WO2021209016A1

WO2021209016A1 - Method for processing message in network device, and related device

Info

Publication number: WO2021209016A1
Application number: PCT/CN2021/087575
Authority: WO
Inventors: 张镇星; 李楠; 黄超
Original assignee: 华为技术有限公司
Priority date: 2020-04-17
Filing date: 2021-04-15
Publication date: 2021-10-21
Also published as: CN113542152A

Abstract

Provided are a method for controlling congestion, and a related device. The method comprises: a policy management entity in a network device determining a message processing policy according to the state of a cached entity, and sending the determined message processing policy to a message processing entity; and the message processing entity processing a new received message according to the message processing policy. In the technical solution, the message processing entity can execute the corresponding message processing policy according to the state of the cached entity, so as to process the new received message. In this way, the cached entity can be better managed, so as to reduce the occurrence of cache management exception of the cached entity.

Description

Method for processing messages in network equipment and related equipment

This application claims the priority of the Chinese patent application filed with the Chinese Patent Office on April 17, 2020, the application number is 202010307569.8, and the application name is "methods for processing messages in network equipment and related equipment", the entire content of which is incorporated by reference In this application.

Technical field

This application relates to the field of information technology, and more specifically, to a method for processing packets in network equipment and related equipment.

Background technique

At present, in order to avoid congestion, some network devices adopt the following strategy: if it is determined that the egress port of the network device is congested, the upstream device is requested to stop sending packets sent through the egress port. However, it takes a certain time for the network device to send a notification to the upstream device, and it may also take a certain time for the upstream device to respond to the notification. During this period of time, the upstream device will continue to send packets to the network device. In order to avoid the occurrence of congestion, the buffer set in the network device can be used to buffer the packets received from the upstream device within this period of time.

However, in some cases, the upstream device may not normally respond to the request sent by the network device and continue to send packets to the network device. At this time, the network device may not be able to buffer the newly received message.

In addition, network devices usually manage the cache space according to pre-configured rules. But if an error occurs in the pre-configured rules, an abnormal situation may also occur in the cache space.

Summary of the invention

The present application provides a method for processing messages in a network device and related devices, which can reduce the problem of cache space.

In a first aspect, an embodiment of the present application provides a method for processing a message in a network device. The network device includes a policy management entity, a cache entity, and a message processing entity. The method includes: the policy management entity obtains first cache status information , The first cache state information indicates the first state of the cache entity at the first moment; the policy management entity determines a first packet processing strategy according to the first cache state information, and sends the first packet processing strategy To the message processing entity; the message processing entity processes the newly received message according to the first message processing strategy. In the above technical solution, the message processing entity may execute a corresponding message processing strategy according to the state of the cache entity to process the newly received message. In this way, the cache entity can be better managed to reduce the occurrence of cache management exceptions in the cache entity.

In a specific design, before the policy management entity obtains the first cache status information, the method further includes: the policy management entity obtains at least one of the following information of the cache entity at the first moment: the length of the sending queue, the The buffer occupancy of the sending queue, the delay of the sending queue, the available capacity of the buffer entity, the used capacity of the buffer entity, the rate of the outgoing port of the buffer entity; the policy management entity obtains the at least one kind of information according to , Determine the first cache status information.

In a specific design, the method further includes: the cache entity determining the first cache state information; the policy management entity obtaining the first cache state information includes: the policy management entity obtains the first cache state from the cache entity information.

In a specific design, that the policy management entity obtains the first cache state information from the cache entity includes: the policy management entity receives the first cache state information sent by the cache entity. In the foregoing embodiment, the cache entity may send the cache status information to the policy management entity by itself. This can reduce the workload of the policy management entity.

In a specific design, before the cache entity sends the first cache state information to the policy management entity, the method further includes: the cache entity determines that the first state satisfies a first preset condition. In the above technical solution, the cache entity sends the cache state information to the policy management entity only when the state meets the preset condition, which can reduce the occupation of the outbound interface of the cache entity and can reduce the workload of the policy management entity.

In a specific design, the first state includes normal or abnormal.

In a specific design, the first state includes at least one of the following: the length of the sending queue, the buffer occupancy of the sending queue, the delay of the sending queue, the available capacity of the buffer entity, and the used capacity of the buffer entity Capacity, the rate of the outgoing port of the cache entity.

In a specific design, before the policy management entity determines the first message processing strategy based on the first cache state information, the method further includes: the policy management entity determines that the first cache state information satisfies a second preset condition.

In a specific design, the cache entity includes a target cache space, the first state includes the usage of the target cache space at the first moment; the policy management entity determines the first message according to the first cache state information The processing strategy includes: the strategy management entity determines the first packet processing strategy according to the usage of the target cache space at the first moment.

In a specific design, the method further includes: the policy management entity obtains second cache state information, the second cache state information indicating the second state of the cache entity at the second moment; the policy management entity according to the second state Cache status information, determine a second message processing strategy, and send the second message processing strategy to the message processing entity; the message processing entity processes the newly received message according to the second message processing strategy.

In a specific design, the second message processing strategy is used to instruct the message processing entity to no longer execute the first message processing strategy.

In a specific design, the second message processing strategy is also used to instruct the message processing entity to use the specified message processing strategy to process the newly received message.

In a specific design, the designated message processing strategy is the default message processing strategy or the message processing strategy used before the first message processing strategy is used.

In a specific design, the newly received message includes a lossless traffic message.

In a specific design, the message processing entity processes the newly received message according to the first message processing strategy, including: the message processing entity adopts one or more of the following processing methods: discarding all new messages Received messages; discard part of the newly received messages; modify the message information of all newly received messages; or modify the message information of some newly received messages.

In a specific design, modifying the message information of the newly received message includes: modifying one or more of the following information of the newly received message: the priority of the newly received message; The discard enable bit of the newly received message; the discard priority of the newly received message; the port number of the newly received message; or, the display congestion flag of the newly received message .

In a second aspect, an embodiment of the present application provides a network device. The network device includes a network device including a policy management entity, a cache entity, and a message processing entity. The policy management entity, the cache entity, and the message processing entity are used to implement the first aspect or any one of the designs of the first aspect.

