WO2021180073A1

WO2021180073A1 - Packet transmission method and device, network node, and storage medium

Info

Publication number: WO2021180073A1
Application number: PCT/CN2021/079756
Authority: WO
Inventors: 杜宗鹏; 耿亮
Original assignee: 中国移动通信有限公司研究院; 中国移动通信集团有限公司
Priority date: 2020-03-09
Filing date: 2021-03-09
Publication date: 2021-09-16
Also published as: CN113382442A; CN113382442B

Abstract

The present application discloses a packet transmission method and device, a network node, and a storage medium. The method comprises: a first network node receiving a first packet; acquiring a first identifier from the first packet, the first identifier being used to indicate that the first packet has a delay-sensitive requirement; upon acquiring the first identifier, configuring the first packet in a specific queue, the specific queue at least being used to buffer delay-sensitive packets to be sent; determining a packet outgoing rate of the specific queue by using a packet incoming rate of the specific queue; and shaping the specific queue on the basis of the determined outgoing rate, and sending the first packet, wherein the first network node is a forwarding node in a network.

Description

Message transmission method, device, network node and storage medium

Cross-references to related applications

This application is filed based on a Chinese patent application with an application number of 202010157089.8 and an application date of March 9, 2020, and claims the priority of the Chinese patent application. The entire content of the Chinese patent application is hereby incorporated into this application by reference.

Technical field

This application relates to the field of Internet Protocol (IP, Internet Protocol) networks, and in particular to a method, device, network node, and storage medium for message transmission.

Background technique

At present, in IP networks, messages are forwarded according to the basic idea of Best Effort (BE, Best Effort), so it is difficult to guarantee deterministic delay. Therefore, some fifth-generation mobile communication technologies (5G) are extremely reliable and reliable. In low-latency communications (URLLC, Ultra-Reliable and Low Latency Communications) scenarios, the deterministic latency requirements of services cannot be met.

Summary of the invention

In order to solve related technical problems, embodiments of the present application provide a message transmission method, device, network node, and storage medium.

The technical solutions of the embodiments of the present application are implemented as follows:

The embodiment of the present application provides a message transmission method, which is applied to a first network node, and includes:

Receive the first message;

Obtain the first identifier from the first message; the first identifier represents the delay-sensitive demand of the first newspaper;

In the case of obtaining the first identifier, set the first message in a specific queue; the specific queue is at least used for buffering the delay-sensitive messages to be sent;

Using the incoming rate of packets in the specific queue to determine the outgoing rate of packets in the specific queue;

Based on the determined output rate, the specific queue is shaped, and the first packet is sent; wherein,

The first network node is a network forwarding node.

In the above solution, the first identifier indicates that the first message has an exclusive sending priority.

In the above solution, the obtaining the first identifier from the first message includes:

Obtain a list of segment identities (SIDs) from the first message; the SID list contains multiple SIDs corresponding to the traffic engineering path;

When it is determined that the current SID indicates a specific queue on the next-hop network node corresponding to the first message, the first message is set in the specific queue.

Obtain the prefix SID from the first message;

Look up the outbound interface corresponding to the obtained prefix SID in the routing and forwarding table;

In a case where the found outgoing interface corresponds to the specific queue, the first packet is set in the specific queue.

In the above solution, the using the incoming rate of packets in the specific queue to determine the outgoing rate of packets in the specific queue includes:

The ingress rate is combined with the queue depth to determine the outbound rate of packets in the specific queue.

In the above solution, the use of the incoming rate in combination with the queue depth to determine the outgoing rate of packets in the specific queue includes:

When the queue depth is less than the threshold, determine that the out rate is a first rate; the first rate is less than the in rate, and the difference between the in rate and the first rate is less than the first value;

or,

When the queue depth is equal to the threshold, it is determined that the out rate is the second rate; the second rate is equal to the in rate;

or,

When the incoming rate is zero, it is determined that the outgoing rate is the third rate; the third rate is the preset rate or the last recorded out rate.

The embodiment of the present application also provides a message transmission method, including:

The second network node obtains the first message; and determines that the service corresponding to the first message is a delay-sensitive service;

The second network node sets a first identifier for the first message; the first identifier represents the delay-sensitive demand of the first message;

The second network node sets the first message with the first identifier in a specific queue for shaping, and then sends the first message with the first identifier;

The first network node receives the first message, and obtains the first identifier from the received first message;

When the first network node obtains the first identifier, set the received first message in a specific queue; the specific queue is at least used to buffer the delay-sensitive messages to be sent;

The first network node uses the incoming rate of the specific queue message to determine the outgoing rate of the specific queue message; and based on the determined outgoing rate, the specific queue is shaped, and the received first message is sent ;in,

The second network node is a network edge node; the first network node is a network forwarding node.

An embodiment of the present application also provides a message transmission device, which is set on a first network node, and includes:

The receiving unit is configured to receive the first message;

The first acquiring unit is configured to acquire a first identifier from a first message; the first identifier represents the delay-sensitive requirement of the first message;

The first processing unit is configured to set the first message in a specific queue when the first identifier is obtained; the specific queue is at least used for buffering the delay-sensitive messages to be sent; The incoming rate of messages in the specific queue is determined, and the outgoing rate of messages in the specific queue is determined; and based on the determined outgoing rate, the specific queue is shaped to send out the first message; wherein,

The first network node is a network forwarding node.

The embodiment of the present application also provides a network node, including: a first communication interface and a first processor; wherein,

The first communication interface is configured to receive a first message;

The first processor is configured to obtain a first identifier from a first message; the first identifier represents the delay-sensitive requirement of the first newspaper; when the first identifier is obtained , Setting the first message in a specific queue; the specific queue is at least used for buffering delay-sensitive messages to be sent; and using the incoming rate of the specific queue messages to determine the Out rate; and based on the determined out rate, the specific queue is shaped, and the first message is sent through the first communication interface; wherein,

The network node is a network forwarding node.

An embodiment of the present application also provides a network node, including: a first processor and a first memory configured to store a computer program that can run on the processor,

Wherein, the first processor is configured to execute the steps of any method on the side of the first network node when running the computer program.

The embodiment of the present application also provides a storage medium on which a computer program is stored, and when the computer program is executed by a processor, the steps of any method on the first network node side described above are implemented.

