WO2023169407A1 - Method for issuing routing table and related apparatus - Google Patents

Abstract

The present application discloses a method for issuing a routing table and a related apparatus. A controller receives network measurement information from a first forwarding device, the network measurement information comprising flow measurement information for a large flow, and the large flow being a flow of which the number of packets within a unit period or a flow rate exceeds a threshold; the controller generates a routing decision according to the network measurement information; the controller sends the routing decision to the first forwarding device, the routing decision comprising a time threshold for aging a Flowlet table entry of a flow segment Flowlet routing table in the first forwarding device, the Flowlet table entry being used for indicating a next-hop port of a target packet in the large flow, and flow identity information of the target packet matching the Flowlet table entry. In the present application, the controller generates a routing decision according to the flow measurement information for the large flow from the forwarding device, and the forwarding device does not need to report flow measurement information for a small flow to the controller, thereby reducing the network overhead of the forwarding device sending the network measurement information, and reducing the data volume of the network measurement information acquired by the controller.

Description

A method and related devices for delivering routing tables

This application claims the priority of the Chinese patent application submitted to the China Patent Office on March 8, 2022, with the application number 202210227513.0 and the invention title "A method of delivering routing tables and related devices", the entire content of which is incorporated by reference. in this application.

Technical field

The embodiments of the present application relate to the field of communication technology, and in particular, to a method of delivering a routing table and related devices.

Background technique

In a data communication network, multiple switches and routers (hereinafter collectively referred to as forwarding devices) form a forwarding network for network traffic. A single data packet in network traffic can be routed and forwarded through different paths through the forwarding device ( routing) to the destination network device (such as a computer). When a large amount of network traffic passes through a forwarding device, in order to reduce network congestion and reduce packet forwarding delays, the forwarding device uses Layer 3 Load balance technology. When there are multiple forwarding paths for the same flow, the forwarding device equipped with Layer 3 load balancing technology chooses to forward all or part of the packets of the flow to different paths to achieve traffic balancing on different paths.

In the existing load balancing technology based on network quality measurement, each forwarding device needs to maintain a flow segment (Flowlet) routing table and a full path quality table. Among them, the full path quality table indicates all the traffic in the current layer 3 load balancing network. The network quality status of the forwarding path (such as queuing status, interface rate, packet loss, delay or flow count, etc.). After each packet reaches the forwarding device, the forwarding device determines the next hop port of the packet based on the Flowlet routing table and the full path quality table.

Since the full path quality table requires each forwarding device to measure the network quality of all forwarding paths in the current network in real time, it brings a huge overhead of network resources to each forwarding device. The load of the forwarding device is too large, resulting in a load balancing effect. reduce.

Contents of the invention

Embodiments of the present application provide a method and related devices for delivering routing tables to improve the effect of load balancing.

In the first aspect, embodiments of the present application provide a method for delivering a routing table. The controller manages each forwarding device in the network. In the initial stage (that is, the forwarding device and the controller have just been deployed to the network and have not yet received the Any message) The controller needs to understand the topology of the network, and the forwarding device needs to report the initialized Flowlet routing table to the controller, so that the controller can subsequently use the Flowlet routing table of all forwarding devices in the network (i.e., the global Flowlet routing table) to perform route calculations. In other words, the controller can obtain the physical topology of the network during the initialization phase by forwarding device reports or controller discovery. In this application, the method of delivering the routing table in this application is only performed on large flows in the network traffic, that is, in this application, the flow to which the target message belongs is a large flow. The forwarding device sends network measurement information to the controller. The network measurement information includes flow measurement information for large flows. For example, it may be delay information, packet loss information, or next hop information.

In some embodiments, in network traffic, a flow is a group of packets containing the same flow identifier (such as a five-tuple or a flow label, etc.). A large flow refers to a flow in which the number of packets or the flow rate in a unit period exceeds the threshold. In some embodiments, a group of packets in a flow in which the number of packets in the flow exceeds a threshold within a period of time is identified as a large flow. A flow that satisfies the large flow condition can be considered a large flow or a large burst. The corresponding flows that do not meet the large flow conditions can It can be considered as a small flow, or it can be considered as a small burst. A large flow can also be called an elephant flow, and a small flow can also be called a mouse flow. Scheduling for large flows can significantly improve the load on the network, while small flows have less impact on the load of the forwarding device due to the small number of messages. In this application, the forwarding device receives When packets come from small flows, they can be routed based on the equal-cost multi-path routing (ECMP) mechanism. Therefore, the forwarding device does not need to save the corresponding Flowlet entries for the small flows, nor does it need to report the Flowlet entries and flow measurement information of the small flows to the controller, thereby further reducing the network overhead of the forwarding device and reducing the collection required by the controller. The amount of data in the Flowlet table item.

The forwarding device referred to in the embodiment of this application may be a network device (such as a switch or router, etc.) with an IP packet forwarding function. Each of the steps implemented by the forwarding device can also be implemented by components in the network device, for example, by a chip system in the network device. The chip system includes logic circuits. The logic circuit can be coupled with the input/output interface and transmit messages through the input/output interface to implement each step in the above method embodiment. The logic circuit can be a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a network processor (network processor, NP), or other possible Programmed logic devices, discrete gate or transistor logic devices, discrete hardware components, or other integrated chips, etc.

The controller needs to maintain a global Flowlet routing table and a global network measurement table. The global Flowlet routing table includes the Flowlet routing table of each forwarding device, and the global network measurement table includes the network measurement information of each forwarding device.

After receiving the network measurement information of the forwarding device, the controller updates the entry corresponding to the forwarding device in the global network measurement table based on the network measurement information. The controller performs routing calculations based on the global Flowlet routing table and the updated global network measurement table. The controller generates a routing decision through the above routing calculation process. The routing decision indicates the aging time threshold of the Flowlet entry in the Flowlet routing table of the forwarding device. Furthermore, it can also indicate the next step for the target packet in the Flowlet routing table of the forwarding device. One-hop port field. The aging time threshold of a Flowlet entry determines whether the Flowlet entry is valid. For example, when the time interval between the forwarding device receiving the target packet does not exceed the above time threshold, the aging flag of the Flowlet entry matching the target packet is set to 1, so that the Flowlet entry is considered valid; when the forwarding device If the time interval for receiving the target packet exceeds the above time threshold, the aging flag of the Flowlet entry matching the target packet is set to 0, and the Flowlet entry is considered invalid.

It should be noted that the controller can set different time thresholds for a Flowlet entry or a group of Flowlet entries on each forwarding device, so that the above Flowlet entries have a better load under the corresponding next-hop route. Balance effect to avoid congestion.