In a third aspect, an embodiment of the present application provides a network device. The network device includes a first processor, a second processor, a first memory, and a second memory. The first memory is used to store the first processor and the second processor. The program code executed by the device. The second memory is used to buffer messages from the upstream device. The first processor executes the program code stored in the first memory to realize the function of the policy management entity in the first aspect or any design of the first aspect, and the second processor executes the program code stored in the first memory to implement the first aspect. The function of the message processing entity in one aspect or any design of the first aspect.

In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, where the computer-readable storage medium stores instructions for implementing the first aspect or the method described in any one of the possible implementation manners of the first aspect.

In the fifth aspect, this application provides a computer program product containing instructions, when the computer program product is run on a computer, the computer can execute the above-mentioned first aspect or any one of the possible implementations of the first aspect. method.

Description of the drawings

Figure 1 is a schematic diagram of a network.

Figure 2 is a schematic structural block diagram of a network device.

Fig. 3 is a schematic flowchart of a method for processing a packet in a network device according to an embodiment of the present application.

Figure 4 is a schematic diagram of reserved cache space, shared cache space and headroom cache space.

Figure 5 is another schematic diagram of reserved cache space, shared cache space and headroom cache space.

Fig. 6 is a schematic flowchart of a method for processing a packet in a network device according to an embodiment of the present application.

Fig. 7 is a structural block diagram of a network device provided according to an embodiment of the present application.

Fig. 8 is a structural block diagram of a network device provided according to an embodiment of the present application.

Detailed ways

The technical solution in this application will be described below in conjunction with the accompanying drawings.

The network device in the embodiment of the present application may be a network device with a routing function (for example, a router) or a network device with a switching function (for example, a switch). The network equipment in the embodiments of this application may be a network equipment in a wired communication network, or a wireless communication network (such as the Global System of Mobile Communication, GSM) system, Code Division Multiple Access, CDMA) systems, Long Term Evolution (LTE) systems, and future 5G networks, etc.).

This application will present various aspects, embodiments, or features around a system that may include multiple devices, components, modules, and the like. It should be understood and understood that each system may include additional devices, components, modules, etc., and/or may not include all the devices, components, modules, etc. discussed in conjunction with the accompanying drawings. In addition, a combination of these schemes can also be used.

In addition, in the embodiments of the present application, words such as "exemplary" and "for example" are used to represent examples, illustrations, or illustrations. Any embodiment or design solution described as an "example" in this application should not be construed as being more preferable or advantageous than other embodiments or design solutions. Rather, the term example is used to present the concept in a concrete way.

In the embodiments of this application, "corresponding (relevant)" and "corresponding" can sometimes be used together. It should be pointed out that the meanings to be expressed are the same when the difference is not emphasized.

Embodiment of the present application, the subscript sometimes as W ₁ may form a clerical error at non-target as W1, while not emphasize the difference, to express their meaning is the same.

The network architecture and business scenarios described in the embodiments of this application are intended to more clearly illustrate the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided in the embodiments of this application. Those of ordinary skill in the art will know that with the network With the evolution of architecture and the emergence of new business scenarios, the technical solutions provided in the embodiments of the present application are equally applicable to similar technical problems.

References described in this specification to "one embodiment" or "some embodiments", etc. mean that one or more embodiments of the present application include a specific feature, structure, or characteristic described in combination with the embodiment. Therefore, the sentences "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. appearing in different places in this specification are not necessarily All refer to the same embodiment, but mean "one or more but not all embodiments" unless it is specifically emphasized otherwise. The terms "including", "including", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

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

Figure 1 is a schematic diagram of a network. As shown in Figure 1, the network 100 includes a network device 110, an upstream device 121, an upstream device 122, an upstream device 123, an upstream device 124, an upstream device 125, a downstream device 131, a downstream device 132, a downstream device 133, a downstream device 134 and Downstream equipment 135. The upstream device can be a terminal device, such as a computer, a mobile phone, a tablet computer, etc., or a network device. Similarly, the downstream device can be a terminal device or a network device.

As shown in FIG. 1, the network device 110 can receive packets from the upstream device 121, the upstream device 122, and the upstream device 123 through the port 111. The network device 110 can receive packets from the upstream device 124 through the port 112. The network device 110 can receive packets from the upstream device 115 through the port 113.

The network device 110 may send packets to the downstream device 131 and the downstream device 132 through the port 114. The network device 110 may send the message to the downstream device 133, the downstream device 134, and the downstream device 135 through the port 115.

It is understandable that FIG. 1 is only a schematic diagram of a network to help those skilled in the art understand the method of the present application, and is not a limitation on the network to which the technical solution of the present application can be applied. For example, in some embodiments, the network device 110 may also receive a packet sent by one or more upstream devices other than the upstream device 121 and the upstream device 122 through the port 111. For another example, the network device 110 may also receive packets sent from one or more upstream devices through another port. For another example, the network device 110 may also send a packet to one or more downstream devices other than the downstream device 131 and the downstream device 132 through the port 114. For another example, the network device 110 may also send packets to one or more downstream devices through another port.

The ports (for example, port 111, port 112, and port 113) in the network device for receiving packets from the upstream device may be referred to as the ingress port of the network device (abbreviated as the ingress port). The ports (for example, port 114 and port 115) in the network device used to send packets to the downstream device may be referred to as the outgoing port of the network device (abbreviated as the outgoing port).

Fig. 2 is a schematic structural block diagram of a network device provided according to an embodiment of the present application. Fig. 2 is a schematic structural block diagram of the network device 110 shown in Fig. 1.

As shown in FIG. 2, the network device 110 includes: a policy management entity 141, a cache entity 142, and a message processing entity 143. In addition, the network device 110 may also include a port 111, a port 112 and a port 113, and a port 114 and a port 115.

Port 111 to port 113 are ingress ports. Port 111 to port 113 can be used to receive messages from upstream devices and send the received messages to the message processing entity 143.

The message processing entity 143 is used for processing messages from the ingress port. For example, the message processing entity 143 can determine whether to allow the message to pass through. If the message is allowed to pass, the message can be sent to the cache entity 142. If the message is not allowed to pass, the message can be deleted, or the discard enable is added to the message header information, and the modified message is sent to the cache entity 142. The cache entity 142 may discard the message after receiving the message containing the discard enable. For another example, the message processing entity 143 can also modify information such as the priority and the outgoing port of the message.

After the cache entity 142 obtains the message from the message processing entity 143, if it determines that the message does not need to be discarded, it can cache the message, and cache the message through the corresponding out port (for example, port 114 or port 115). The message is sent to the downstream device.