According to the message transmission method, device, network node, and storage medium provided in the embodiments of the present application, the first network node receives the first message; obtains the first identifier from the first message; the first identifier represents the first Sending stationery to delay-sensitive requirements; in the case of obtaining the first identifier, set the first message in a specific queue; the specific queue is at least used for buffering delay-sensitive messages to be sent; Using the incoming rate of messages in the specific queue to determine the outgoing rate of messages in the specific queue; based on the determined outgoing rate, shaping the specific queue to send out the first message; wherein, the first message A network node is a network forwarding node. On the network forwarding node, each low-latency service flow is not recognized. It is only identified as a flow with low-latency requirements based on the characteristics of the packet as a whole, and packets with low-latency requirements are Set in a specific queue for shaping, try to ensure that there is a certain queue depth in the specific queue, instead of using the BE forwarding mechanism as soon as possible. In this way, the network forwarding node can ensure that low-latency traffic is sent in an orderly manner on demand, and the network forwarding node's The processing minimizes the packet loss and buffering delay caused by the micro-burst of packets in the network, and can meet the delay requirements of the business.

Description of the drawings

Figure 1 is a schematic diagram of a deterministic network architecture;

Figure 2 is a schematic diagram of a micro burst of an IP device;

Figure 3 is a schematic diagram of an IP network architecture;

FIG. 4 is a schematic flowchart of a method for message transmission on the side of a second network node in an embodiment of the present application;

5a and 5b are schematic diagrams of adjacent SID formats according to an embodiment of this application;

Fig. 6 is a schematic diagram of a prefix SID format according to an embodiment of the application;

FIG. 7 is a schematic flowchart of a method for message transmission on the side of a first network node in an application embodiment;

FIG. 8 is a flowchart of a method for message transmission according to an application embodiment;

FIG. 9 is a schematic diagram of the corresponding relationship between the inbound interface and the outbound interface of a network forwarding node according to an application embodiment of this application;

FIG. 10 is a schematic diagram of the relationship between a physical interface and a virtual interface according to an embodiment of the application;

FIG. 11 is a schematic structural diagram of a message transmission device according to an embodiment of the application;

FIG. 12 is a schematic structural diagram of another message transmission device according to an embodiment of the application;

FIG. 13 is a schematic diagram of a network node structure according to an embodiment of this application;

FIG. 14 is a schematic diagram of another network node structure according to an embodiment of the application.

Detailed ways

The application will be further described in detail below in conjunction with the drawings and embodiments.

Time Sensitive Networking (TSN, Time Sensitive Networking) evolved from the audio and video bridging (AVB, Ethernet Audio/Video Bridging) used for audio and video networks. It is a protocol set defined by the Institute of Electrical and Electronics Engineers (IEEE). Mainly used for small dedicated networks (Ethernet with a delay of 10 to 100 μs), such as in-vehicle networks (which can be understood as a network formed by multiple devices installed in a vehicle) or industrial control Network), which has also been defined for larger networks, for example, for fronthaul networks. The main idea is high priority and packet preemption (expressed as Packet in the text).

The scheduling mechanism of TSN network mainly includes the following aspects:

1. Credit-based shaper (CBS, credit-based sharper): A scheduling mechanism for a queue; the queue will get the credit value at the agreed rate, and the credit value is greater than or equal to 0 to send data packets. When sending data packets, The credit value will be reduced; the effect of this shaper is: to shape the packets in the queue and send them one by one at the agreed rate (also called pacing); after shaping, this kind of traffic will generally coexist with BE traffic on the sending port, and This kind of traffic requires a higher priority to ensure that it is not interfered by BE traffic, so as to maintain the effect of the previous queue shaping;

2. Time Sensitive Queues (TSQ, Time Sensitive Queues): Using a gating mechanism, all queues on the device use a circular scheduling mechanism (based on a gating table with a granularity of ns to control the opening and closing of the queue), The synchronization between devices relies on the Precision Time Protocol (PTP), through the coordination of each device on the path, the gate can be opened and closed accurately, and the fastest forwarding of TSN traffic is supported;

3. Transmission preemption: packet preemption strategy, supports high-priority packets to interrupt the low-priority packets being sent;

4. Ingress scheduling and cyclic queue forwarding (CQF, Input scheduling and Cyclic Queuing and Forwarding) mechanism: arrive at the correct time window at the entrance, and then ensure that it is sent from the exit at a certain time window, using several queues that cycle periodically as the sending queue ；

5. Urgency-based scheduling (UBS, Urgency Based Scheduling), also known as Asynchronous Traffic Shaping (ATS, Asynchronous Traffic Shaping): The purpose is to provide better overall delay than CQF, lower cost, and does not require inter-device The mechanism currently supports two shaping methods, namely, Length Rate Ratio (LRQ, Length Rate Quotient) and Token Bucket Emulation (TBE, Token Bucket Emulation). The scheduling effects of these two methods are similar to CBS. , Are applicable to the pacing of queue traffic;

6. Packet Replication and Elimination (PRE, Packet Replication and Elimination): Multiple transmission and selective reception, also called Frame Replication and Elimination (FRER, Frame Replication and Elimination).

The deterministic demand is by no means limited to the local two-layer network. More than 20 authors from different organizations have jointly written a Use Case manuscript, which elaborates the needs in nine major industries, including: professional audio and video ( pro audio&video), electrical utilities, building automation systems, wireless for industrial, cellular radio, industrial machine-to-machine communication (industrial M2M), mining (mining industry) , Private blockchain and network slicing (private blockchain and network slicing); at the same time, the scale of demand scenarios may be very large, including a national network, a large number of equipment and ultra-long distances. Based on this, Deterministic Networking (DetNet, Deterministic Networking) was created. DetNet is a deterministic network architecture defined by the Internet Engineering Task Force (IETF). It focuses on the determinism of the three-layer network and expands the capabilities of TSN from the second layer. Go to the third floor, as shown in Figure 1.

TSN is characterized by its small network scale and relatively simple traffic model, which supports the identification of each stream, or network synchronization; therefore, the related mechanism of TSN is mainly developed for small-scale networks. In large-scale networks, it is difficult to directly It is applied to IP forwarding equipment; moreover, the processing mechanism of deterministic traffic is relatively complicated by the relevant mechanism of TSN.