Further, after the controller generates a new routing decision, it updates the Flowlet routing table for the forwarding device in the global Flowlet routing table. The updated global Flowlet routing table can continue to be used in subsequent routing calculations to generate routing decisions.

After the forwarding device receives the routing decision from the controller, it updates the Flowlet entry in the Flowlet routing table locally saved by the forwarding device based on the routing decision, specifically including updating the aging time threshold of the Flowlet entry or the Flowlet The next hop port field in the entry. After the forwarding device receives the target message, it can The packet is forwarded based on the next hop port indicated in the updated Flowlet entry.

In this application, the controller generates routing decisions based on the flow measurement information for large flows from the forwarding device. The forwarding device does not need to report the flow measurement information for small flows to the controller, thereby reducing the number of network messages sent by the forwarding device to the controller. The network overhead of measurement information reduces the amount of network measurement information that the controller needs to collect, reduces the computing overhead required by the controller to generate routing decisions, and improves the load balancing effect.

Based on the first aspect, in an optional implementation manner, the routing decision further includes a next-hop port field for the target packet in the Flowlet entry. Specifically, in the forwarding device, each Flowlet entry in the Flowlet routing table mainly includes the following fields: identification field, valid information field, and next hop port field. The identification field in the Flowlet entry includes identification. If the flow identity information of the target packet is hashed and the obtained ID is the same as the ID included in the ID field of a Flowlet entry, then the target packet can be considered to match the Flowlet entry. In this case, you can further determine how to process the packet based on the valid information field and next hop port field in the matching Flowlet entry. The valid information field in the Flowlet entry is used to indicate whether the Flowlet entry is currently valid, thereby determining whether matching packets can be forwarded based on the next hop port recorded in the next hop port field in the Flowlet entry. The aging time threshold of a Flowlet entry determines whether the Flowlet entry is valid. Therefore, in addition to delivering the aging time threshold of the Flowlet entry to the forwarding device, the controller can also deliver the next-hop port field for the target packet to the forwarding device, thus improving the update of the Flowlet entry of the forwarding device. Efficiency also improves the efficiency of message forwarding.

Based on the first aspect, in an optional implementation, the number of forwarding devices is not limited, that is, the forwarding device may include a first forwarding device, a second forwarding device, or other forwarding devices. Therefore, there may be multiple forwarding devices that send network measurement information to the controller, or there may be only one. If the controller receives the network measurement information of a forwarding device (i.e., the first forwarding device in this application), it generates a routing decision based on the network measurement information of the forwarding device (the first forwarding device) and issues it to the forwarding device. device (first forwarding device); if the controller receives multiple network measurement information from multiple forwarding devices (such as all forwarding devices or part of the forwarding devices in the entire network), where each forwarding device corresponds to one network measurement information, The controller then performs routing calculations based on the received multiple network measurement information, thereby generating routing decisions and issuing them to each forwarding device. In some possible implementations, the controller generates a routing decision based on network measurement information from the second forwarding device and/or the first forwarding device, and sends the routing decision to the second forwarding device and/or the first forwarding device. In this application, the controller can perform routing calculations based on multiple network measurement information from multiple forwarding devices and generate routing decisions corresponding to each forwarding device, which improves the flexibility of routing calculations and is more suitable for networks with many network devices. , complex network topology.

Based on the first aspect, in some possible implementations, when the forwarding device is forwarding messages and network congestion occurs on the port of the forwarding device, the forwarding device needs to obtain the network measurement information of its own device and report it to the controller. The network measurement information also includes port information on the forwarding device where network congestion occurs, which is used to reflect the network status of the forwarding device itself or the network quality of the next-hop port of the forwarding device, and so on. Exemplary, specific content forms of network measurement information include but are not limited to: port traffic where network congestion occurs, queue information, or other information reflecting network quality. One or more types of information, the specifics are not limited here. For those forwarding devices that do not experience network congestion, there is no need to report their network measurement information to the controller. The forwarding device can forward packets according to the local Flowlet routing table, thereby reducing the network overhead of the forwarding device and reducing the time required by the controller. The amount of network measurement information that needs to be collected reduces the computing overhead required by the controller to generate routing decisions. Furthermore, the network measurement information reported by the forwarding device, in addition to the information of the ports in the forwarding device where network congestion occurs, may also include the information of the ports in the forwarding device where network congestion does not occur, so that the controller can more comprehensively Understand the network quality of the forwarding device to perform routing calculations.

In the second aspect, embodiments of the present application provide a method for updating a routing table, including:

The forwarding device sends network measurement information to the controller, where the network measurement information includes flow measurement information for large flows, where the large flow is a flow whose number of packets or flow rate exceeds a threshold in a unit period;

The forwarding device receives a routing decision from the controller, where the routing decision is generated by the controller based on the network measurement information;

The forwarding device updates the aging time threshold of the Flowlet entry in the flow segment Flowlet routing table according to the routing decision. The Flowlet entry is used to indicate the next hop port of the target message in the large flow, where, The flow identity information of the target packet matches the Flowlet entry.

Based on the second aspect, in an optional implementation, the method further includes: the forwarding device updating the next hop port field in the Flowlet entry according to the routing decision.

Based on the second aspect, in an optional implementation, the routing decision includes multiple next-hop port fields in the Flowlet entry for the target message, and the method further includes: forwarding The device determines the target next hop port used to forward the target message according to the multiple next hop port fields.

Based on the second aspect, in an optional implementation manner, the forwarding device sends network measurement information to the controller, including: in response to network congestion occurring on a port of the forwarding device, the forwarding device sends The network measurement information further includes port information where network congestion occurs.

Based on the second aspect, in an optional implementation manner, the flow to which the target packet belongs is a large flow, and the large flow is a flow whose number of packets or flow rate in a unit period exceeds a threshold.

Based on the second aspect, in an optional implementation manner, the flow to which the target packet belongs is a large flow, and a large flow is a flow in which the number of packets in a unit period exceeds a threshold.

The information interaction and execution process of the embodiment shown in this aspect are based on the same concept as the embodiment shown in the first aspect. Therefore, for a description of the beneficial effects shown in this aspect, please refer to the above-mentioned first aspect. The details will not be described here.

In a third aspect, embodiments of the present application provide a controller, including:

The transceiver unit is used to receive network measurement information from the forwarding device. The network measurement information includes flow measurement information for large flows. A large flow is a flow in which the number of packets or the flow rate in a unit period exceeds a threshold;

a processing unit for generating routing decisions based on network measurement information;

The transceiver unit is also used to send routing decisions to the forwarding device. The routing decisions include flow segmentation Flowlet in the forwarding device. The aging time threshold of the Flowlet entry in the routing table. The Flowlet entry is used to indicate the next hop port of the target packet in the large flow. The flow identity information of the target packet matches the Flowlet entry.

