CN1219386C - Transmission Delay Scheduling Method Using TTL and Service Type Fields in Packet Header - Google Patents

Abstract

A joint scheduling method using TTL and service type fields in the packet header to realize multi-service packet transmission delay guarantee. The method realizes the end-to-end delay guarantee of packets of different business types during the transmission process of the communication network, uses the TTL field of the packet in the existing communication network, calculates the time T1 needed for the packet to be sent from the current moment to the destination, and uses T1 and TTL The difference DT is the forwarding weight, which is combined with the service type field to perform forwarding scheduling of packets in this node, so as to accelerate the forwarding efficiency of packets in this node and improve service quality. In this way, the validity and stability of packet transmission in the network are guaranteed. While ensuring the transmission delay of different types of packets, the network load is minimized.

Description

Transmission Delay Scheduling Method Using TTL and Service Type Fields in Packet Header

technical field

The invention relates to the technical field of joint scheduling for guaranteeing packet transmission delay guarantee. In particular, it relates to a joint scheduling method that utilizes the TTL and service type fields in the packet header to realize multi-service packet transmission delay guarantee, especially a method that utilizes the remaining lifetime (TTL) and service type fields in the packet header to realize different service types A joint scheduling and routing method for end-to-end delay guarantee during packet transmission in a communication network.

Background technique

As we know, the scale of IP network is expanding day by day, and the applications supported are also increasing day by day. IP network has been widely recognized as a convergence network of various services. But real-time applications such as VoIP, video conferencing, distance learning, and telemedicine, which bring bright prospects to IP networks, pose serious challenges to the loose structure, simple service and control mechanism of IP networks. How to meet the transmission delay of different types of packets on the IP network, so as to provide corresponding service quality assurance for real-time applications, has become an urgent problem that the computer Internet faces.

As early as the 1980s, the research work on real-time communication has already begun, and the problem of delay guarantee has been studied on various network structures, including CSMA/CD, token ring, FDDI, ATM, MPLS and point-to-point network. The connection of a real-time application is the relationship between the application and the network. On the one hand, the real-time application needs to provide service flow characteristic information to the network, and on the other hand, the network needs to provide the real-time application with the required service quality. Scalable connection access control is the key technology to realize connection-oriented real-time communication.

Contents of the invention

In view of the insufficiency of existing technologies in ensuring the real-time performance of different types of packet transmission, the main purpose of the present invention is to provide a joint scheduling method that utilizes the TTL and service type fields in the packet header to realize multi-service packet transmission delay guarantee, Especially the joint scheduling method that guarantees the real-time transmission of different types of packets in the network. The method can provide end-to-end quality of service guarantees for different types of packets together with the above-mentioned other methods, especially real-time guarantees. This method can also reduce the network load to a certain extent.

The present invention is achieved through the following technical solutions:

A joint scheduling method that uses the TTL and service type fields in the packet header to realize the transmission delay guarantee of multi-service packets, using the remaining lifetime TTL and service type fields of the packets transmitted in the network, and according to the service type to which the packets belong, the The difference between the TTL value and the time required for transmission from the current node to the destination DT value is used as the weight of packet transmission scheduling to allocate network resources and select the best packet transmission path, and based on this, it performs priority forwarding to ensure different types of packet transmission. The end-to-end delay, that is, the length of time from when the sender sends a letter to when the receiver receives the letter, reduces the network load.

If the DT value obtained after the calculation is negative, it indicates that the transmission delay of the packet cannot meet the requirements. At this time, the current node should delete the packet from the node, which means that the packet is removed from the network and sent to the sender. The source of the letter responds with an error notification message packet, telling the source that the packet sent by the source cannot be sent to the sink correctly through the node; this can also effectively reduce the load on the network; otherwise, it indicates that the packet can still be sent to the sink according to the required delay; this , the current node will hand over the packet to the forwarding scheduler and the routing control unit for processing according to the service type it belongs to.

When the packet is handed over to the routing control unit for routing selection processing, the routing control unit will select a path from the routing table that can meet the TTL transmission delay requirements of this type of packet according to the size of the DT value, and then transfer the packet to the corresponding The corresponding service type buffer queue of the output port is forwarded.