On the other hand, the current scheduling mechanism of IP networks belongs to the weighted round-robin (WRR) type. In this scheduling mechanism, the core idea of packet forwarding is to send data packets as quickly as possible. The core indicators are line speed and Throughput rate. Here, the line rate refers to: for a certain type of message, such as a series of 128 bytes (bytes), the device can achieve port rate entry and port rate transmission when forwarding. Under the current IP forwarding mechanism, as shown in Figure 2, even in a relatively lightly loaded network, the current scheduling of IP devices (ie network nodes) will produce a certain degree of microbursts, that is, pacing is good before scheduling. After scheduling, the messages will be gathered together. Therefore, in the design concept, the scheduling mechanism of IP network does not match some requirements of TSN and DetNet (100% reliability and deterministic delay must be guaranteed), so it is difficult to directly apply TSN scheduling in some details. mechanism.

As can be seen from the above description, the characteristics of IP forwarding as the old mechanism are statistical multiplexing, low cost, large throughput, and the design concept conforms to the burst traffic model of IP; the scheduling characteristic of TSN as the new mechanism is determinism. For specific traffic, it achieves a scheduling effect similar to constant bit rate (CBR).

Therefore, to support deterministic transmission on IP networks, the following operations must be avoided:

First, avoid requiring network forwarding nodes (that is, network nodes that forward packets, which can also be called P nodes, or P routers, or intermediate routers, as shown in Figure 3), such as operations in backbone networks The quotient node, to identify the business flow by flow.

Specifically, the network forwarding node (generally, the forwarding pressure is high) cannot be used to analyze each flow (Flow). This is because: the network forwarding node is generally responsible for forwarding according to the packet, which is not suitable for supporting too many flows. Identification work.

Second, avoid requiring too much state to be sensed/maintained on the network forwarding node.

Specifically, the number of flows on the network forwarding node is large. If a single flow changes, it is not recommended to constantly change the bandwidth reservation of the network forwarding node on the control plane. This also follows the current design concept of IP forwarding routers: In the IP network, there are many bursts of traffic. At this time, it is recommended that even if a certain/some flow access requirements change, the network forwarding nodes do not have to perceive them all.

Based on this, in various embodiments of the present application, on the network forwarding node, each low-latency flow is not identified, and only the flow with low-latency requirements is identified as a whole based on the characteristics of the packet, so as to ensure low-latency Traffic is sent on demand.

The embodiment of the present application provides a message transmission method, which is applied to a second network node. As shown in FIG. 4, the method includes:

Step 401: The second network node obtains the first message; and determines that the service corresponding to the first message is a delay-sensitive service;

Step 402: The second network node sets a first identifier for the first message;

Here, the first identifier represents the time delay sensitive demand of the first newspaper.

Step 403: The second network node sets the first message with the first identifier in a specific queue for shaping, and then sends the first message with the first identifier.

Wherein, the second network node is a network edge node, which may be referred to as a PE node, a PE router, etc., such as an operator edge node in a backbone network.

In actual application, in step 401, when the first packet is accessed from a specific virtual local area network (VLAN) or from a specific interface, or when the first packet is accessed from a certain If a specific interface or VLAN is accessed and has an agreed priority, the second network node considers the service corresponding to the first message to be a delay-sensitive service, that is, a low-latency service. Of course, other methods may also be used to identify that the service corresponding to the first message is a delay-sensitive service, which is not limited in the embodiment of the present application.

The first identifier characterizes the delay-sensitive requirement of the first newspaper and stationery, and can also be understood as the first identifier characterizing that the type of the first message is a delay-sensitive type.

In practical applications, the first identifier may be a priority identifier, which may identify that low-latency traffic occupies a higher exclusive priority, such as 6.

Based on this, in an embodiment, the first identifier indicates that the first packet has an exclusive sending priority.

In practical applications, the first identifier may also be an adjacent SID of a segment routing (SR) (it can be expressed as adj SID in English, and the type of adjacent SID is the End.X function in SRv6).

Figures 5a and 5b show the formats of two adjacent SIDs. In practical applications, you can specify a specific algorithm in the algorithm section, such as 200, which is used for low-latency traffic transmission.

In practical applications, the first identifier may also be a prefix SID of an SR (English may be expressed as Prefix SID, and the type of the prefix SID is End function in SRv6). Among them, the prefix SID in the format shown in FIG. 5a is suitable for a point-to-point (P2P) connection scenario (Type: 43), and the prefix SID in the format shown in FIG. 5b is suitable for a local area network (LAN) scenario (Type: 44).

Figure 6 shows the format of a prefix SID. In actual application, when the End function is released, it is a sub-TLV of the Locator TLV in the Intermediate System to Intermediate System (ISIS). Part, you can specify a specific algorithm, such as 200, which is used for low-latency traffic transmission.

In actual application, in step 403, the second network node may perform separate shaping for each access delay-sensitive service, so that the packets are sent out one by one at a certain rate (the rate required by the service flow); Wherein, the second network node can obtain the rate through a certain implementation method, for example, through manual/network management configuration (ie static configuration), or control plane transfer mode (which can also be understood as control plane notification), or Data plane transfer method (also called data plane notification method).

The method for shaping the second network node may use CBS, or LRQ, TBE of ATS, etc., which is not limited in the embodiment of the present application.

Correspondingly, an embodiment of the present application also provides a message transmission method, which is applied to a first network node, as shown in FIG. 7, including:

Step 701: Receive the first message;

Step 702: Obtain the first identifier from the first message;

Step 703: When the first identifier is obtained, set the first message in a specific queue;

Here, the specific queue is at least used for buffering (also can be understood as placing) delay-sensitive messages to be sent.

Step 704: Use the incoming rate of packets in the specific queue to determine the outgoing rate of packets in the specific queue;

Step 705: Shape the specific queue based on the determined output rate, and then send the first message.

Wherein, the first network node is a network forwarding node, which may be called a P node, a P router, etc., such as an operator node in a backbone network.

In an actual application, the first network node receives the first message from a previous hop node, and the previous hop node may be a second network node or a network forwarding node.