The policy management entity 141 may determine the message processing policy according to the state of the cache entity 142. The message processing entity 143 may obtain the message processing strategy determined by the policy management entity 141, and execute the message processing strategy.

The specific functions and beneficial effects of the policy management entity 141, the cache entity 142, and the message processing entity 143 will be described in detail with reference to the following embodiments.

The method for processing packets in the network device provided by the present application will be introduced below in conjunction with FIG. 1 to FIG. 5.

The network device (for example, the network device 110) receives the message sent by the upstream device through the ingress port, and sends the message to the corresponding downstream device through the egress port. In some cases, the network device is unable to send packets from the upstream device to the downstream device in time due to some reasons (such as congestion on the outbound port, the flow of packets received on the ingress port is greater than the flow of packets sent on the outbound port, etc.) . In this case, the network device can cache the messages that were not sent in time to the cache entity (such as the cache entity 142) of the network and the device, and wait for the outgoing port to be able to send these messages, and then cache the messages cached by the cache entity. The document is sent to the downstream device.

According to different functions, the buffer space used for buffering messages in the buffer entity can be divided into two types, which can be called the first buffer space and the second buffer space respectively.

The first buffer space can be shared by different types of messages. In other words, the first buffer space can store any one or more types of messages, and for any one or more types of messages, as long as the first buffer space has available capacity, the messages can be cached to The first buffer space. For example, the first buffer space can all be used to store type 1 messages.

The second cache space may include multiple sub-cache spaces. The multiple sub-cache spaces, the multiple sub-cache spaces have a one-to-one correspondence with the multiple types, and each sub-cache space in the multiple sub-cache spaces is used to store messages of a corresponding type. For example, assuming that the total capacity of the second cache space is 512MB, the second cache space includes 8 sub-cache spaces, and the capacity of each sub-cache space in the 8 sub-cache spaces may be 64 MB. The 8 sub-buffer spaces correspond to the 8 types in a one-to-one manner. The 8 types can be called type 0, type 1, ..., type 7, respectively. The eight sub-cache spaces may be referred to as sub-cache space 0, sub-cache space 1, ..., sub-cache space 7, respectively. Sub-buffer space 0 is used to store type 0 messages, sub-buffer space 1 is used to store type 1 messages, sub-buffer space 2 is used to store type 2 messages, and so on. In other words, the second buffer space can store at most type 0 messages with a size of 64 MB, type 1 messages with a size of 64 MB, type 2 messages with a size of 64 MB, and so on.

In the foregoing embodiment, the capacity of the sub-buffer space of different types of messages is the same. In other embodiments, the sizes of different sub-buffer spaces may be different. Also take the second cache space with a total size of 512MB as an example. Assume that the second cache space includes 6 sub-cache spaces in total, which are sub-cache space 0 to sub-cache space 5, respectively. Sub-buffer space 0 to sub-buffer space 5 are respectively used for buffering type 0 to type 5 messages. The size of the sub-cache space 0 and the sub-cache space 1 may be 128MB; the size of each sub-cache space from the sub-cache space 2 to the sub-cache space 5 is 64MB.

The above-mentioned sub-buffer space is divided according to capacity. In other embodiments, the sub-buffer space may also be divided according to the maximum number of messages that can be buffered. Also take the foregoing sub-cache space 0 to sub-cache space 7 as an example. In some embodiments, the number of messages that can be buffered in each sub-buffer space is the same. For example, each of the 8 sub-buffer spaces can buffer 100 messages. In other embodiments, the number of messages that can be buffered in different sub-buffer spaces may be different. For example, each sub-cache space from sub-cache space 0 to sub-cache space 3 can buffer 200 messages, and each sub-cache space from sub-cache space 4 to sub-cache space 7 can buffer 100 messages.

In some embodiments, the type of the message may be determined according to the class of service (CoS) of the message. The Institute of Electrical and Electronics Engineers (IEEE) 802.1p defines CoS. IEEE802.1p stipulates a total of 8 CoS, which are 0 to 7. A CoS value of 0 has the lowest priority, and a CoS value of 7 has the highest priority. The network device can determine the CoS of the message according to the CoS field in the Layer 2 header of the received message and the virtual local area network (Virtual Local Area Network) identification (ID). For example, if the message is a message that carries the CoS value and VLAN ID, the CoS of the message is the CoS value carried in the message; if the message only carries the CoS value (the VLAN ID is 0 at this time) If the message is a message, the CoS of the message is the CoS value carried by the message; if the message is a message carrying a label, then the CoS of the message is 0 by default or a new CoS can be specified by changing it.

In some embodiments, the type of the message may correspond to the CoS of the message on a one-to-one basis. For example, the correspondence between the message type and the CoS of the message can be as shown in Table 1.

Table 1

类型type	CoSCoS
00	00
11	11
22	22
33	33
44	44
55	55
66	66
77	77

As shown in Table 1, if the CoS of the message is 0, the type of the message is 0; if the CoS of the message is 1, the type of the message is 1, and so on.

In other embodiments, the correspondence between the type of the message and the CoS of the message may also have multiple CoS corresponding to one type. For example, the correspondence between the message type and the CoS of the message can be as shown in Table 2.

Table 2

As shown in Table 2, the type of packets with CoS 0 and 1 is 0; the type of packets with CoS 2 and 3 is 1; the type of packets with CoS 4 and 5 is 2; CoS is 6 and 7 The type of the message is 3.

For another example, the type of packets with CoS of 0, 1 and 2 is 0; the type of packets with CoS of 3, 4, and 5 is 1, and the type of packets with CoS of 6 and 7 are 2.

In other embodiments, the type of the message may be determined according to the Internet Protocol (IP) priority of the message. The network device can determine the Type of Service (ToS) field in the IP packet header to determine the IP priority. Similar to CoS priority, the correspondence between IP priority and message type can be one-to-one correspondence, or multiple IP priorities can correspond to one type. For the sake of brevity, it will not be repeated here.

In addition, the type of the message can also be determined according to the priority (priority, Pri) field in the VLAN header of the message or the priority in the differentiated services code point (DSCP) field in the IP header. The specific determination method is the same as that of determining the type of the message according to the CoS. For the sake of brevity, it will not be repeated here.