Based on the third aspect, in an optional implementation manner, the routing decision further includes a next-hop port field for the target packet in the Flowlet entry.

Based on the third aspect, in an optional implementation manner, the number of forwarding devices is multiple, and the processing unit is specifically used for:

Generate routing decisions based on network measurement information from multiple forwarding devices.

Based on the third aspect, in an optional implementation manner, the network measurement information is sent by the forwarding device where the port where network congestion occurs is located, and the network measurement information also includes information about the port where network congestion occurs.

The information interaction and execution process of the embodiment shown in this aspect are based on the same concept as the embodiment shown in the first aspect. Therefore, for a description of the beneficial effects shown in this aspect, please refer to the above-mentioned first aspect. The details will not be described here.

In the fourth aspect, embodiments of the present application provide a device for updating a routing table, including:

The transceiver unit is used to send network measurement information to the controller. The network measurement information includes flow measurement information for large flows. A large flow is a flow whose number of packets or flow rate exceeds a threshold in a unit period;

The transceiver unit is also used to receive routing decisions from the controller. The routing decisions are generated by the controller based on network measurement information;

The processing unit is used to update the aging time threshold of the Flowlet entry in the flow segment Flowlet routing table according to the routing decision. The Flowlet entry is used to indicate the next hop port of the target packet in the large flow, where the target packet's The flow identity information matches the Flowlet entry.

Based on the fourth aspect, in an optional implementation, the processing unit is also used to:

Update the next hop port field in the Flowlet entry based on the routing decision.

Based on the fourth aspect, in an optional implementation, the routing decision includes multiple next-hop port fields for the target packet in the Flowlet entry, and the processing unit is also used to:

Determine the target next-hop port used to forward the target packet based on multiple next-hop port fields.

Based on the fourth aspect, in an optional implementation, the transceiver unit is specifically used for:

When network congestion occurs on a port of the packet forwarding device, network measurement information is sent to the controller, and the network measurement information also includes information on the port where network congestion occurs.

The information interaction and execution process of the embodiment shown in this aspect are based on the same concept as the embodiment shown in the first aspect. Therefore, for a description of the beneficial effects shown in this aspect, please refer to the above-mentioned first aspect. The details will not be described here.

In a fifth aspect, embodiments of the present invention provide a computer device, including a communication interface and a processor; the communication interface is used to communicate with other devices under the control of the processor; the processor is used to execute the computer Programs or instructions to cause the computer device to execute the method described in any of the above aspects.

In a sixth aspect, embodiments of the present application provide a computer-readable storage medium that stores a computer program that, when run on a computer, causes the computer to execute the following steps described in any of the above aspects. Method to distribute routing table.

In the seventh aspect, embodiments of the present application provide a computer program product or computer program. The computer program product or computer program includes a computer program or instructions that, when run on a computer, cause the computer to execute any of the above aspects. Method for delivering routing tables.

In an eighth aspect, embodiments of the present application provide a chip system. The chip system includes a processor for implementing the functions involved in the above aspects, for example, sending or processing the data and/or information involved in the above methods. . In a possible design, the chip system further includes a memory, and the memory is used to store necessary program instructions and data for the server or communication device. The chip system may be composed of chips, or may include chips and other discrete devices.

It can be seen from the above technical solutions that the embodiments of the present application have the following advantages:

This application discloses a method and related devices for delivering a routing table. The controller receives network measurement information from the first forwarding device. The network measurement information includes flow measurement information for large flows. A large flow is a flow within a unit period. The number of packets or the flow rate exceeds the threshold; the controller generates a routing decision based on the network measurement information; the controller sends the routing decision to the first forwarding device, and the routing decision includes the aging of the Flowlet entries in the flow segmentation Flowlet routing table in the first forwarding device. Time threshold, Flowlet entry is used to indicate the next hop port of the target packet in the large flow, where the flow identity information of the target packet matches the Flowlet entry. In this application, the controller generates routing decisions based on the flow measurement information for large flows from the forwarding device. The forwarding device does not need to report the flow measurement information for small flows to the controller, thereby reducing the number of network messages sent by the forwarding device to the controller. The network overhead of measurement information reduces the amount of network measurement information that the controller needs to collect, reduces the computing overhead required by the controller to generate routing decisions, and improves the load balancing effect.

Description of the drawings

In order to explain the embodiments of the present application or the technical solutions in the prior art more clearly, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings in the following description are only This is an embodiment of the present application. For those of ordinary skill in the art, other drawings can be obtained based on the provided drawings without exerting creative efforts.

Figure 1 is a schematic diagram of a message forwarding scenario based on Flowlet routing table;

Figure 2 is a schematic diagram of the architecture of a load balancing network based on network quality measurement;

Figure 3 is a schematic diagram of a forwarding device making routing decisions based on the full path quality table;

Figure 4 is a schematic diagram of an application scenario of the method of delivering a routing table in the embodiment of the present application;

Figure 5 is a schematic flowchart of a method for delivering a routing table in an embodiment of the present application;

Figure 6 is a schematic diagram of routing calculation based on the load balancing algorithm in this application;

Figure 7 is a schematic diagram of a scenario in which the controller delivers the next hop port field and time threshold in this embodiment of the present application;

Figure 8 is a schematic structural diagram of a controller provided by an embodiment of the present application;

Figure 9 is a schematic structural diagram of a device for updating a routing table provided by an embodiment of the present application;

Figure 10 is a schematic structural diagram of a computer device provided by an embodiment of the present application.

Detailed ways

Embodiments of the present application provide a method and related devices for delivering routing tables to improve the effect of load balancing.

The embodiments of the present invention will be described below with reference to the drawings in the embodiments of the present invention. The terms used in the embodiments of the present invention are only used to explain specific embodiments of the present invention and are not intended to limit the present invention. Persons of ordinary skill in the art know that with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

In this application, "at least one" refers to one or more, and "plurality" refers to two or more. "And/or" describes the relationship between associated objects, indicating that there can be three relationships, for example, A and/or B, which can mean: A exists alone, A and B exist simultaneously, and B exists alone, where A, B can be singular or plural. The character "/" generally indicates that the related objects are in an "or" relationship. "At least one of the following" or similar expressions thereof refers to any combination of these items, including any combination of a single item (items) or a plurality of items (items). For example, at least one of a, b, or c can mean: a, b, c, a-b, a-c, b-c, or a-b-c, where a, b, c can be single or multiple .