In practical applications, when the first identifier may be a priority identifier, for each egress delay-sensitive message, the first network node configures a special queue to uniformly shape all delay-sensitive messages Send out later, that is, all packets with the exclusive priority will be aggregated into this queue.

When the first identifier is an adjacent SID of an SR, the first network node recognizes that the current SID is issued by itself, and forwards the message to a pre-configured specific queue to unify all delay-sensitive messages Send after shaping, that is, in actual application, all the packets whose current SID is the adjacent SID will be aggregated into this queue.

Based on this, in an embodiment, the SID list is obtained from the first message; the SID list includes multiple SIDs corresponding to the traffic engineering path;

That is, when it is determined that the current SID indicates a specific queue on the next-hop network node corresponding to the first packet, it is determined that the first identifier is acquired.

Here, it can be understood that: for the second network node, the current SID refers to: the SID corresponding to the destination address (DA) in the received first message; accordingly, the first message The next-hop network node corresponding to the text refers to the network node corresponding to the DA in the first packet, that is, the second network node that receives the first packet.

When the first identifier is a prefix SID of an SR, the second network node recognizes that the SID is not issued by itself, searches the routing and forwarding table according to the SID, and the outbound interface is in a pre-configured specific queue, thereby forwarding the message To a specific queue, in this queue, all delay-sensitive packets will be uniformly shaped and sent out. That is, in actual applications, all packets whose outbound interface is the specific queue in the routing table will be aggregated into this queue.

Based on this, in an embodiment, the prefix SID is obtained from the first message;

Look up the next hop and outbound interface corresponding to the obtained prefix SID in the routing and forwarding table;

That is, when it is determined that the found outbound interface corresponds to the specific queue, it is determined that the first identifier is acquired.

In actual application, the output speed of the message is limited according to the "incoming rate of the packet of the destination virtual interface". In this way, for the entire IP device (ie P node), each port is based on the received low-latency traffic The speed of the message (that is, the incoming rate of the virtual interface), send these low-latency traffic messages, while ensuring a certain buffer (buffer) depth (can also be understood as a certain message buffer size), rather than according to the previous The mechanism of forwarding as soon as possible (ie BE forwarding mechanism).

Based on this, in an embodiment, the specific implementation of step 705 may include:

Using the ingress rate in combination with the queue depth to determine the outgoing rate of packets in the queue.

Wherein, the queue depth can be understood as the number of packets in the queue.

In an embodiment, the using the inbound rate in combination with the queue depth to determine the outbound rate of packets in the specific queue includes:

Wherein, when the queue depth is equal to the threshold, it is determined that the out rate is the second rate; the second rate is equal to the in rate.

Here, in actual application, a statistical period for determining the message rate can be set, for example, 20 μs, counting the number of messages entering the specific queue within 20 μs, and determining the incoming rate based on this.

In practical applications, the first value can be set according to needs, as long as the incoming rate is slightly greater than the outgoing rate.

When the queue depth is equal to the threshold, and the inbound rate is too large (for example, exceeds the inbound rate threshold (which can be set as required)), the outbound rate may be the fourth rate, and the fourth rate is the set outbound rate threshold (It can be set as required, and is less than the input rate).

In practical applications, the preset rate can be set as required.

The shaping method of the first network node may adopt CBS, or LRQ, TBE of ATS, etc., which is not limited in the embodiment of the present application.

The embodiment of the present application also provides a message transmission method. As shown in FIG. 8, the method includes:

Step 801: The second network node obtains the first message; and determines that the service corresponding to the first message is a delay-sensitive service;

Step 802: The second network node sets a first identifier for the first message; the first identifier represents the delay-sensitive demand of the first message;

Step 803: The second network node sets the first message with the first identifier in a specific queue for shaping, and then sends the first message with the first identifier;

Step 804: The first network node receives the first message, and obtains the first identifier from the received first message, and when the first identifier is obtained, sets the received first message in a specific queue; The specific queue is at least used for buffering delay-sensitive messages to be sent;

Step 805: The first network node uses the incoming rate of the specific queue message to determine the outgoing rate of the specific queue message; and based on the determined outgoing rate, the specific queue is shaped, and then the received The first message.

Wherein, the second network node is a network edge node; the first network node is a network forwarding node.

Here, it needs to be explained that the specific processing procedures of the first network node and the second network node have been described in detail above, and will not be repeated here.

In the message transmission method provided by the embodiments of the present application, a first network node receives a first message; obtains a first identifier from the first message; and the first identifier represents the delay-sensitive requirement of the first message When the first identifier is obtained, the first message is set in a specific queue; the specific queue is at least used for buffering delay-sensitive messages to be sent; using the specific queue message Incoming rate, determining the outgoing rate of messages in the specific queue; based on the determined outgoing rate, shaping the specific queue, and sending out the first message; wherein, the first network node is a network forwarding node; On the network forwarding node, each low-latency service flow is not identified, and only the flow with low-latency requirements is identified as a whole based on the characteristics of the packet, and the packets with low-latency requirements are set in a specific queue for shaping. Try to ensure that there is a certain queue depth in a specific queue instead of using BE's forwarding mechanism as soon as possible. In this way, the network forwarding node can ensure that low-latency traffic is sent in an orderly manner, and the processing of the network forwarding node minimizes the number of packets in the network The packet loss and buffer delay caused by the micro-burst can meet the delay requirements of the business.

The application will be further described in detail below in conjunction with application examples.

Application Example One

In this application embodiment, the transmission of low-latency services uses exclusive priority.

In a common IP forwarding scenario, low-latency traffic and BE traffic use the same destination IP. At this time, low-latency traffic needs to be distinguished, and these low-latency traffic need exclusive priority.

With reference to Figure 9, the message transmission process of this application embodiment includes:

Step 1: The network edge node (such as the PE1 node) identifies the flow of the packet packet Packet1 according to the related technology, and shapes the queue where the packet packet Packet1 is located, and confirms that the correct priority is assigned, the exclusive priority, such as Priority 6;

Here, after the flow identification is performed, it is identified that the packet packet Packet1 is a low-latency flow.

The specific identification and shaping process needs to be carried out in accordance with the pre-configured identification and shaping methods. Shaping can be done by CBS or ATS.

After this step is completed, low-latency traffic and BE traffic enter the network together.