In other embodiments, the type of the message can be divided according to other information received in the message. For example, the type of the message can be determined based on the lossy traffic message and the lossless traffic message. For another example, the type of the message can be determined according to IP version 4 (IPv4) and IP version 6 (IPv6).

In other embodiments, the message type can also be determined simultaneously based on priority information and other information (for example, lossy traffic messages and lossless traffic messages, or IPv4 messages and IPv6 messages).

The number of message types supported by different network devices can be different. For example, some network devices can support 8 types of messages, and other networks can support 4 types of messages.

The number of message types supported by the network device may be the same as the number of sub-buffer spaces included in the second buffer space. For example, if the number of message types that the network device can support is 8, then the second buffer space of the network device includes 8 sub-buffer spaces. The 8 sub-buffer spaces have a one-to-one correspondence with the 8 message types.

In some embodiments, the cache entity may include at least one first cache space and at least one second cache space. In other words, in some embodiments, the cache entity may include a first cache space and a second cache space. In other embodiments, the cache entity may include multiple first cache spaces and one second cache space. In other embodiments, the cache entity may include a first cache space and a plurality of second cache spaces.

In some embodiments, the second cache space included in the cache entity may have a one-to-one correspondence with the ingress port. In other words, the multiple second buffer spaces have a one-to-one correspondence with the multiple ingress ports. Each second buffer space user in the plurality of second buffer spaces buffers the message received through the corresponding port.

Also take the network device 110 shown in FIG. 2 as an example. The cache space of the network device 110 cache entity 142 may include three second cache spaces. The three second cache spaces may be referred to as the second cache space 111, the second cache space 112, and the second cache space 113, respectively. The second buffer space 111 corresponds to the port 111. The second buffer space 111 is used to buffer the message received from the port 111. The second buffer space 112 corresponds to the port 112. The second buffer space 112 is used to buffer the message received from the port 112. The second buffer space 113 corresponds to the port 113. The second buffer space 113 is used to buffer the message received from the port 113.

In other embodiments, the second buffer space has a one-to-one correspondence with the egress port. In other words, there is a one-to-one correspondence between the multiple second buffer spaces and the multiple egress ports. Each second buffer space in the plurality of second buffer spaces is used for buffering the message sent through the corresponding egress port. Take the network device 110 shown in FIG. 2 as an example. The cache entity 142 of the network device 110 may include two second cache spaces. The two second cache spaces may be referred to as the second cache space 114 and the second cache space 115, respectively. The second buffer space 114 corresponds to the port 114. The second buffer space 114 is used to buffer the message sent through the port 114. The second buffer space 115 corresponds to the port 115. The second buffer space 115 is used to buffer the message sent through the port 115.

In other embodiments, the second cache space in the cache entity may not correspond to the egress port or the ingress port. In other words, no matter which ingress port the message received by the network device comes from or which egress port is sent through, the second buffer space can buffer the message. For ease of description, this second buffer space that does not distinguish between the ingress port and the egress port may be referred to as a port-independent second buffer space. The foregoing second buffer space corresponding to the egress port or the ingress port may be referred to as a port-related second buffer space. Furthermore, the second cache space corresponding to the egress port may be referred to as the second cache space related to the egress port. The second cache space corresponding to the ingress port may be referred to as the second cache space related to the ingress port.

In some embodiments, the multiple second cache spaces in the cache entity may all be port-related second cache spaces. For example, the multiple second cache spaces in the cache entity are all second cache spaces related to the egress port. For another example, the multiple second cache spaces in the cache entity are all second cache spaces related to the ingress port. For another example, a part of the second cache space among the plurality of second cache spaces in the cache entity is the second cache space related to the ingress port, and the other part of the second cache space is the second cache space related to the egress port.

In other embodiments, the second cache space in the cache entity is a port-independent second cache space. For example, the cache entity includes a port-independent second cache space.

In other embodiments, the second cache space in the cache entity may have multiple port-related second cache spaces and at least one port-independent second cache space. For example, the cache entity may include multiple second cache spaces related to outbound ports and one second cache space independent of ports. For another example, the cache entity may include multiple second cache spaces related to ingress ports and one second cache space independent of ports. For another example, the cache entity may include multiple second cache spaces related to ingress ports, multiple second cache spaces related to egress ports, and one port-independent second swap space.

The cache entity may use the first cache space and the second cache space in a predetermined order.

The order of use of the first cache space and the second cache space can be configured by the user or can be set when the network device is shipped from the factory (in other words, the user cannot change the order of use).

In some embodiments, if the cache entity includes a first cache space, a port-independent second cache space, and multiple port-related second cache spaces, the cache entity may first use the port-independent second cache space , And then use the first buffer space, and finally use the port-related second buffer space.

In other embodiments, if the cache entity includes a first cache space and multiple port-related second cache spaces, the cache entity may first use the first cache space, and then use the port-related second cache space.

In other embodiments, if the cache entity includes a first cache space and a port-independent second cache space, the cache entity may first use the first cache space, and then use the port-independent second cache space.

It can be understood that the above-mentioned sequential use is the order in the case that the buffer space can also cache the corresponding type of messages. If a certain buffer space has no remaining buffer space to buffer the message, the buffer space is skipped. For example, if the first buffer space has no remaining space to buffer the message, the second buffer space is directly used to buffer the message.

It can be understood that using the second buffer space first and then the first buffer space refers to: first using the sub-buffer space in the second buffer space corresponding to a certain type of message to buffer the corresponding type of message. In the case that the sub-buffer space cannot buffer the corresponding type of message, the first buffer space is used to cache the corresponding type of message (assuming that the first buffer space has available capacity).

Using the first buffer space and then the second buffer space means: first use the first buffer space to buffer the message, if the first buffer space does not have enough available capacity to continue to buffer the message, then use the second buffer space and the message The sub-buffer space corresponding to the message type caches the message.

Taking the network 100 shown in FIG. 1 as an example, it is assumed that the packets in the network 100 shown in FIG. 1 may include lossless traffic packets and lossy traffic packets.