The terms "first", "second", "third", "fourth", etc. (if present) in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects without necessarily using Used to describe a specific order or sequence. It is to be understood that the figures so used are interchangeable under appropriate circumstances so that the embodiments of the invention described herein, for example, can be practiced in sequences other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, e.g., a process, method, system, product, or apparatus that encompasses a series of steps or units and need not be limited to those explicitly listed. Those steps or elements may instead include other steps or elements not expressly listed or inherent to the process, method, product or apparatus.

In a data communication network, multiple switches and routers (hereinafter collectively referred to as forwarding devices) form a forwarding network for network traffic. A single data packet in network traffic can be routed and forwarded through different paths through the forwarding device ( routing) to the destination network device (such as a computer). When a large amount of network traffic passes through a forwarding device, in order to reduce network congestion and reduce packet forwarding delays, the forwarding device uses Layer 3Load balance technology. When there are multiple forwarding paths for the same flow, the forwarding device equipped with Layer 3 load balancing technology chooses to forward all or part of the packets of the flow to different paths to achieve traffic balancing on different paths.

Traditional Layer 3 load balancing technology uses equal-cost multi-path routing (ECMP). Forwarding devices that use ECMP will save ECMP routing tables. After receiving the data packet, the forwarding device calculates an index (index) based on the flow identity information of the data packet, and then queries the ECMP routing table for an entry matching the index. Based on the next hop included in the entry, the forwarding device The port forwards the data packet. In practical applications, the bandwidth, delay and reliability of each path in the network are different, and the ECMP mechanism treats the cost of each path as the same and cannot make good use of bandwidth, especially when the differences between paths are large. It would be very unsatisfactory. For example, if the router has two exits and two paths, one has a bandwidth of 100M and the other has a bandwidth of 2M. If ECMP is deployed, the total network bandwidth can only achieve a utilization rate of 4M.

In view of the above-mentioned significant shortcomings of load balancing based on ECMP technology, load balancing technology based on flow segmentation (Flowlet) was proposed. The load balancing system of the forwarding device using this technology divides all packets belonging to a flow into one or multiple Flowlets based on the interval between two packets. There is a Flowlet routing table in the forwarding device, and each The table entry specifies the next hop route. Please refer to Figure 1, which is a schematic diagram of a packet forwarding scenario based on the Flowlet routing table. As shown in Figure 1, the forwarding device selects a forwarding path for the packet based on querying the Flowlet routing table.

In the existing load balancing system based on network quality measurement, each forwarding device needs to maintain a Flowlet routing table and a full path quality table. Among them, the full path quality table indicates the network quality of all forwarding paths in the current layer 3 load balancing network. Status (such as queuing status, interface rate, packet loss, delay or flow count, etc.). When a packet reaches the forwarding device, the forwarding device determines the next hop port of the packet based on the Flowlet routing table and full path quality table.

Please refer to Figure 2, which is a schematic diagram of the architecture of a load balancing network based on network quality measurement. As shown in Figure 2, the load balancing network includes multiple forwarding devices. In order to achieve load balancing, each forwarding device needs to maintain the Flowlet routing table and the full path quality table. The forwarding device obtains all forwarding by maintaining the full path quality table. Real-time network quality status of the path (such as queuing status, interface rate, packet loss, delay or flow count, etc.), so as to make routing decisions based on the quality of each forwarding path. In Figure 2, when a packet reaches the ingress port of forwarding device 1, forwarding device 1 selects different egress ports for it based on the Flowlet routing table and full path quality table, thereby forwarding the packet to different next-hop devices (forwarding device 1). Device 2 or forwarding device 3), so that the packet finally reaches the destination device through different paths in the network.

Specifically, each forwarding device stores a Flowlet routing table and a full path quality table. Among them, the full path quality table needs to be maintained based on the real-time network quality status so that the forwarding device can select the forwarding path with better network quality status. Forward the message. The following takes forwarding device 1 in Figure 2 as an example to introduce the processing flow of packets after they pass through forwarding device 1. Please refer to Figure 3. Figure 3 is a schematic diagram of a forwarding device making routing decisions based on the full path quality table. As shown in Figure 3, forwarding device 1 stores a Flowlet routing table and a full path quality table. After receiving the message, the forwarding device 1 obtains the identifier of the message by performing a hash operation on the flow identity information of the message. Forwarding device 1 uses this identifier to retrieve the Flowlet entry in the Flowlet routing table. If the identifier can match a certain Flowlet entry in the Flowlet routing table, the next destination indicated by the matched Flowlet entry can be used. Hop ports to forward the packet.

As shown in Figure 3, the Flowlet entry in the Flowlet routing table includes a valid bit field to indicate whether the Flowlet entry is valid. Forwarding device 1 maintains a timer, which can set a time threshold T. The time threshold T is the minimum time interval to determine whether two messages belong to the same Flowlet. The value of the aging flag field in the Flowlet entry is determined based on whether the time interval between two packet processing exceeds the time threshold T. When the time interval between forwarding device 1 receiving the message does not exceed the time threshold T, the aging flag of the Flowlet entry matching the message is set to 1, thus considering the Flowlet entry to be valid; when forwarding device 1 receives the If the time interval between packets has exceeded the time threshold T, the aging flag of the Flowlet entry matching the packet is set to 0, and the Flowlet entry is considered invalid.

When forwarding device 1 receives the message, if the matched Flowlet entry is shown to be valid, forwarding device 1 can forward the matched message based on the next hop port recorded in the next hop port field in the Flowlet entry. packet; if the matched Flowlet entry is invalid, the Flowlet entry cannot be used as a routing entry for processing packets matching the Flowlet entry. In this case, it indicates that the routing table entry records the next hop record of the previous Flowlet, and forwarding device 1 regards the message received this time as the first message of a new Flowlet. At this time, in order to achieve load balancing, it is necessary to find the egress port with the best (or better) network quality leading to the destination device of the packet in the full path quality table as the next hop in the Flowlet entry. The new value of the port field and the new Flowlet The valid bit field of the entry is set to 1, so that the packet is forwarded based on the new next-hop port field in the Flowlet entry.

It can be seen from the above that when the load balancing algorithm is enabled on the forwarding device, the network quality of all forwarding paths needs to be queried through the full path quality table. Since the network quality of each path changes in real time, the forwarding device also needs to update and maintain the full path quality table in real time. The specific process for the forwarding device to update and maintain the full path quality table is as follows:

1. Based on the network measurement of a single forwarding device, discover the network measurement information of the forwarding path from the forwarding device to the neighbor forwarding device; or, based on the device status when a flow passes through the forwarding device, update the network measurement information of the forwarding device itself.