Step 2: The packet packet Packet1 arrives at the inbound interface Iin1 of the network forwarding node P1, and the network forwarding node P1 searches for the outbound interface as Iout3 according to the DA of the packet packet Packet1;

Here, a specific queue corresponding to Priority 6 is configured on Iout3. This specific queue serves low-latency traffic, and this specific queue is associated with Priority 6.

The priority of the packet packet Packet1 is Priority 6. Therefore, the outbound interface is a specific queue pre-configured on Iout3 to serve low-latency traffic.

In order to simplify the description, suppose that the network forwarding node P1 has four interfaces, and a board has only one interface. Of course, in actual applications, the processing methods for multiple interfaces of a board are similar.

Step 3: The packet packet Packet1 arrives at the outbound interface Iout3 of the network forwarding node P1. For Iout3, all Priority 6 packets are placed in the same queue (ie a specific queue) (the source of the packet may be the inbound interface Iin1, Iin2, Iin4, That is, as long as the packets with priority 6 are all placed in the specific queue), they are shaped (such as CBS, ATS, etc.) according to the overall incoming rate as the outgoing rate.

Specifically, at the beginning or when the queue depth is low, it can be sent according to the incoming rate slightly at the outgoing rate; when the queue depth reaches a certain threshold, it will be sent according to the incoming rate = the outgoing rate; when the incoming rate = 0, the outgoing rate will be sent according to The preset fixed rate or the last recorded rate sends out the packets in the buffer.

Among them, an appropriate period of packet rate statistics can be set, for example, statistics are performed once every 20 μs.

When the statistical inbound rate is too large, a certain threshold can be set to assign the rate, and there is no need to forward according to the inbound rate. At this time, the specific queue depth will become larger. Generally speaking, in actual application, this situation will not last too long. This is because:

First, the speed limit is set at the network entrance;

Second, the micro-burst of low-latency traffic in the network has been reduced;

Third, the proportion of low-latency traffic at this time is not high in the entire network, mainly BE traffic.

That is to say, the "incoming rate = outgoing rate" here is just a concept, and the specific implementation may adjust internal parameters according to the network and traffic conditions.

For Iout3, for the specific queue where all Priority 6 packets are located, it is scheduled with other queues (that is, the queue of BE traffic, for example, the traffic of Priority 0 is BE traffic) according to the relevant mechanism, and sent from Iout3, such as low-latency traffic The priority is higher, so a message will be sent out as soon as there is a specific queue.

Application Example Two

In this application embodiment, the low-latency service uses a dedicated adjacent SID.

In the SR forwarding scenario, different adjacent SIDs can be used for low-latency traffic and BE traffic, so that low-latency traffic and BE traffic can be distinguished directly according to SID. At this time, there is no need to set exclusive priority, but it is still A separate queue is required, and a higher priority is configured (to ensure priority forwarding).

Step 1: The network edge node (such as the PE1 node) identifies the flow of the packet packet Packet1 according to related technologies, and shapes the queue where the packet packet Packet1 is located, and confirms that the correct SID list is marked (these SIDs are strict Engineering traffic (TE) path, and point to a specific queue resource on each node);

Step 2: The packet packet Packet1 arrives at the inbound interface Iin1 of the network forwarding node P1, the current SID is SID13, and the forwarding device searches for the specific queue Queue3 whose outbound interface is Iout3 according to SID13;

It should be noted that the packets matching SID13 on the board where the device inbound interfaces Iin2 and Iin4 are located will also be sent to Iout3 (SID13 is an adjacent SID that points to the specific queue Queue3 of Iout3, that is, packets corresponding to low-latency traffic will be sent to Iout3. Set in the specific queue Queue3.

Step 3: The packet packet Packet1 arrives at the outbound interface Iout3 of the network forwarding node P1, and the specific queue Queue3 of Iout3 is shaped (such as CBS, ATS, etc.) according to the overall queue ingress rate as the outbound rate.

Here, the specific processing procedure of shaping according to the overall queue ingress rate as the egress rate is the same as the application embodiment 1, and will not be repeated here.

For Iout3, the specific queue Queue3 where all SID13 messages are located is scheduled with other queues according to the existing mechanism and sent from Iout3.

In practical applications, these special adjacent SIDs can be notified to the network for network programming to meet the demands of low-latency services.

Application Example Three

In this application embodiment, the low-latency service uses a special prefix SID.

In the SR forwarding scenario, different prefix SIDs can be used for low-latency traffic and BE traffic, so that low-latency traffic and BE traffic can be distinguished directly according to SID. At this time, there is no need to set exclusive priority, but it is still A separate queue is required, and a higher priority is configured (to ensure priority forwarding).

Step 1: At the network edge node (such as the PE1 node), according to the related technology, identify the traffic of the packet packet 1, and shape the queue where the packet packet 1 is located, and confirm that the correct prefix SID is marked (in the SRv6 scenario In the SID, the location (Locator) corresponding to the SID has a related identifier. For example, Flex Algo ID is used. The outbound interface of its forwarding entry points to a specific queue resource on each node. The node generates a related forwarding entry, and the Locator is the address part of the SID in SRv6, which is used to route to the node that publishes the SID);

The difference from Application Example 2 is that the adjacent SID generally needs to be specified per hop, and is a label stack (SR-TE), here only one SID (SR-BE) is required, which is a global label (for example, corresponding to a PE node) .

Step 2: The packet packet Packet1 arrives at the inbound interface Iin1 of the network forwarding node P1, looks up the forwarding table according to SID9, and the outbound interface is the specific queue Queue3 of Iout3;

It should be noted that the other prefix SID of the board where the inbound interface Iin1 of the device is located may also correspond to the queue of the outbound interface as Queue3. This prefix SID is similar to SID9 and represents low-latency traffic;

The inbound interfaces Iin2 and Iin4 of the device may also receive packets with SID9 or other prefixed SID packets. The queue corresponding to the outbound interface is Queue3.

In other words, the packets corresponding to the low-latency traffic will be set in the specific queue Queue3.

Step 3: The packet packet Packet1 arrives at the outbound interface Iout3 of the forwarding node P1 of the network. For the specific Queue3 of Iout3, the overall queue ingress rate is used as the outbound rate for shaping (such as CBS, ATS, etc.) to be sent.

For Iout3, Queue3 and other queues are scheduled according to the existing mechanism and sent from Iout3.

