embodiment is that the priority is adopted.

  Thereby, in the priority control procedure of the fourth embodiment, even if the DSCP value is large, the packet having a long packet has a low priority. Conversely, even if the DSCP value is small, a packet with a short packet has a higher priority. In this way, by using an index that considers not only the DSCP but also the packet length as the priority, more detailed and flexible priority control can be performed, and the network can be operated efficiently. Can do.

  In the fourth embodiment, the embodiment in which the method for obtaining the priority is changed based on the first embodiment has been described. However, the method for obtaining the priority is changed based on the second or third embodiment. Forms are also possible.

Example 5
As a fifth embodiment of the present invention, another IP network device is cited. The schematic diagram of the transmission part of the IP network device of the fifth embodiment is shown in FIG. 1 as in the first embodiment. However, the processing of the program Tx_Thread operating in the packet control unit 120 is different, so Different. . FIG. 6 is a flowchart showing a specific procedure for priority control of the IP network device according to the fifth embodiment.

  The priority control procedure of the fifth embodiment is basically the same as that of the first embodiment, except for the following points. In other words, when another packet is stored in the processing-waiting packet storage unit 101 during packet transmission, the Tx Thread uses the transmitting packet buffer monitoring unit 114 to determine the length of the transmitted packet. It is determined whether or not it is longer than the threshold value C (S508). If the already transmitted part is longer than the threshold C, the remaining untransmitted part is transmitted to the end (S311). If the already transmitted part is shorter than the threshold C, the Tx Thread acquires the priority of the waiting packet (S309), and compares the priority of the packet being transmitted with the priority of the waiting packet (S310). . The threshold C is set when the IP network device is activated.

  When another packet is stored in the processing-waiting packet storage unit 101 during packet transmission, in the IP network device according to the first to fourth embodiments, the remaining untransmitted portion of the packet being transmitted is below a certain level. In the case of length, the rest is sent to the end (S311). Therefore, in the case of a packet having a long overall length, even if a considerable length portion has already been transmitted, if the remaining untransmitted portion is longer than a certain length, the transmission is canceled ( S308), the already transmitted part is lost. On the other hand, in the IP network device of the fifth embodiment, when the already transmitted portion of the packet being transmitted is longer than a certain length, the packet is transmitted to the end regardless of the length of the remaining untransmitted portion. It is configured to send out. Thereby, while suppressing a decrease in the network utilization efficiency due to the suspension / retransmission of packet transmission, the priority control of packets faithful to the priority as described in the first to fourth embodiments is also achieved, We are trying to achieve both.

  In the embodiments described so far, the priority control procedure based on DSCP (and the combination of DSCP and packet length) in the Diff-serv method has been described, but the application of the present invention is not limited to this. As the priority in the present invention, for example, a priority based on a known precedence or the like may be considered. Further, as an evaluation formula for determining the priority, for example, (DSCP) ÷ (packet length) ^ 2 can be considered in addition to the formula in the fourth embodiment. In this case, as compared with the fourth embodiment, when the length of the packet is increased, the priority is rapidly decreased, which is more advantageous for a short packet. Which evaluation formula is adopted is determined based on a network design policy such as what data is to be given higher priority. Further, in the second and third embodiments, the packet whose transmission is stopped is divided into two (S413 in FIG. 4), but this does not necessarily require the division number to be two. For example, two or more arbitrary division numbers such as three divisions and four divisions are conceivable. In the above embodiment, the implementation on the IP network of the present invention has been described. However, the implementation of the present invention is not limited to the IP protocol. The present invention can be implemented on any protocol that transfers variable-length frames in a best-effort manner.

It is the schematic which shows the IP network apparatus which is one Example of this invention (Example 1 thru | or Example 5, only a transmission part is shown). It is the schematic which shows an example of the conventional IP network apparatus (only a transmission part is shown). It is a flowchart which shows the specific procedure of the priority control in the IP network apparatus of the 1st and 4th Example of this invention. It is a flowchart which shows the specific procedure of the priority control in the IP network apparatus of 2nd Example of this invention. It is a flowchart which shows the specific procedure of the priority control in the IP network apparatus of the 3rd Example of this invention. It is a flowchart which shows the specific procedure of the priority control in the IP network apparatus of the 5th Example of this invention. It is a conceptual diagram which shows the position of DS field in IP header (Diff-serve system). It is a conceptual diagram which shows the position of DSCP within DS field.