2. The network measurement information is transmitted to the target device along with the packet forwarding. The target device calculates the network quality along the entire path and returns it to the corresponding forwarding device along the path.

3. After receiving feedback from the network measurement information, all forwarding devices update the local full path quality table.

Since the full path quality table requires each forwarding device to measure the network quality of all forwarding paths in the current network in real time, it brings a huge overhead of network resources to each forwarding device. The load of the forwarding device is too large, resulting in a load balancing effect. reduce. On the other hand, the forwarding process that is too cumbersome and expensive has limited applicable scenarios, making it difficult to support large-scale network topologies.

In view of this, this application discloses a method and related device for delivering a routing table to improve the load balancing effect. In the method of delivering routing tables in this application, one or more controllers are deployed in the entire load balancing network. Please refer to Figure 4. Figure 4 is a schematic diagram of an application scenario of the method of delivering a routing table in an embodiment of the present application. As shown in Figure 4, the controller receives network measurement information from each forwarding device (such as forwarding device 1, forwarding device 2 or forwarding device 3 in Figure 4) for the port where network congestion occurs, and then generates a The routing decision is sent to each forwarding device. The routing decision includes the aging time threshold of the Flowlet entry in the flow segment Flowlet routing table of the forwarding device. The Flowlet entry is used to indicate the next hop of the target packet in the large flow. port, where the flow identity information of the target packet matches the Flowlet entry. Each forwarding device forwards packets according to the received routing policy.

Next, the process of the above method of delivering routing tables is introduced. Please refer to Figure 5. Figure 5 is a schematic flow chart of a method of delivering a routing table in an embodiment of the present application. As shown in Figure 5, the method of delivering a routing table in an embodiment of the present application includes:

101. The forwarding device sends network measurement information to the controller.

The forwarding device in the network has basic telemetry capabilities. The forwarding device can collect the delay flow size or packet loss information of specific traffic in real time. Telemetry technology generally refers to the remote real-time high-speed collection of data from physical network elements or virtual network elements to achieve real-time, high-speed and more sophisticated information collection technology for the network. Compared with traditional network information collection technologies, such as simple network management protocol (SNMP), Telemetry actively pushes data information to the controller as a collector through push mode, providing more real-time, faster and more accurate information. Accurate network information collection function.

The controller manages each forwarding device in the network. In the initial stage (that is, the forwarding device and the controller have just been deployed to the network and have not yet received any messages) the controller needs to master the topology of the network, and the forwarding device needs to initialize The optimized Flowlet routing table is reported to the controller so that the controller can subsequently perform route calculations based on the Flowlet routing tables of all forwarding devices in the network (i.e., the global Flowlet routing table). In other words, the controller can obtain the physical topology of the network during the initialization phase by forwarding device reports or controller discovery. In this application, the method of delivering the routing table in this application is only performed on large flows in the network traffic, that is, in this application, the flow to which the target message belongs is a large flow. The forwarding device sends network measurement information to the controller. The network measurement information includes flow measurement information for large flows. For example, it may be delay information, packet loss information, or next hop information.

In some embodiments, in network traffic, a flow is a group of packets containing the same flow identifier (such as a five-tuple or a flow label, etc.). A large flow refers to a flow in which the number of packets or the flow rate in a unit period exceeds the threshold. In a specific implementation, a group of packets in a flow in which the number of packets in the flow exceeds a threshold within a period of time is identified as a large flow. A flow that satisfies the large flow condition can be considered a large flow or a large burst. The corresponding flow that does not meet the large flow condition can be considered as a small flow or a small burst. A large flow can also be called an elephant flow, and a small flow can also be called a mouse flow. Scheduling for large flows can significantly improve the load on the network, while small flows have less impact on the load of the forwarding device due to the small number of messages. In this application, the forwarding device receives When packets come from small flows, they can be routed based on the ECMP mechanism. Therefore, the forwarding device does not need to save the corresponding Flowlet entries for the small flows, nor does it need to report the Flowlet entries and flow measurement information of the small flows to the controller, thereby further reducing the network overhead of the forwarding device and reducing the collection required by the controller. The amount of data in the Flowlet table item.

In some possible implementations, when the forwarding device is forwarding messages, when network congestion occurs on the port of the forwarding device, the forwarding device needs to obtain the network measurement information of its own device and report it to the controller, where the network measurement information It also includes port information on the forwarding device where network congestion occurs, which is used to reflect the network status of the forwarding device itself or the network quality of the next hop port of the forwarding device, etc. For example, the specific content form of the network measurement information includes but is not limited to: one or more of port traffic where network congestion occurs, queue information, or other information reflecting network quality, which is not limited here. For those forwarding devices that do not experience network congestion, there is no need to report their network measurement information to the controller. The forwarding device can forward packets according to the local Flowlet routing table, thereby reducing the network overhead of the forwarding device and reducing the time required by the controller. The amount of network measurement information that needs to be collected reduces the computing overhead required by the controller to generate routing decisions. Furthermore, the network measurement information reported by the forwarding device, in addition to the information of the ports in the forwarding device where network congestion occurs, may also include the information of the ports in the forwarding device where network congestion does not occur, so that the controller can more comprehensively Understand the network quality of the forwarding device to perform routing calculations.

In practical applications, the forwarding device can obtain its network measurement information in a variety of ways. For example, the forwarding device can discover the network quality of the forwarding path from the forwarding device to the neighbor forwarding device through network measurements; or the forwarding device can determine the network quality of the forwarding path based on a flow. The device status when passing through the forwarding device is used to update the network measurement information of the forwarding device itself; or other methods of obtaining network measurement information may be used, which are not limited here.

In some possible implementations, the network measurement information may also include load conditions of the computing power of the forwarding device.

The forwarding device referred to in the embodiment of this application may be a network device (such as a switch or router) with a network protocol IP packet forwarding function. Each of the steps implemented by the forwarding device can also be implemented by components in the network device, for example, by a chip system in the network device. The chip system includes logic circuits. The logic circuit can be coupled with the input/output interface and transmit messages through the input/output interface to implement each step in the above method embodiment. steps. The logic circuit can be a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a network processor (NP), or other possible Programmed logic devices, discrete gate or transistor logic devices, discrete hardware components, or other integrated chips, etc.