Assume that the network device 110 only includes three ingress ports and two egress ports as shown in FIG. 1. Then, the cache entity of the network device 110 may include three second cache spaces and one first cache space. The one first cache space may be referred to as a shared (share) cache space. The three second buffer spaces can be divided into reserve buffer spaces and headroom buffer spaces. The three second cache spaces may include one reserved cache space and three headroom cache spaces. The reserved buffer space is a port-independent second buffer space. The three headroom buffer spaces are the second buffer space related to the ingress port. In other words, each of the three headroom buffer spaces corresponds to an ingress port of the network device 110. The three headroom buffer spaces may be referred to as headroom buffer space 111, headroom buffer space 112, and headroom buffer space 113, respectively. The headroom cache space 111 corresponds to the ingress port 111, the headroom cache space 112 corresponds to the ingress port 112, and the headroom cache space 113 corresponds to the ingress port 113. The headroom buffer space 111 is used for buffering messages received from the port 111, the headroom buffer space 112 is used for buffering messages received from the port 112, and the headroom buffer space 113 is used for buffering messages received from the port 113.

For lossless traffic packets in the network 100, the predetermined order of use of the cache space of the cache entity of the network device 100 may be: reserved cache space, shared cache space, and clear cache space.

For lossy traffic packets in the network 100, the network device 100 may use the reserved cache space and the shared cache space in the cache entity to cache.

Fig. 3 is a schematic flowchart of a method for processing a message in a network device according to an embodiment of the present application. For ease of description, the packets received and processed by the network device in the method shown in FIG. 3 are all packets with lossless traffic.

301: The policy management entity receives cache status information 1 sent by the cache entity.

The cache status information 1 includes the used capacity of the target cache space at time 1.

The target buffer space can be used to buffer target messages. The target message is a message with a target type. The target type is any one of multiple message types. For example, if the type of the message includes 0 to 7 as shown in Table 1, then the target type can be any of types 0 to 7.

The target cache space includes a first cache space, a first target sub-cache space and a second target sub-cache space.

The first buffer space has the same function as the first buffer space described above. The first buffer space can be shared by different types of messages. In other words, as long as the first buffer space has available capacity, any type of message can be buffered in the first buffer space, and there is no upper limit for each type of message.

In some embodiments, the first cache space may be a shared cache space.

The target sub-cache space is one of the multiple sub-cache spaces included in the second cache space. The multiple sub-buffer spaces have a one-to-one correspondence with multiple message types supported by the network device. Each sub-buffer space is used to buffer corresponding types of messages.

The first target sub-cache space and the second target sub-cache space may be different sub-cache spaces in the second cache space. For example, the first target sub-cache space may be one of multiple sub-cache spaces included in the second cache space 1. For example, the second cache space 1 may be reserved cache space. The second target sub-cache space may be one of multiple sub-cache spaces included in the second cache space 2. For example, the second buffer space 2 may be a headroom buffer space.

Take the headroom buffer space as an example. For example, suppose that the network device supports a total of 4 message types, which can be called type 0, type 1, type 2 and type 3 respectively. Then the headroom cache space in the cache entity of the network device includes four sub-cache spaces, and the four sub-cache spaces may be referred to as sub-cache space 0, sub-cache space 1, sub-cache space 2 and sub-cache space 3, respectively. Sub-cache space 0 corresponds to type 0. Sub-buffer space 0 is used to buffer type 0 messages. Sub-cache space 1 corresponds to type 1. Sub-buffer space 1 is used to buffer type 1 messages. Sub-buffer space 2 corresponds to type 2 messages. The sub-buffer space 2 is used to buffer type 2 messages. The sub-buffer space 3 corresponds to type 3 messages. The sub-buffer space 3 is used to buffer type 3 messages. Assuming that the target message is of type 1, then the second target sub-buffer space is sub-buffer space 1. Assuming that the type of the target message is type 0, then the second target sub-cache space is sub-cache space 0.

The cache entity of the network device preferentially uses the second cache space 1 to cache the message. In the case that the second buffer space 1 cannot buffer the message, the first buffer space is used to buffer the message. In the case that the first cache space cannot cache the message, the cache entity uses the second cache space 2 to cache the message. In other words, the cache entity may first use the first target sub-cache space to cache the target message; in the case that the first target sub-cache space cannot continue to cache the target message, use the first cache space to cache the target message; In the case that the cache space cannot continue to cache the target message, the second target sub-cache space is used to cache the target message.

The cache entity of the network device can monitor the target cache space, determine the used capacity of the target cache space at time 1, and send the used capacity of the target cache space at time 1 to the policy management entity through the cache state 1. The used capacity of the target cache in the cache state 1 may include the used capacity of the first cache space at time 1, the used capacity of the first target sub-cache space at time 1, and the used capacity of the second target sub-cache space at time 1. Used capacity.

302. The policy management entity determines whether the target cache space meets the preset condition 1 according to the used capacity of the target cache space at time 1.

303. If the target cache space satisfies the preset condition 1, the policy management entity may determine the message processing strategy 1 and send the message processing strategy 1 to the message processing entity.

In some embodiments, the target cache space meeting the preset condition 1 may include: the first target sub-cache space cannot continue to cache the target message, the first cache space cannot continue to cache the target message, and the second target The sub-buffer space cannot continue to cache the target message.

In other words, the policy management entity may first determine whether the first target sub-cache space can continue to cache the target message. If the first target sub-cache space can continue to cache the target message, the policy management entity can determine that the target cache space does not meet the preset condition 1 and can continue to use the first target sub-cache space to cache the target message . If the first target sub-cache space cannot continue to cache the target message, the policy management entity may continue to determine whether the first cache space can continue to cache the target message. If the first cache space can continue to cache the target message, the policy management entity can determine that the target cache space does not satisfy the preset condition 1 and can continue to use the first cache space to cache the target message. If the first cache space cannot continue to cache the target message, the policy management entity may determine whether the second target sub-cache space can continue to cache the target message. If the second target sub-cache space can cache the target message, the policy management entity may determine that the target cache space does not meet the preset condition 1 and may continue to use the second target sub-cache space to cache the target message . If the second target sub-buffer space cannot cache the target message, the policy management entity may determine that the target buffer space satisfies the preset condition 1.

In other words, when the policy management entity determines that the first target sub-cache space, the first cache space and the second target sub-cache space cannot cache the target message at the same time, the policy management entity determines that the target cache space satisfies the first target packet. A preset condition.

In some embodiments, the inability of the first buffer space to buffer the target message may include: the first buffer space does not have sufficient available capacity for buffering the message. In other words, all the available capacity of the first buffer space has been occupied by the buffered messages.