In actual applications, these special prefix SIDs can be advertised to the network for network programming to meet the demands of low-latency services.

In addition, each port of each node can be configured with a specific queue, which becomes the interface of these special prefix SIDs, and supports the forwarding of low-latency traffic.

As can be seen from the above description, the application embodiment constructs a mechanism to ensure low-latency transmission in a larger IP three-layer network, which specifically includes:

Network edge nodes perform traffic identification and speed limit;

Each port of the network forwarding node monitors the speed of the received low-latency traffic, shapes and forwards low-latency packets according to the monitored speed.

Among them, as shown in Figure 10, on the network forwarding node, each physical outbound interface is divided into a specific virtual interface, corresponding to a specific queue provided for all low-latency traffic, and according to the aggregated low-latency traffic , Perform plastic surgery. In this way, on the network forwarding node, low-latency traffic can be guaranteed to be sent on demand, and the processing of the network forwarding node minimizes the micro-burst of messages, and maintains a suitable buffer depth for low-latency flows.

Among them, the identification mechanism used in the above mechanism to identify low-latency traffic includes:

The first method uses exclusive IP/Multiprotocol Label Switching (MPLS) priority to represent low-latency traffic, which is suitable for IP/MPLS networks.

The second method, using a specific SID to represent low-latency traffic, is suitable for SR networks. The specific SID may specifically be the adjacent SID of the SR-TE or the node SID of the SR-BE (that is, the prefix SID).

Using the scheme of the embodiment of this application, a relatively simple mechanism (1. identification and overall identification of low-latency traffic; 2. special shaping and forwarding of low-latency traffic) is adopted to reduce the formation of micro-bursts in IP forwarding (Messages will not form bursts due to being forwarded as soon as possible) to ensure fast and orderly forwarding of low-latency traffic in the IP network (messages are forwarded in the pacing mode as far as possible); when in a larger network, the main time is When the delay is optical fiber propagation, as long as the network needs to ensure normal forwarding as much as possible (that is, forwarding messages in an orderly manner so that the messages should not be gathered together as much as possible), and not forming micro bursts that cause packet loss, it can provide services with lower delay; There is no need to introduce too complicated flow recognition or too much state control.

In order to implement the method of the embodiment of the present application, the embodiment of the present application also provides a message transmission device, which is set on a first network node. As shown in FIG. 11, the device includes:

The receiving unit 111 is configured to receive the first message;

The first acquiring unit 112 is configured to acquire a first identifier from a first message; the first identifier represents the delay-sensitive requirement of the first message;

The first processing unit 113 is configured to, when the first identifier is obtained, set the first message in a specific queue; the specific queue is at least used to buffer the delay-sensitive messages to be sent; Determining the incoming rate of messages in the specific queue, determining the outgoing rate of messages in the specific queue; and shaping the specific queue based on the determined outgoing rate, and sending out the first message; wherein,

The first network node is a network forwarding node.

Wherein, in an embodiment, the first obtaining unit 112 is configured to obtain an SID list from the first message; the SID list includes multiple SIDs corresponding to the traffic engineering path;

Correspondingly, the first processing unit 113 is configured to set the first packet in the first packet when it is determined that the current SID indicates a specific queue on the next-hop network node corresponding to the first packet. In a specific queue.

In an embodiment, the first obtaining unit 112 is configured to:

Obtain the prefix SID from the first message;

Correspondingly, the first processing unit 113 is configured to set the first packet in the specific queue when the found outbound interface corresponds to the specific queue.

In an embodiment, the first processing unit 113 is configured to:

Wherein, in an embodiment, the using the inbound rate in combination with the queue depth to determine the outbound rate of packets in the specific queue includes:

When the queue depth is less than the threshold, the first processing unit 113 determines that the out rate is the first rate; the first rate is less than the in rate, and the difference between the in rate and the first rate is less than First value

or,

When the queue depth is equal to the threshold, the first processing unit 113 determines that the out rate is a second rate; the second rate is equal to the in rate.

or,

When the incoming rate is zero, the first processing unit 113 determines that the outgoing rate is a third rate; the third rate is a preset rate or the last recorded out rate.

In practical applications, the receiving unit 111 can be implemented by a communication interface in a message transmission device; the first acquiring unit 112 and the first processing unit 113 can be implemented by a processor in the message transmission device.

In order to implement the method on the second network node side of the embodiment of the present application, the embodiment of the present application also provides a message transmission device, which is set on the second network node. As shown in FIG. 12, the device includes:

The second obtaining unit 121 is configured to obtain the first message; and determine that the service corresponding to the first message is a delay-sensitive service;

The second processing unit 122 is configured to set a first identifier for the first message; the first identifier represents the delay-sensitive requirement of the first message; and the first message with the first identifier is set. The message is set in a specific queue for shaping, and then the first message with the first identifier is sent.

In practical applications, the second acquisition unit 121 may be implemented by a processor in a message transmission device in combination with a communication interface; the second processing unit 122 may be implemented by a processor in the message transmission device.

It should be noted that when the message transmission device provided in the above embodiment performs message transmission, only the division of the above-mentioned program modules is used as an example for illustration. In practical applications, the above-mentioned processing can be allocated to different program modules according to needs. Complete, that is, divide the internal structure of the device into different program modules to complete all or part of the processing described above. In addition, the message transmission device provided in the foregoing embodiment and the message transmission method embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

Based on the hardware implementation of the above program modules, and in order to implement the method on the first network node side of the embodiment of the present application, the embodiment of the present application also provides a network node. As shown in FIG. 13, the network node 130 includes:

The first communication interface 131 can exchange information with other network nodes;

The first processor 132 is connected to the first communication interface 131 to implement information interaction with other network nodes, and is configured to execute the method provided by one or more technical solutions on the first network node side when it is configured to run a computer program. The computer program is stored in the first memory 133.

Specifically, the first communication interface 131 is configured to receive the first message;

The first processor 132 is configured to obtain a first identifier from a first message; the first identifier represents the delay-sensitive requirement of the first newspaper; when the first identifier is obtained Next, the first message is set in a specific queue; the specific queue is used at least to buffer the delay-sensitive messages to be sent; and the incoming rate of the specific queue message is used to determine the specific queue message And based on the determined output rate, the specific queue is shaped, and the first message is sent out through the first communication interface; wherein,

The network node is a network forwarding node.