In this application, the number of forwarding devices is not limited, that is, the forwarding device may include a first forwarding device, a second forwarding device, or other forwarding devices. Therefore, there may be multiple forwarding devices that send network measurement information to the controller, or there may be only one. If the controller receives the network measurement information of a forwarding device (i.e., the first forwarding device in this application), it generates a routing decision based on the network measurement information of the forwarding device (the first forwarding device) and issues it to the forwarding device. device (first forwarding device); if the controller receives multiple network measurement information from multiple forwarding devices (such as all forwarding devices or part of the forwarding devices in the entire network), where each forwarding device corresponds to one network measurement information, The controller then performs routing calculations based on the received multiple network measurement information, thereby generating routing decisions and issuing them to each forwarding device. In some possible implementations, the controller generates a routing decision based on network measurement information from the second forwarding device and/or the first forwarding device, and sends the routing decision to the second forwarding device and/or the first forwarding device. In this application, the controller can perform routing calculations based on multiple network measurement information from multiple forwarding devices and generate routing decisions corresponding to each forwarding device, which improves the flexibility of routing calculations and is more suitable for networks with many network devices. , complex network topology.

102. The controller generates routing decisions based on network measurement information.

As shown in Figure 5, the controller needs to maintain a global Flowlet routing table and a global network measurement table. The global Flowlet routing table includes its own Flowlet routing for each forwarding device (such as forwarding device 1 and forwarding device 2 in Figure 5). The global network measurement table includes the network measurement information of each forwarding device (for example, forwarding device 1 and forwarding device 2 in Figure 5).

After receiving the network measurement information of the forwarding device, the controller updates the entry corresponding to the forwarding device in the global network measurement table based on the network measurement information. The controller performs routing calculations based on the global Flowlet routing table and the updated global network measurement table. The controller generates a routing decision through the above routing calculation process. The routing decision indicates the aging time threshold of the Flowlet entry in the Flowlet routing table of the forwarding device. Furthermore, it can also indicate the next step for the target packet in the Flowlet routing table of the forwarding device. One-hop port field. Optionally, the route calculation process can also be further combined with historical network data. For example, based on historical network data, it can be concluded that there is obvious congestion on some paths in certain time periods, and you can give priority to avoid these future routes in advance. Paths where congestion may occur.

For example, please refer to Figure 6, which is a schematic diagram of route calculation based on the load balancing algorithm in this application. As shown in Figure 6, when a packet is transmitted from the source device to the destination device, it needs to pass through forwarding path 1, forwarding path 2, forwarding path 3, and forwarding path 4. Because the load on forwarding path 2 and forwarding path 3 is too large at this time, network congestion occurs. After receiving the network measurement information reported by the forwarding device where the port where the network congestion occurs is located, the controller performs route calculation based on the load balancing algorithm. , generate a new routing decision, thereby selecting forwarding path 5 and forwarding path 6 with less load as new forwarding paths, thereby replacing forwarding path 2 and forwarding path 3 where network congestion occurs. Based on the instructions of the new routing decision, the packet is delivered from the source device to the destination device and passes through forwarding path 1, forwarding path 5, forwarding path 6, and forwarding path 4 in sequence.

Further, after the controller generates a new routing decision, it updates the global Flowlet routing table for the forwarding The device's Flowlet routing table. The updated global Flowlet routing table can continue to be used in subsequent routing calculations to generate routing decisions.

In this application, the routing decision includes the aging time threshold of the Flowlet entry in the Flowlet routing table in the forwarding device. The Flowlet entry is used to indicate the next hop port of the target packet in the large flow, where the flow identity information of the target packet Matches the Flowlet entry. In some possible implementations, the routing decision also includes the next-hop port field for the target packet in the Flowlet entry. Specifically, in the forwarding device, each Flowlet entry in the Flowlet routing table mainly includes the following fields: identification field, valid information field, and next hop port field. The identification field in the Flowlet entry includes identification. If the flow identity information of the target packet is hashed and the obtained ID is the same as the ID included in the ID field of a Flowlet entry, then the target packet can be considered to match the Flowlet entry. In this case, you can further determine how to process the packet based on the valid information field and next hop port field in the matching Flowlet entry. The valid information field in the Flowlet entry is used to indicate whether the Flowlet entry is currently valid, thereby determining whether matching packets can be forwarded based on the next hop port recorded in the next hop port field in the Flowlet entry. The aging time threshold of a Flowlet entry determines whether the Flowlet entry is valid. Therefore, in addition to delivering the aging time threshold of the Flowlet entry to the forwarding device, the controller can also deliver the next-hop port field for the target packet to the forwarding device, thus improving the update of the Flowlet entry of the forwarding device. Efficiency also improves the efficiency of message forwarding.

In some embodiments, when the time interval between the forwarding device receiving the target packet does not exceed the above time threshold, the aging flag of the Flowlet entry matching the target packet is set to 1, thereby considering the Flowlet entry to be valid; when If the time interval at which the forwarding device receives the target packet exceeds the above time threshold, the aging flag of the Flowlet entry matching the target packet is set to 0, and the Flowlet entry is considered invalid.

For example, please refer to Figure 7. Figure 7 is a schematic diagram of a scenario in which the controller delivers the next hop port field and time threshold in this embodiment of the present application. As shown in Figure 7, forwarding device 1 and forwarding device 2 send their respective network measurement information to the controller, and the controller generates the next hop port field and the next hop port field of each forwarding device based on the global Flowlet routing table and global network measurement table. time threshold, and then send the next hop port field and time threshold to each forwarding device for execution by the forwarding device.

In some embodiments, the controller can set different time thresholds for a Flowlet entry or a group of Flowlet entries on each forwarding device, so that the above Flowlet entries have better performance under the corresponding next-hop route. Load balancing effect to avoid congestion.

103. The forwarding device updates the Flowlet entry in the Flowlet routing table based on the routing decision.

After the forwarding device receives the routing decision from the controller, it updates the Flowlet entry in the Flowlet routing table locally saved by the forwarding device based on the routing decision, specifically including updating the aging time threshold of the Flowlet entry or the Flowlet The next hop port field in the entry. After the forwarding device receives the target packet, it can forward the packet based on the next hop port indicated in the updated Flowlet entry.

In this application, the controller generates routing decisions based on the flow measurement information for large flows from the forwarding device. The forwarding device does not need to report the flow measurement information for small flows to the controller, thereby reducing the number of network messages sent by the forwarding device to the controller. The network overhead of measurement information reduces the amount of network measurement information that the controller needs to collect, reduces the computing overhead required by the controller to generate routing decisions, and improves the load balancing effect.