In other embodiments, the inability of the first buffer space to buffer the target message may include: the used capacity in the first buffer space reaches the upper limit of the used capacity. In other words, an upper limit of used capacity can be set for the first cache space, for example, it can be referred to as a preset threshold 1. If the used capacity in the first buffer space reaches the preset threshold 1, the first buffer space may no longer continue to buffer packets. For example, the preset threshold value 1 can be set to 85%. In other words, if 85% of the buffer space of the first buffer space is already occupied by the message, then the first buffer space cannot continue to buffer the message. The remaining 15% of the capacity in the first cache space after reaching the upper limit of the used capacity can be activated under special circumstances. For example, it can be used to buffer burst messages or messages that cannot be discarded.

It can be understood that the 85% preset threshold value 1 in the foregoing embodiment is just an example. The preset threshold value 1 may also be a proportional value, such as 95%, 90%, or 80%. It is understandable that if the preset threshold value 1 is 100%, it means that the first buffer space cannot buffer the target message when all the available capacity of the first buffer space has been occupied.

In other embodiments, the preset threshold value 1 may be a buffer capacity value. For example, assuming that the total size of the first cache space is 2048MB, the preset threshold 1 may be 1800MB, 1900MB or 1700MB.

In other embodiments, the preset threshold 2 may also be the number of packets. For example, assuming that the first buffer space can buffer 1000 packets in total, the preset threshold 1 may be 800, 900, or 950.

For another example, the preset threshold value 1 in the foregoing embodiment is a preset value for the used capacity. In other embodiments, a lower limit of the available capacity can also be set according to the available capacity of the first cache space. If the available capacity of the first cache space reaches the lower limit of available capacity, the first cache space may not continue to cache messages (correspondingly, the available capacity of the target cache space reported by the cache entity at this time).

The upper limit of used capacity and the lower limit of available capacity of the aforementioned first cache space can be configured by the user, or set by the network device when it leaves the factory (in other words, the user cannot change the upper limit of used capacity and the lower limit of available capacity).

In other embodiments, the inability of the first cache space to cache the target message may also include that the first cache space fails to continue to cache the target message.

Similarly, in some embodiments, the inability of the first target sub-buffer space to cache the target message may include: the first target sub-buffer space does not have sufficient available capacity for buffering messages. In other words, all the available capacity of the first target sub-buffer space has been occupied by the cached messages.

In other embodiments, the inability of the first target sub-buffer space to cache the target message may include: the used capacity in the first target sub-buffer space reaches the upper limit of the used capacity. In other words, an upper limit of used capacity can be set for the first target sub-cache space, for example, it can be referred to as a preset threshold 2. If the used capacity in the first target sub-buffer space reaches the preset threshold 2, the first target sub-buffer space may no longer continue to buffer packets. For example, the preset threshold 2 can be set to 85%. In other words, if 85% of the buffer space of the first target sub-buffer space is already occupied by messages, then the first target sub-buffer space cannot continue to buffer messages. The remaining 15% of the capacity in the first target sub-cache space after reaching the upper limit of the used capacity can be activated under special circumstances. For example, it can be used to buffer burst messages or messages that cannot be discarded.

Similarly, in some embodiments, the inability of the second target sub-buffer space to cache the target message may include: the second target sub-buffer space does not have sufficient available capacity for buffering messages. In other words, all the available capacity of the second target sub-buffer space has been occupied by the cached messages.

In other embodiments, the inability of the second target sub-buffer space to cache the target message may include: the used capacity in the second target sub-buffer space reaches the upper limit of the used capacity. In other words, an upper limit of the used capacity can be set for the second target sub-cache space, for example, also the preset threshold 2. If the used capacity in the second target sub-buffer space reaches the preset threshold 2, the second target sub-buffer space may no longer continue to buffer packets. For example, the preset threshold 2 can be set to 85%. In other words, if 85% of the buffer space of the second target sub-buffer space is already occupied by the message, the second target sub-buffer space cannot continue to buffer the message. The remaining 15% of the capacity in the second target sub-cache space after reaching the upper limit of the used capacity can be activated under special circumstances. For example, it can be used to buffer burst messages or messages that cannot be discarded.

It can be understood that the 85% preset threshold 2 in the foregoing embodiment is just an example. The preset threshold 2 may also be a proportional value, such as 95%, 90%, or 80%. It can be understood that if the preset threshold 2 is 100%, it means that the target sub-cache space cannot cache the target message when the entire capacity of the target sub-cache space has been occupied.

In other embodiments, the preset threshold 2 may be a buffer capacity value. For example, assuming that the total size of the first target sub-cache space is 2048MB, then the preset threshold 2 may be 1800MB, 1900MB or 1700MB.

In other embodiments, the preset threshold 2 may also be the number of packets. For example, assuming that the first target sub-buffer space can buffer 1000 packets in total, the preset threshold 2 may be 800, 900, or 950.

For another example, the preset threshold 2 in the foregoing embodiment is a preset value for available capacity. In other embodiments, a lower limit of available capacity can also be set according to the available capacity of the target sub-cache space. If the available capacity of the target sub-cache space reaches the lower limit of the available capacity (correspondingly, the available capacity of the target cache space reported by the cache entity at this time), then the target sub-cache space may no longer continue to cache messages.

The upper limit of the used capacity and the lower limit of the available capacity of the above-mentioned first target sub-cache space and the second target sub-cache space can be configured by the user, or can be set by the network device when it leaves the factory (in other words, the user cannot change the The upper limit of available capacity and the lower limit of available capacity).

In addition, in the foregoing embodiment, the upper limit of the used capacity of the first cache space, the first target sub-cache space and the second target sub-cache space are the same. In other embodiments, the upper limit of the used capacity of the first cache space, the first target sub-cache space and the second target sub-cache space may also be different. For example, the upper limit of the used capacity of the first cache space is 85%, the upper limit of the used capacity of the first target sub-cache space is 90%, and the upper limit of the used capacity of the second target sub-cache space is 95%.

In other embodiments, the inability of the first target sub-cache space to cache the target message may also include that the first target sub-cache space fails to continue to cache the target message. Similarly, the inability of the second target sub-cache space to cache the target message may also include that the second target sub-cache space fails to continue to cache the target message.