After the transmission path of the packet is selected, the packet will be scheduled and managed by the forwarding scheduling manager. At this time, the forwarding scheduler will send the packet to the corresponding service of the corresponding output port according to the size of the DT value and the service type to which the packet belongs. The priority queue of the type is waiting to be sent; at this time, the smaller the DT value, the higher the forwarding priority of the packet.

The source sends various packets to the network. After the current node receives the packets, it will process them as follows:

1) According to the destination address and routing table of the packet sending, calculate the minimum time required to send the packet from the node to the destination. This time is recorded as T1, and the TTL value is subtracted from T1 to obtain the new value DT, and then the following processing is taken:

A) If DT is less than 0, it indicates that the transmission delay of the packet cannot meet the requirements. At this time, the current node should delete the packet from the node (that is, remove the packet from the network), and respond to the source with an error notification message packet , telling the source that the packet sent cannot be sent to the sink correctly through this node. This can also effectively reduce the load on the network.

B) If the DT of the packet is greater than or equal to 0, it indicates that the packet can still be sent to the destination with the required time delay. At this time, the intermediate node will hand over the packet to the forwarding scheduler and routing control component shown in Figure 1 for processing.

2) When the packet is handed over to the routing control unit for routing selection processing, the routing control unit will select a transmission path that can meet the packet transmission delay requirement according to the size of the DT value and the routing table.

3) After the transmission path of the packet is selected, the packet will be scheduled and managed by the forwarding scheduling manager. At this time, the forwarding scheduler will send the packet to the corresponding service type of the corresponding output port according to the size of the DT value and the service type to which the packet belongs The corresponding priority queue is waiting to be sent. At this time, the smaller the DT value, the higher the forwarding priority of the packet, so that the remaining lifetime and service type fields can provide guarantees for the delay requirements of various industries.

Description of drawings

Figure 1 is a node structure diagram for implementing TTL and service type associated routing and scheduling;

Figure 1 shows the node system structure that can meet the above requirements. The functions of the main parts in the figure have been described above.

The above-mentioned invention can be implemented independently to ensure the end-to-end delay of different types of packets transmitted in the network, and can also be integrated with other existing technologies to jointly ensure the real-time guarantee of network transmission of different types of packets.

Any equivalent transformation based on the idea of the present invention shall belong to the protection scope of the present invention.

Claims (4)

1. A joint scheduling method utilizing TTL and service type field in the packet header to realize multi-service packet transmission delay guarantee, characterized in that, utilizing the remaining lifetime TTL and service type field of the packet header, according to the service type to which the packet belongs , use the difference DT between the TTL value and the time required for transmission from the current node to the destination as the weight of packet transmission scheduling to allocate network resources and select the best packet transmission path, and perform priority forwarding based on this, thus ensuring different types of The end-to-end delay of packet transmission, that is, the length of time from when the sender sends a letter to when the receiver receives the letter, reduces the network load. 2. A kind of joint scheduling method utilizing the TTL and service type field in the packet header to realize multi-service packet transmission delay guarantee according to claim 1, it is characterized in that, if the DT value obtained after calculation is negative, it indicates The transmission delay of the packet can no longer meet the requirements. At this time, the current node should delete the packet from the node, which means that the packet is removed from the network, and respond to the source of the letter with an error notification message packet, telling The packet sent by the source cannot be sent to the sink correctly through the node; otherwise, it indicates that the packet can still be sent to the sink according to the required delay; at this time, the current node will hand over the packet to the forwarding scheduler and route selection according to the type of service it belongs to The control unit is processed. 3. a kind of joint scheduling method that utilizes TTL and service type field in the packet header to realize multi-service packet transmission delay guarantee according to claim 1, it is characterized in that, when the packet is handed over to the routing control component to carry out routing selection process , the routing control component will select a path from the routing table that can meet the TTL transmission delay requirement of this type of packet according to the size of the DT value, and then transfer the packet to the corresponding service type buffer queue of the corresponding output port for forwarding. 4. A kind of joint scheduling method utilizing the TTL and service type field in the packet header to realize multi-service packet transmission delay guarantee according to claim 1, characterized in that, after the transmission path of the packet is selected, the packet will be forwarded by The scheduling manager performs scheduling management. At this time, the forwarding scheduler will send the packet to the priority queue of the corresponding service type corresponding to the output port according to the size of the DT value and the service type to which the packet belongs to wait to be sent; at this time, the higher the DT value Smaller, the higher the forwarding priority of the packet.