In some possible implementations, the routing decision generated by the controller indicates multiple next-hop ports for one Flowlet entry. That is, after the controller calculates the route, it considers that multiple next-hop ports can be used to deliver the target packet. Therefore, after receiving the routing policy, the forwarding device updates the Flowlet entry in the locally saved Flowlet routing table, and obtains the next-hop port field in the updated Flowlet entry, which includes the above instructions indicated by the controller. Multiple next hop ports. After receiving a target packet that matches the Flowlet entry, one port (target next hop port) is determined from multiple next hop ports to forward the target packet. Specifically, the forwarding device may determine the target next-hop port for forwarding the target message from multiple next-hop ports based on random selection; or may use other methods to determine the target next-hop port, such as , when the controller delivers the routing policy, it also indicates the network quality of each next-hop port, so that the forwarding device can select the next-hop port with the best (or better) network quality from multiple next-hop ports. As the target next hop port. This application does not limit the way in which the forwarding device selects a target next-hop port from multiple next-hop ports. Therefore, the controller instructs multiple next-hop ports to the forwarding device, and the forwarding device selects one of them to perform packet forwarding based on the needs of the actual scenario, thus improving the flexibility of packet forwarding.

On the basis of the embodiment corresponding to Figure 5, in order to better implement the above solution of the embodiment of the present application, the following also provides relevant equipment for implementing the above solution. Specifically, please refer to Figure 8, which is a schematic structural diagram of a controller provided by an embodiment of the present application. The controller includes:

The transceiver unit 201 is configured to receive network measurement information from the forwarding device. The network measurement information includes flow measurement information for large flows. A large flow is a flow in which the number of packets or the flow rate in a unit period exceeds a threshold;

The processing unit 202 is used to generate routing decisions based on network measurement information;

The transceiver unit 201 is also used to send routing decisions to the forwarding device. The routing decisions include the aging time threshold of the Flowlet entries in the flow segmentation Flowlet routing table in the forwarding device. The Flowlet entries are used to indicate the next destination of the target packet in the large flow. Hop port, where the flow identity information of the target packet matches the Flowlet entry.

In one possible design, the routing decision also includes the next-hop port field of the target packet in the Flowlet entry.

In one possible design, the number of forwarding devices is multiple, and the processing unit 202 is specifically used to:

Generate routing decisions based on network measurement information from multiple forwarding devices.

In a possible design, the network measurement information is sent by the forwarding device where the port where network congestion occurs is located, and the network measurement information also includes information about the port where network congestion occurs.

The information interaction, execution process, etc. between the modules/units in the message forwarding device are based on the same concept as the method embodiment corresponding to Figure 5 in this application. For specific content, please refer to the method embodiments shown above in this application. Narration will not be repeated here.

Please refer to Figure 9. Figure 9 is a schematic structural diagram of a device for updating a routing table provided by an embodiment of the present application. The device for updating a routing table includes:

The transceiver unit 301 is configured to send network measurement information to the controller and receive routing decisions from the controller. The network measurement information includes flow measurement information for large flows. The large flow is the number of packets in a unit period or the flow rate exceeds Threshold flow,routing decisions are generated by the controller based on,network measurement information;

The processing unit 302 is configured to update the aging time threshold of the Flowlet entry in the flow segment Flowlet routing table according to the routing decision. The Flowlet entry is used to indicate the next hop port of the target packet in the large flow, where the target packet The flow identity information matches the Flowlet entry.

In a possible design, the processing unit 302 is also used to:

Update the next hop port field in the Flowlet entry based on the routing decision.

In one possible design, the routing decision includes multiple next-hop port fields for the target packet in the Flowlet entry. The processing unit 302 is also used to:

Determine the target next-hop port used to forward the target packet based on multiple next-hop port fields.

In one possible design, the transceiver unit 301 is specifically used for:

When network congestion occurs on a port of the packet forwarding device, network measurement information is sent to the controller, and the network measurement information also includes information on the port where network congestion occurs.

The information interaction, execution process, etc. between the modules/units in the message forwarding device are based on the same concept as the method embodiment corresponding to Figure 5 in this application. For specific content, please refer to the method embodiments shown above in this application. Narration will not be repeated here.

The embodiment of the present application also provides a computer device. Please refer to Figure 10. Figure 10 is a schematic structural diagram of the computer device provided by the embodiment of the present application. The controller described in the corresponding embodiment of Figure 8 can be deployed on the computer device. And/or the message forwarding device described in the embodiment corresponding to Figure 9 is used to implement various steps performed by the controller and/or the forwarding device in the embodiment corresponding to Figure 5. Specifically, the computer equipment is implemented by one or more servers. The computer equipment may vary greatly due to different configurations or performance, and may include one or more central processing units (CPU) 422 (for example, one or one or more processors) and memory 432, one or more storage media 430 (eg, one or more mass storage devices) that stores applications 442 or data 444. Among them, the memory 432 and the storage medium 430 may be short-term storage or persistent storage. The program stored in the storage medium 430 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations on the computer device. Furthermore, the central processor 422 may be configured to communicate with the storage medium 430 and execute a series of instruction operations in the storage medium 430 on the computer device 400 .

The computer device may also include one or more power supplies 426, one or more wired or wireless network interfaces 450, one or more input and output interfaces 458, and/or, one or more operating systems 441, such as Windows Server ^™ , Mac OS X ^TM , Unix ^TM , Linux ^TM , FreeBSD ^TM and more.

It should be noted that the information interaction, execution process, etc. between various modules/units in the computer equipment are based on the same concept as the method embodiment corresponding to Figure 5 in this application. For specific content, please refer to the method implementation shown above in this application. The description in the example will not be repeated here.

An embodiment of the present application also provides a computer program product that, when run on a computer, causes the computer to perform the steps performed by the controller and/or the steps performed by the forwarding device in the method described in the embodiment shown in Figure 5. .

Embodiments of the present application also provide a computer-readable storage medium. The computer-readable storage medium stores a program for performing signal processing. When it is run on a computer, it causes the computer to execute the embodiment shown in Figure 5. Steps performed by the controller and/or steps performed by the forwarding device in the described method.

The image processing device provided by the embodiment of the present application may specifically be a chip. The chip may include a processing unit and a communication unit. The processing unit may be, for example, a processor. The communication unit may be, for example, an input/output interface, a pin or a circuit, etc. . The processing unit can execute computer execution instructions stored in the storage unit, so that the chip executes the method described in the embodiment shown in FIG. 5 . Optionally, the storage unit is a storage unit within the chip, such as a register, cache, etc. The storage unit may also be a storage unit located outside the chip in the wireless access device, such as Read-only memory (ROM) or other types of static storage devices that can store static information and instructions, random access memory (random access memory, RAM), etc.