Explanation of symbols

101 Processing packet storage (Tx Queue)
104 Transmitting packet buffer (Tx Buffer)
105 Enqueue Processing Unit 106 Packet Transmission Processing Unit 107 Packet Determination Unit 108 Packet Header Information Saving Unit 110 Processing Waiting Packet Priority Acquisition Unit 111 Queue Monitoring Unit 114 Transmitting Packet Buffer Monitoring Unit 120 Packet Control Unit

Claims (39)

A transmission control device for transferring packets,
A function of comparing the priority of the first packet currently being transmitted with the priority of the subsequent second packet;
A function of stopping transmission of the first packet and transmitting the second packet first when the priority of the second packet is higher than that of the first packet;
A function of transmitting all the data included in the first packet after the transmission of the second packet is completed;
A transmission control apparatus comprising: The transmission control apparatus according to claim 1, wherein the packet includes a variable length frame. 3. The transmission control apparatus according to claim 1, wherein the packet is transferred by a best effort method. The transmission control apparatus according to any one of claims 1 to 3,
The transmission control apparatus, wherein the communication is to transfer a packet using an Internet protocol (IP). The transmission control apparatus according to any one of claims 1 to 4,
A transmission control apparatus comprising: a processing unit that sends back the first packet to a queue in which packets arriving at the transmission control apparatus are stored. The transmission control apparatus according to any one of claims 1 to 6, wherein one of the criteria for determining whether or not to stop transmission of the first packet is the remaining portion of the first packet. A transmission control apparatus using the length of The transmission control apparatus according to any one of claims 1 to 6,
A function of comparing the length of the remaining portion of the first packet with a first threshold;
A function of transmitting the first packet to the end when the length of the remaining portion of the first packet is shorter than the first threshold;
A transmission control apparatus comprising: The transmission control apparatus according to any one of claims 1 to 6,
A function of comparing the length of the remaining portion of the first packet with a first threshold;
A function of transmitting the first packet to the end when the length of the remaining portion of the first packet is equal to or less than the first threshold;
A transmission control apparatus comprising: 9. The transmission control apparatus according to claim 7 or 8, wherein the first threshold value can be changed. The transmission control apparatus according to any one of claims 4 to 9, wherein the priority is a Diff-serv Code Diff-serv Code Point (DSCP). The transmission control apparatus according to any one of claims 4 to 9, wherein the priority is
For the first packet, an evaluation formula that uses the DSCP and the length of the untransmitted portion of the packet as arguments,
For the second packet, an evaluation expression with the DSCP and the total length of the packet as arguments,
A transmission control apparatus characterized by being represented respectively. The transmission control apparatus according to claim 11, wherein the priority is a positive real number x,
The evaluation formula for the first packet is (DSCP) ÷ (length of untransmitted portion of packet) ^ x,
The evaluation formula for the second packet is (DSCP) ÷ (total length of packet) ^ x,
A transmission control apparatus characterized by being represented respectively. 13. The transmission control apparatus according to claim 12, wherein x is 1 or 2. The transmission control apparatus according to any one of claims 4 to 9, wherein the priority is a precedence in an IP header. The transmission control apparatus according to any one of claims 4 to 9, wherein the priority is
With respect to the first packet, an evaluation expression that takes Precedence and the length of the untransmitted portion of the packet as arguments,
For the second packet, it is an evaluation expression that takes Precedence and the total length of the packet as arguments,
A transmission control apparatus characterized by being represented respectively. The transmission control apparatus according to claim 15, wherein the priority is a positive real number x,
The evaluation formula for the first packet is (Precedence) ÷ (length of untransmitted portion of packet) ^ x,
The evaluation formula for the second packet is (Precedence) ÷ (total length of packet) ^ x,
A transmission control apparatus characterized by being represented respectively. 17. The transmission control apparatus according to claim 16, wherein x is 1 or 2. The transmission control apparatus according to any one of claims 1 to 17,
A transmission control apparatus having a function of dividing the first packet after transmission of the first packet is stopped. The transmission control apparatus according to claim 18, wherein the number of divisions is two. The transmission control apparatus according to any one of claims 1 to 19, wherein one of the criteria for determining whether or not to stop transmission of the first packet is the total length of the first packet. A transmission control apparatus characterized by using the above. The transmission control apparatus according to any one of claims 1 to 19,