Wherein, in an embodiment, the first processor 132 is configured to:

Obtain the SID list from the first message; the SID list contains multiple SIDs corresponding to the traffic engineering path;

In an embodiment, the first processor 132 is configured to:

Obtain the prefix SID from the first message;

In an embodiment, the first processor 132 is configured to:

When the queue depth is less than the threshold, the first processor 132 determines that the out rate is a first rate; the first rate is less than the in rate, and the difference between the in rate and the first rate is less than First value

or,

When the queue depth is equal to the threshold, the first processor 132 determines that the out rate is a second rate; the second rate is equal to the in rate.

or,

When the in rate is zero, the first processor 132 determines that the out rate is a third rate; the third rate is a preset rate or the last recorded out rate.

It should be noted that the specific processing process of the first processor 132 can be understood with reference to the foregoing method.

Of course, in actual applications, the various components in the network node 130 are coupled together through the bus system 134. It can be understood that the bus system 134 is configured to implement connection and communication between these components. In addition to the data bus, the bus system 134 also includes a power bus, a control bus, and a status signal bus. However, for clarity of description, various buses are marked as the bus system 134 in FIG. 13.

The first memory 133 in the embodiment of the present application is configured to store various types of data to support the operation of the network node 130. Examples of such data include: any computer program used to operate on the network node 130.

The methods disclosed in the foregoing embodiments of the present application may be applied to the first processor 132 or implemented by the first processor 132. The first processor 132 may be an integrated circuit chip with signal processing capability. In the implementation process, the steps of the foregoing method may be completed by an integrated logic circuit of hardware in the first processor 132 or instructions in the form of software. The aforementioned first processor 132 may be a general-purpose processor, a digital signal processor (DSP, Digital Signal Processor), or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, and the like. The first processor 132 may implement or execute the methods, steps, and logical block diagrams disclosed in the embodiments of the present application. The general-purpose processor may be a microprocessor or any conventional processor or the like. Combining the steps of the method disclosed in the embodiments of the present application, it may be directly embodied as being executed and completed by a hardware decoding processor, or executed and completed by a combination of hardware and software modules in the decoding processor. The software module may be located in a storage medium, and the storage medium is located in the first memory 133. The first processor 132 reads the information in the first memory 133 and completes the steps of the foregoing method in combination with its hardware.

In an exemplary embodiment, the network node 130 may be configured by one or more Application Specific Integrated Circuits (ASIC, Application Specific Integrated Circuit), DSP, Programmable Logic Device (PLD, Programmable Logic Device), and Complex Programmable Logic Device (CPLD). , Complex Programmable Logic Device, Field-Programmable Gate Array (FPGA, Field-Programmable Gate Array), general-purpose processor, controller, microcontroller (MCU, Micro Controller Unit), microprocessor (Microprocessor), or other electronics Component implementation, used to perform the aforementioned method.

Based on the hardware implementation of the above program modules, and in order to implement the method on the second network node side of the embodiment of the present application, the embodiment of the present application also provides a network node. As shown in FIG. 14, the network node 140 includes:

The second communication interface 141 can exchange information with other network nodes;

The second processor 142 is connected to the second communication interface 141 to implement information interaction with other network nodes, and is configured to execute the method provided by one or more technical solutions on the second network node side when it is configured to run a computer program. The computer program is stored in the second storage 143.

Specifically, the second communication interface 141 is configured to obtain the first message;

The second processor 142 is configured to:

It is determined that the service corresponding to the first message is a delay-sensitive service; and the first message with the first identifier is set in a specific queue for shaping, and then sent through the second communication interface 141 with the first identifier The first message.

It should be noted that the specific processing procedures of the second processor 142 and the second communication interface 141 can be understood with reference to the foregoing method.

Of course, in actual applications, the various components in the network node 140 are coupled together through the bus system 144. It can be understood that the bus system 144 is configured to implement connection and communication between these components. In addition to the data bus, the bus system 144 also includes a power bus, a control bus, and a status signal bus. However, for clear description, various buses are marked as the bus system 144 in FIG. 14.

The second memory 143 in the embodiment of the present application is configured to store various types of data to support the operation of the network node 140. Examples of such data include: any computer program used to operate on the network node 140.

The method disclosed in the foregoing embodiment of the present application may be applied to the second processor 142 or implemented by the second processor 142. The second processor 142 may be an integrated circuit chip with signal processing capability. In the implementation process, the steps of the foregoing method may be completed by an integrated logic circuit of hardware in the second processor 142 or instructions in the form of software. The aforementioned second processor 142 may be a general-purpose processor, a DSP, or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. The second processor 142 may implement or execute the methods, steps, and logical block diagrams disclosed in the embodiments of the present application. The general-purpose processor may be a microprocessor or any conventional processor or the like. Combining the steps of the method disclosed in the embodiments of the present application, it may be directly embodied as being executed and completed by a hardware decoding processor, or executed and completed by a combination of hardware and software modules in the decoding processor. The software module may be located in a storage medium, and the storage medium is located in the second memory 143. The second processor 142 reads the information in the second memory 143 and completes the steps of the foregoing method in combination with its hardware.

In an exemplary embodiment, the network node 140 may be implemented by one or more ASICs, DSPs, PLDs, CPLDs, FPGAs, general-purpose processors, controllers, MCUs, Microprocessors, or other electronic components for performing the aforementioned methods.