In other embodiments, the policy management entity may estimate the time when the target buffer space is full and start a timer according to the transmission rate of the incoming port message and the available capacity of the target buffer space. If the timer reaches the estimated time and the transmission rate of the ingress port does not change during this period, the policy management entity can determine that the target buffer space meets the preset condition 1. If the transmission rate of the ingress port or the transmission rate of the egress port changes before the timer reaches the estimated time, the policy management entity can adjust the estimated time according to the changed transmission rate, and according to the adjusted estimate Time determines whether the target buffer space meets the preset condition 1.

In other embodiments, the policy management entity may determine whether the first target sub-cache space and the first cache space can cache the target message according to the used capacity of the first target sub-cache space and the used capacity of the first cache space. . If the first target sub-buffer space and the first buffer space cannot buffer the target message, the policy management entity may estimate the second target message according to the transmission rate of the ingress port receiving the target message and the size of the second target sub-buffer space. The time when the target sub-buffer space is full, and a timer is started. If the timer reaches the estimated time and the transmission rate of the target message received by the ingress port does not change, the policy management entity may determine that the second target buffer space meets the preset condition 1. If the transmission rate of the ingress port used to receive the target packet or the transmission rate of the egress port used to send the target packet changes before the timer reaches the estimated time, then the policy management entity can change according to the change The estimated time is adjusted for the subsequent transmission rate, and it is determined whether the second target buffer space satisfies the preset condition 1 according to the adjusted estimated time.

304: The message processing entity processes the newly received message according to the message processing strategy 1.

In some embodiments, the message processing strategy 1 may be to delete the newly received target message. For example, the message processing entity may process the newly received target message, change the message header information, add discard enable to the message header information, and send the modified message to the cache entity. In this way, the cache entity can delete the target message containing the discard enable. For another example, the message processing entity may directly discard the newly received target message.

In other embodiments, the message processing strategy 1 may be to modify the type of the newly received target message. In this way, the message processing entity can modify the type of the newly received target message. After the newly received target message is modified, it will be cached using the sub-buffer space corresponding to the modified type. For example, the target message is type 0, and after the type is modified, the type of the target message is type 1. In this way, the sub-cache space used for the cache type 1 in the second cache space can be used to cache the target message with the modified type.

In some embodiments, the modification to the type of the target message may be modified in the order of priority. The type priority of the target packet after modification is lower than the type before modification. Assume that the priority of type 0 is greater than the priority of type 1 than the priority of type 2. If the target message has a priority of type 0, the type of the target message can be modified to type 1.

The sub-buffer space corresponding to the re-determined message type has available capacity to buffer the reclassified message. The policy management entity may first determine the sub-buffer space with available capacity, and then determine that a corresponding type in these sub-buffer spaces is the type that needs to be modified for the target message, and the type that needs to be modified can be sent to as the message processing strategy Message processing entity. For example, assuming that the type of the target message is type 0, the sub-buffer space corresponding to type 1 can no longer continue to cache the message, and the sub-buffer space corresponding to type 2 and type 3 can still continue to buffer the message. In this case, the message processing strategy 1 determined by the policy management entity may be to modify the type of the newly received target message to type 2 or type 3. After the message processing entity newly receives the target message, it can modify the type of the target message to type 2 or type 3. In this way, the cache entity can continue to use the type 2 or type 3 sub-cache space to cache newly received messages.

In other embodiments, if the cache entity includes a plurality of second cache spaces and the plurality of second cache spaces have a one-to-one correspondence with a plurality of egress ports. Then, performing exception management on the newly received target message may be to modify the outgoing port used to send the newly received target message. As described above, each second buffer space in the buffer entity may be used to buffer the message sent through the corresponding out port. Therefore, it may happen that the sub-storage space used for buffering the target message in one second buffer space of the plurality of second buffer spaces is full but the sub-buffer used for buffering the target message in the other or more second buffer spaces The space can also continue to cache target messages. Therefore, the egress port used to send the target message can be modified, so that the target sub-buffer space in the other second buffer space can be used to cache the target message.

In the above technical solution, the cache management of the network device can enable the target message to not preempt the child corresponding to other types of messages in the unified second buffer space when the target sub-buffer space cannot continue to cache the target message. Cache space. If the network device also includes a port-independent second buffer space, the target message will not preempt the port-independent second buffer space corresponding to other types of sub-buffer spaces. In this way, it can be ensured that the network device has the remaining buffer space to use when other types of messages need to be buffered.

In some embodiments, the network device may also notify the upstream sending the target packet to stop sending the target packet to the network device when it is determined that the target buffer space satisfies the preset condition 1.

The message processing entity may also stop exception management of newly received target messages when the cache management returns to normal.

In some embodiments, the policy management entity may receive the cache status 2 sent by the cache entity. The cache state 2 carries the used capacity of the target cache at time 2. The policy management entity may determine whether the target cache space meets the preset condition 2 according to the used capacity of the target cache space at time 2. If the target cache space satisfies the preset condition 2, the policy management entity can determine the message processing strategy 2 and send the message processing strategy 2 to the message processing entity, where the message processing strategy 2 is different from the message processing strategy 1. In this case, the message processing entity implements message processing strategy 2. If the target buffer space does not meet the preset condition 2, the measurement management entity is not sure of a new message processing strategy. In this case, the message processing entity continues to execute message processing strategy 1.

In some embodiments, the policy management entity may first determine whether there are still cached target packets in the second target sub-cache space. If at least one target message is still cached in the second target sub-cache space, the policy management entity may determine that the target cache space does not satisfy the preset condition 2. For example, if the available capacity of the second target sub-cache space is equal to the total capacity of the second target sub-cache space, the policy management entity may determine that no target packet is cached in the second target sub-cache space. If the available capacity of the second target sub-cache space is less than the total capacity of the second target sub-cache space, the policy management entity may determine that at least one target message is cached in the second target sub-cache space.

In some embodiments, if there is no cached target message in the second target sub-cache space, the policy management entity determining whether the target cache space satisfies the preset condition 2 may include: determining the used capacity of the first cache space Is it less than the preset threshold 3; if the used capacity of the first cache space is less than the preset threshold 3, it can be determined that the target cache space satisfies the preset condition 2; if the used capacity of the first cache space is greater than or If it is equal to the preset threshold 3, it can be determined that the target cache space does not satisfy the preset condition 2.

