Packet sending method, network node, and system

Patent metadata

Field	Value
Publication number	US-11394646-B2
Application number	US-202017094529-A
Country	US
Kind code	B2
Filing date	Nov 10, 2020
Priority date	May 11, 2018
Publication date	Jul 19, 2022
Grant date	Jul 19, 2022

Abstract

A packet sending method, a network node, and a system, where a controller obtains a forwarding latency requirement of a service flow and a destination address of the service flow, determines a forwarding path that meets the forwarding latency requirement, and determines a number of a first cycle time in which an ingress node forwards a packet and a number of a second cycle time in which an intermediate node forwards the packet. The controller sends a first entry to the ingress node. The first entry includes a sequence number of the packet and the first cycle time number. The controller sends a second entry to the intermediate node. The second entry includes the sequence number of the packet and the second cycle time number.

First claim

What is claimed is:

1. A method implemented by a controller, wherein the method comprises:
obtaining a forwarding latency requirement of a service flow and a destination address of the service flow;
determining, based on the forwarding latency requirement and the destination address, a forwarding path for forwarding the service flow, wherein a first latency of the forwarding path meets the forwarding latency requirement, wherein the forwarding path passes through a first network node and a second network node, wherein the first network node is an ingress node on the forwarding path, and wherein the second network node is an intermediate node on the forwarding path;
determining a first cycle time number in which the first network node forwards a packet and a second cycle time number in which the second network node forwards the packet, wherein the packet is in the service flow;
generating a first entry and a second entry, wherein the first entry comprises a first correspondence between a sequence number of the packet and the first cycle time number, and wherein the second entry comprises a second correspondence between the sequence number and the second cycle time number;
sending the first entry to the first network node indicating to send the packet during a first time period corresponding to the first cycle time number; and
sending the second entry to the second network node indicating to send the packet during a second time period corresponding to the second cycle time number.

2. The method of claim 1 , further comprising:
obtaining a first node latency of the first network node and a second node latency of the second network node, wherein the first node latency comprises a first packet processing latency and a first packet sending latency of the first network node, and wherein the second node latency comprises a second packet processing latency and a second packet sending latency of the second network node; and
obtaining a link latency of a link on the forwarding path,
wherein the latency meets the forwarding latency requirement when a sum of the first node latency, the second node latency, and the link latency falls within a range of the forwarding latency requirement.

3. The method of claim 1 , further comprising:
obtaining a first duration of a first cycle time of the first network node and a second duration of a second cycle time of the second network node;
obtaining a third cycle time number of the first network node and a fourth cycle time number of the second network node that correspond to a current time of the controller;
determining the first cycle time number based on the first duration and the third cycle time number; and
determining the second cycle time number based on the first duration, a second latency of a link from the first network node to the second network node, a second packet processing latency of the second network node, the fourth cycle time number, and the second duration.

4. The method of claim 1 , further comprising:
determining a first adjacent segment identifier of the first network node and a second adjacency segment identifier of the second network node;
generating a label stack comprising the first adjacency segment identifier and the second adjacency segment identifier; and
sending the label stack to the first network node indicating to forward the packet through the forwarding path.

5. A method implemented by a first network node, wherein the method comprises:
obtaining a forwarding latency requirement of a service flow and a destination address of the service flow;
obtaining, based on the forwarding latency requirement and the destination address, a forwarding path for forwarding the service flow, wherein a first latency of the forwarding path meets the forwarding latency requirement, and wherein the first network node is an ingress node on the forwarding path;
obtaining a first entry comprising a first correspondence between a sequence number of a packet and a first cycle time number in which the first network node forwards the packet, wherein the packet is in the service flow; and
sending, based on the first entry, the packet to a second network node during a first time period corresponding to the first cycle time number,
wherein the second network node is an intermediate node on the forwarding path.

6. The method of claim 5 , wherein obtaining the forwarding path comprises calculating the forwarding path based on the forwarding latency requirement and the destination address.

7. The method of claim 6 , further comprising:
generating a second entry comprising a second correspondence between the sequence number and a second cycle time number in which the second network node forwards the packet; and
sending the second entry to the second network node indicating to send the packet during a second time period corresponding to the second cycle time number.

8. The method of claim 5 , wherein obtaining the forwarding path comprises:
sending the forwarding latency requirement and the destination address to a controller instructing the controller to determine, based on the forwarding latency requirement and the destination address, the forwarding path and to generate the first entry; and
receiving, from the controller, the first entry and information about the forwarding path.

9. The method of claim 8 , wherein the information comprises address information of a third node on the forwarding path, wherein the method further comprises generating, by triggering based on the address information, a label switched path (LSP) corresponding to the forwarding path, wherein the LSP is used to forward the packet.

10. The method of claim 8 , further comprising:
obtaining, from the controller, a second entry comprising a second correspondence between the sequence number and a second cycle time number in which the second network node forwards the packet; and
sending the second entry to the second network node to indicate the second network node to send the packet during a second time period corresponding to the second cycle time number.

11. The method of claim 9 , wherein the information is a label stack corresponding to the forwarding path, wherein the label stack comprises a first adjacency segment identifier of the first network node and a second adjacency segment identifier of the second network node, and wherein the label stack indicates to forward the packet through the forwarding path.

12. The method of claim 5 , wherein sending the packet to the second network node comprises:
matching the first entry based on the sequence number;
adding the packet to a packet queue corresponding to the first cycle time number; and
sending the packet to the second network node when the packet queue is a sending queue.

13. A controller comprising:
a memory configured to store instructions;
a processor coupled to the memory, wherein the instructions cause the processor to be configured to:
obtain a forwarding latency requirement of a service flow and a destination address of the service flow;
determine, based on the forwarding latency requirement and the destination address, a forwarding path for forwarding the service flow, wherein a first latency of the forwarding path meets the forwarding latency requirement, wherein the forwarding path passes through a first network node and a second network node, wherein the first network node is an ingress node on the forwarding path, and wherein the second network node is an intermediate node on the forwarding path;
determine a first cycle time number in which the first network node forwards a packet and a second cycle time number in which the second network node forw

Assignees

• Huawei Tech Co Ltd

Inventors

• Geng Xuesong
• Chen Guoyi

Classifications

• H04L47/56 Primary

  implementing delay-aware scheduling · CPC title

• H04L45/70

  Routing based on monitoring results · CPC title

• H04L47/28

  in relation to timing considerations · CPC title

• H04L47/24

  Traffic characterised by specific attributes, e.g. priority or QoS · CPC title

• H04L45/302

  Route determination based on requested QoS · CPC title