It can be understood that the memory (the first memory 133, the second memory 143) of the embodiment of the present application may be a volatile memory or a non-volatile memory, and may also include both volatile and non-volatile memory. Among them, the non-volatile memory can be a read-only memory (ROM, Read Only Memory), a programmable read-only memory (PROM, Programmable Read-Only Memory), an erasable programmable read-only memory (EPROM, Erasable Programmable Read- Only Memory, Electrically Erasable Programmable Read-Only Memory (EEPROM), Ferromagnetic Random Access Memory (FRAM), Flash Memory, Magnetic Surface Memory , CD-ROM, or CD-ROM (Compact Disc Read-Only Memory); magnetic surface memory can be magnetic disk storage or tape storage. The volatile memory may be a random access memory (RAM, Random Access Memory), which is used as an external cache. By way of exemplary but not restrictive description, many forms of RAM are available, such as static random access memory (SRAM, Static Random Access Memory), synchronous static random access memory (SSRAM, Synchronous Static Random Access Memory), and dynamic random access memory. Memory (DRAM, Dynamic Random Access Memory), Synchronous Dynamic Random Access Memory (SDRAM, Synchronous Dynamic Random Access Memory), Double Data Rate Synchronous Dynamic Random Access Memory (DDRSDRAM, Double Data Rate Synchronous Dynamic Random Access Memory), enhanced Type synchronous dynamic random access memory (ESDRAM, Enhanced Synchronous Dynamic Random Access Memory), synchronous connection dynamic random access memory (SLDRAM, SyncLink Dynamic Random Access Memory), direct memory bus random access memory (DRRAM, Direct Rambus Random Access Memory) ). The memories described in the embodiments of the present application are intended to include, but are not limited to, these and any other suitable types of memories.

In order to implement the method in the embodiment of the present application, the embodiment of the present application also provides a message transmission system. The system includes a plurality of first network nodes and second network nodes.

It should be noted that the specific processing procedures of the first network node and the second network node have been described in detail above, and will not be repeated here.

In an exemplary embodiment, the embodiment of the present application also provides a storage medium, that is, a computer storage medium, specifically a computer-readable storage medium, such as a first memory 133 storing a computer program, which can be used by the network node 130 Is executed by the first processor 132 to complete the steps described in the foregoing first network node-side method. For another example, a second memory 143 storing a computer program is included. The computer program can be executed by the second processor 142 of the network node 140 to complete the steps described in the second network node-side method. The computer-readable storage medium may be a memory such as FRAM, ROM, PROM, EPROM, EEPROM, Flash Memory, magnetic surface memory, optical disk, or CD-ROM.

It should be noted that: "first", "second", etc. are used to distinguish similar objects, and not necessarily used to describe a specific sequence or sequence.

In addition, the technical solutions described in the embodiments of the present application can be combined arbitrarily without conflict.

The above are only preferred embodiments of the present application, and are not used to limit the protection scope of the present application.

Claims

A message transmission method, applied to a first network node, includes:

Receive the first message;

Obtain the first identifier from the first message; the first identifier represents the delay-sensitive demand of the first newspaper;

In the case of obtaining the first identifier, set the first message in a specific queue; the specific queue is at least used for buffering the delay-sensitive messages to be sent;

Using the incoming rate of packets in the specific queue to determine the outgoing rate of packets in the specific queue;

Based on the determined output rate, the specific queue is shaped, and the first packet is sent; wherein,

The first network node is a network forwarding node.
The method according to claim 1, wherein the first identifier characterizes that the first message has an exclusive sending priority.
The method according to claim 1, wherein said obtaining the first identifier from the first message comprises:

Obtain a segment identification list SID list from the first message; the SID list contains multiple SIDs corresponding to the traffic engineering path;

When it is determined that the current SID indicates a specific queue on the next-hop network node corresponding to the first message, the first message is set in the specific queue.
The method according to claim 1, wherein said obtaining the first identifier from the first message comprises:

Obtain the prefix SID from the first message;

Look up the outbound interface corresponding to the obtained prefix SID in the routing and forwarding table;

In a case where the found outgoing interface corresponds to the specific queue, the first packet is set in the specific queue.
The method according to any one of claims 1 to 4, wherein the determining the outgoing rate of packets in the specific queue by using the incoming rate of packets in the specific queue comprises:

The ingress rate is combined with the queue depth to determine the outbound rate of packets in the specific queue.
The method according to claim 5, wherein said determining the outgoing rate of packets in the specific queue by using the ingress rate in combination with the queue depth comprises:

When the queue depth is less than the threshold, determine that the out rate is a first rate; the first rate is less than the in rate, and the difference between the in rate and the first rate is less than the first value;

or,

When the queue depth is equal to the threshold, it is determined that the out rate is the second rate; the second rate is equal to the in rate;

or,

When the incoming rate is zero, it is determined that the outgoing rate is the third rate; the third rate is the preset rate or the last recorded out rate.
A message transmission method includes:

The second network node obtains the first message; and determines that the service corresponding to the first message is a delay-sensitive service;

The second network node sets a first identifier for the first message; the first identifier represents the delay-sensitive demand of the first message;

The second network node sets the first message with the first identifier in a specific queue for shaping, and then sends the first message with the first identifier;

The first network node receives the first message, and obtains the first identifier from the received first message;

When the first network node obtains the first identifier, set the received first message in a specific queue; the specific queue is at least used to buffer the delay-sensitive messages to be sent;

The first network node uses the incoming rate of the specific queue message to determine the outgoing rate of the specific queue message; and based on the determined outgoing rate, the specific queue is shaped, and the received first message is sent ;in,

The second network node is a network edge node; the first network node is a network forwarding node.
A message transmission device, which is arranged on a first network node, includes:

The receiving unit is configured to receive the first message;

The first acquiring unit is configured to acquire a first identifier from a first message; the first identifier represents the delay-sensitive requirement of the first message;

The first processing unit is configured to set the first message in a specific queue when the first identifier is obtained; the specific queue is at least used for buffering the delay-sensitive messages to be sent; The incoming rate of messages in the specific queue is determined, and the outgoing rate of messages in the specific queue is determined; and based on the determined outgoing rate, the specific queue is shaped to send out the first message; wherein,

The first network node is a network forwarding node.
A network node includes: a first communication interface and a first processor; wherein,

The first communication interface is configured to receive a first message;

The first processor is configured to obtain a first identifier from a first message; the first identifier represents the delay-sensitive requirement of the first newspaper; when the first identifier is obtained , Setting the first message in a specific queue; the specific queue is at least used for buffering delay-sensitive messages to be sent; and using the incoming rate of the specific queue messages to determine the Out rate; and based on the determined out rate, the specific queue is shaped, and the first message is sent through the first communication interface; wherein,

The network node is a network forwarding node.
A network node includes: a first processor and a first memory configured to store a computer program that can run on the processor,

Wherein, the first processor is configured to execute the steps of the method according to any one of claims 1 to 6 when running the computer program.
A storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are realized.