A function of comparing the overall length of the first packet with a second threshold;
If the overall length of the first packet is shorter than the second threshold, the first packet's priority is independent of the priority of the first packet and the priority of the second packet. A function of transmitting the first packet to the end without stopping transmission;
A transmission control apparatus comprising: The transmission control apparatus according to any one of claims 1 to 19,
A function of comparing the overall length of the first packet with a second threshold;
When the entire length of the first packet is equal to or smaller than the second threshold, the transmission of the first packet is performed regardless of the priority of the first packet and the priority of the second packet. A function of transmitting the first packet to the end without stopping
A transmission control apparatus comprising: The transmission control apparatus according to claim 21 or 22, wherein the second threshold value can be changed. The transmission control apparatus according to any one of claims 1 to 23, wherein one of the criteria for determining whether or not to stop transmission of the first packet is the transmission completion of the first packet. A transmission control apparatus using a length of a portion. The transmission control apparatus according to any one of claims 1 to 23, wherein:
A function of comparing the length of the transmitted portion of the first packet with a third threshold;
A function of transmitting the first packet to the end when the length of the transmitted portion of the first packet is longer than the third threshold;
A transmission control apparatus comprising: 26. The transmission control apparatus according to claim 25, wherein the third threshold value is changeable. A communication control apparatus including at least the transmission control apparatus according to any one of claims 1 to 26. A communication system including at least the transmission control device according to any one of claims 1 to 26 or the communication control device according to claim 27. A communication network including at least the transmission control device according to any one of claims 1 to 26 or the communication control device according to claim 27. A transmission method for forwarding packets,
Comparing the priority of the first packet currently being transmitted with the priority of the subsequent second packet;
Stopping the transmission of the first packet and transmitting the second packet first when the priority of the second packet is higher than that of the first packet;
Transmitting all data included in the first packet after the transmission of the second packet is completed;
A transmission method characterized by comprising: The transmission method according to claim 30, wherein the packet is transferred using an Internet protocol (IP). The transmission method according to claim 30 or claim 31, wherein
A transmission method comprising the step of sending back the first packet to a queue in which packets arriving at the transmission control apparatus are stored. The transmission method according to any one of claims 30 to 32, wherein one of the criteria for determining whether or not to stop transmission of the first packet is the remaining part of the first packet. A transmission method characterized by using a length. The transmission method according to any one of claims 30 to 32, wherein:
Comparing the length of the remaining portion of the first packet with a first threshold;
If the length of the remaining portion of the first packet is shorter than the first threshold, transmitting the first packet to the end;
A transmission method characterized by comprising: The transmission method according to any one of claims 30 to 34, wherein:
A transmission method comprising the step of dividing the first packet after transmission of the first packet is stopped. 36. The transmission method according to claim 30, wherein one of the criteria for determining whether or not to stop transmission of the first packet is the total length of the first packet. The transmission method characterized by using. 36. The transmission method according to any one of claims 30 to 35, wherein:
Comparing the overall length of the first packet with a second threshold;
If the overall length of the first packet is shorter than the second threshold, the first packet's priority is independent of the priority of the first packet and the priority of the second packet. Transmitting the first packet to the end without stopping transmission;
A transmission method characterized by comprising: 38. The transmission method according to any one of claims 30 to 37, wherein one of the criteria for determining whether or not to stop transmission of the first packet is a transmitted part of the first packet. The transmission method characterized by using the length of. The transmission method according to any one of claims 30 to 37, wherein:
Comparing the length of the transmitted portion of the first packet with a third threshold;
If the length of the transmitted portion of the first packet is longer than the third threshold, transmitting the first packet to the end;
A transmission method characterized by comprising:

Images