It should be noted that the device embodiments described above are only illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physically separate. Rather than being a physical unit, it can be located in one place, or it can be distributed across multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. In addition, in the drawings of the device embodiments provided in this application, the connection relationship between modules indicates that there are communication connections between them, which can be specifically implemented as one or more communication buses or signal lines.

Through the above description of the embodiments, those skilled in the art can clearly understand that the present application can be implemented by software plus necessary general hardware. Of course, it can also be implemented by dedicated hardware including dedicated integrated circuits, dedicated CPUs, dedicated memories, Special components, etc. to achieve. In general, all functions performed by computer programs can be easily implemented with corresponding hardware. Moreover, the specific hardware structures used to implement the same function can also be diverse, such as analog circuits, digital circuits or special-purpose circuits. circuit etc. However, for this application, software program implementation is a better implementation in most cases. Based on this understanding, the technical solution of the present application can be embodied in the form of a software product in essence or that contributes to the existing technology. The computer software product is stored in a readable storage medium, such as a computer floppy disk. , U disk, mobile hard disk, ROM, RAM, magnetic disk or optical disk, etc., including several instructions to cause a computer device (which can be a personal computer, training device, or network device, etc.) to execute the steps described in various embodiments of this application. method.

In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it may be implemented in whole or in part in the form of a computer program product.

The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in or transmitted from one computer-readable storage medium to another, for example, the computer instructions may be transferred from a website, computer, training device, or data The center transmits to another website site, computer, training equipment or data center through wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can store, or a data storage device such as a training device or a data center integrated with one or more available media. The available media may be magnetic media (eg, floppy disk, hard disk, magnetic tape), optical media (eg, DVD), or semiconductor media (eg, solid state disk (Solid State Disk, SSD)), etc.

Claims (20)

A method for delivering routing tables, which is characterized by including: The controller receives network measurement information from the first forwarding device, where the network measurement information includes flow measurement information for a large flow, where the large flow is a flow whose number of packets or flow rate exceeds a threshold in a unit period; The controller generates routing decisions based on the network measurement information; The controller sends the routing decision to the first forwarding device, where the routing decision includes an aging time threshold of the Flowlet entry of the flow segmentation Flowlet routing table in the first forwarding device, and the Flowlet entry is used to Indicates the next hop port of the target packet in the large flow, where the flow identity information of the target packet matches the Flowlet entry. The method according to claim 1, characterized in that the routing decision further includes a next hop port field for the target message. The method according to claim 1 or 2, characterized in that the controller generates a routing decision based on the network measurement information, including: the controller generates a routing decision based on information from the second forwarding device and/or the first forwarding device. The network measurement information is used to generate routing decisions. The method according to any one of claims 1 to 3, characterized in that the network measurement information is sent by the forwarding device where the port where network congestion occurs is located, and the network measurement information also includes the time when network congestion occurs. Port information. A method for updating a routing table, which is characterized by including: The forwarding device sends network measurement information to the controller, where the network measurement information includes flow measurement information for large flows, where the large flow is a flow in which the number of packets or the flow rate in a unit period exceeds a threshold; The forwarding device receives a routing decision from the controller, where the routing decision is generated by the controller based on the network measurement information; The forwarding device updates the aging time threshold of the Flowlet entry in the flow segment Flowlet routing table according to the routing decision. The Flowlet entry is used to indicate the next hop port of the target message in the large flow, where, The flow identity information of the target packet matches the Flowlet entry. The method according to claim 5, further comprising: The forwarding device updates the next hop port field in the Flowlet entry according to the routing decision. The method according to claim 5 or 6, wherein the routing decision includes multiple next-hop port fields in the Flowlet entry for the target message, and the method further includes: The forwarding device determines a target next hop port for forwarding the target message according to the multiple next hop port fields. The method according to any one of claims 5 to 7, characterized in that the forwarding device sends network measurement information to the controller, including: In response to network congestion occurring at a port of the forwarding device, the forwarding device sends the network measurement information to the controller, where the network measurement information further includes port information where network congestion occurs. A controller, characterized by including: A transceiver unit configured to receive network measurement information from the first forwarding device, where the network measurement information includes Flow measurement information on large flows, where the number of packets or the flow rate in a unit period exceeds a threshold; A processing unit configured to generate routing decisions based on the network measurement information; The transceiver unit is also configured to send the routing decision to the first forwarding device. The routing decision includes the aging time threshold of the Flowlet entry of the flow segmentation Flowlet routing table in the forwarding device. The Flowlet entry Used to indicate the next hop port of a target packet in a large flow, where the flow identity information of the target packet matches the Flowlet entry. The controller according to claim 9, wherein the routing decision further includes a next hop port field for the target message. The controller according to claim 9 or 10, characterized in that the processing unit is specifically used for: Generate routing decisions based on network measurement information of the second forwarding device and/or the first forwarding device. The controller according to any one of claims 9 to 11, characterized in that the network measurement information is sent by the forwarding device where the port where network congestion occurs is located, and the network measurement information also includes network congestion. port information. A device for updating routing tables, characterized by including: A transceiver unit configured to send network measurement information to the controller, where the network measurement information includes flow measurement information for large flows, where the large flow is a flow whose number of messages or flow rate exceeds a threshold in a unit period; The transceiver unit is also configured to receive routing decisions from the controller, where the routing decisions are generated by the controller based on the network measurement information; A processing unit configured to update the aging time threshold of the Flowlet entry in the flow segment Flowlet routing table according to the routing decision. The Flowlet entry is used to indicate the next hop port of the target message in the large flow, wherein, The flow identity information of the target packet matches the Flowlet entry. The device according to claim 13, characterized in that the processing unit is also used to: Update the next hop port field in the Flowlet entry according to the routing decision. The device according to claim 13 or 14, wherein the routing decision includes multiple next hop port fields for the target message in the Flowlet entry, and the processing unit is further configured to : According to the plurality of next hop port fields, a target next hop port for forwarding the target message is determined. The device according to any one of claims 13 to 15, characterized in that the transceiver unit is specifically used for: When network congestion occurs on a port of the message forwarding device, the network measurement information is sent to the controller, where the network measurement information also includes information on the port where network congestion occurs. A computer device, characterized in that it includes a processor, The processor is configured to execute a computer program so that the computer device executes the method according to any one of claims 1 to 8. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented. A computer program product, characterized in that it includes a computer program that implements the method according to any one of claims 1 to 8 when the computer program is executed by a computer. A chip system, characterized in that the chip system includes at least one processor, and when the program instructions are executed in the at least one processor, the chip system executes as described in any one of claims 1 to 8 Methods.