In other embodiments, if there is no cached target message in the second target sub-cache space, the policy management entity determining whether the target cache space satisfies the preset condition 2 may include: determining the value of the first target sub-cache space Whether the used capacity is less than the preset threshold 4; if the used capacity of the first target sub-cache is less than the preset threshold 4, it can be determined that the target cache space satisfies the preset condition 2; if the first target sub-cache If the used capacity of the space is greater than or equal to the preset threshold 4, it can be determined that the target cache space does not satisfy the preset condition 2.

In other embodiments, if there is no cached target message in the second target sub-cache space, the policy management entity determining whether the target cache space satisfies the preset condition 2 may include: determining the used amount of the first cache space Is the capacity less than the preset threshold 3 and the used capacity of the first target sub-cache space is less than the preset threshold 4; if the used capacity of the first cache space is less than the preset threshold 3 and the used capacity of the first target sub-cache space If the used capacity is less than the preset threshold 4, it can be determined that the target cache space meets the preset condition 2; otherwise, it can be determined that the target cache space does not meet the preset condition 2.

Similar to the preset threshold value 1 and the preset threshold value 2, the preset threshold value 3 and the preset threshold value 4 may be a ratio value or a specific capacity value or the number of packets. The preset threshold 3 and the preset threshold 4 may be configured by the user or may be set when the network device leaves the factory.

When the network device sends the messages buffered in the buffer entity, they will be sent in the reverse order of the buffer order. For example, the second target sub-buffer space starts to buffer the target message after the first buffer space cannot continue to buffer the message. Therefore, the network device may first send the target message buffered in the second target sub-buffer space through the corresponding out port, then send the target message buffered in the first buffer space, and finally send the first target message. Target message in the sub-buffer space. Therefore, through the above solution, it can be achieved that the second target sub-buffer space has available capacity to buffer the new target message. In this way, if the target message needs to be cached, the second target sub-buffer space can be used to cache the target message.

In other embodiments, the policy management entity may determine whether the used capacity of the second target sub-cache space is less than a preset threshold of 5. If the used capacity of the second target sub-cache space is less than the preset threshold value 5, the policy management entity may determine that the target cache space meets the preset condition 2. If the used capacity of the second target sub-cache space is greater than or equal to the preset threshold value 5, the policy management entity may determine that the target cache space does not satisfy the preset condition 2.

In other embodiments, the policy management entity may start a timer after sending the message processing policy 1 to the message processing entity, if the timer reaches the preset time and is used to send the target message's egress port During this period, the target message has been sent, then the policy management entity can determine that the target buffer space satisfies the preset condition 2. The network device always sends the target message through the outgoing port. Therefore, after the message processing entity executes the message processing strategy 1, the target message cached in the target buffer space is always decreasing. After a preset time has elapsed, there is already available capacity in the target buffer space to continue to buffer the target message. In this case, the policy management entity may determine another message processing strategy (for example, message processing strategy 2), and send the newly determined message processing strategy to the message processing entity.

In some embodiments, the preset time may be determined according to the sending speed of the target message and the capacity of the second target sub-buffer space. For example, if the sending speed of the target message does not change, when the timer reaches the preset time, all target messages in the second target sub-buffer space have been sent. In other words, the preset time may be sufficient for the network device to send out the target message buffered in the second target sub-buffer space. For another example, in other embodiments, the preset time may cause the network device to send out most of the target message cached in the second target sub-cache space (for example, 80%, 90%, or 95%, etc.).

In some embodiments, the message processing strategy 2 may be to stop executing the message processing strategy 1. In this way, the message processing entity can stop executing the message processing strategy 1 after receiving the message processing strategy 2 (for example, not deleting the newly received target message). After receiving the message processing strategy 2, the message processing entity can use the default message processing strategy to process the newly received message or use the message processing strategy used before the message processing strategy 1 to process the newly received message. Arts.

The effect of the solution shown in FIG. 3 will be introduced below in conjunction with FIG. 4 and FIG. 5.

Fig. 4 is a schematic diagram of how to buffer the target message in the buffer space without adopting the method shown in Fig. 3.

As shown in FIG. 4, the network device 110 starts to receive type 1 packets from the upstream device 121 through the port 111 at time t0, and the transmission rate for receiving type 1 packets is 60 Gbps. The network device 110 starts to receive type 1 packets from the upstream device 122 through the port 111 at time t0, and the transmission rate for receiving type 1 packets is 60 Gps.

Assume that type 1 packets need to be sent through port 114 and the maximum transmission speed of port 114 is 100 Gbps. Then, in this case, the port 114 cannot send the type 1 packets received by the port 111 and the port 112 in time. In this case, the cache entity in the network device 110 may first cache the received type 1 message in a sub-cache space in the reserved cache space for caching type 1 messages (hereinafter referred to as type 1 reserved cache space) .

At t1, the reserved buffer space of type 1 is all occupied by type 1 packets. The cache entity can determine whether there is any available capacity in the shared cache space. If there is remaining space in the shared cache space, the cache entity may cache type 1 messages in the shared cache space.

At t2, the entire capacity of the shared buffer space is occupied by type 1 packets. In this case, the cache entity may use the sub-buffer space in the headroom cache space for caching type 1 messages (hereinafter referred to as type 1 headroom cache space) to cache type 1 messages.

At t3, the headroom buffer space of type 1 is all occupied by type 1 packets. In this case, the cache entity uses the sub-caching space for caching other types to cache type 1 messages. As shown in Figure 4, type 1 messages start to preempt the reserved buffer space and the sub-buffer space in the headroom buffer space for buffering other types of messages.

As shown in Figure 4, at t4, the reserved buffer space, shared buffer space, and headroom buffer space have all been occupied by type 1 messages. The network device 110 starts to receive type 2 packets from the upstream device 123 through the port 111 at time t4, and the transmission rate for receiving type 2 packets is 100 Gps. Assume that type 2 messages still need to be sent through port 114. It can be seen that at t4, the port 111 is still receiving Type 1 packets from the upstream device 111 and the upstream device 112 at a transmission speed of 60 Gps respectively. Therefore, type 2 messages cannot be sent from port 114 in time. In this case, the cache entity needs to cache type 2 messages, but since the reserved cache space and headroom cache space, the sub-buffer space that was used to cache type 2 messages has been preempted by type 1 messages, so the cache entity cannot Cache type 2 packets in time. This caused an abnormal resource management.

Fig. 5 is a schematic diagram of reserved cache space, shared cache space and clear cache space after using the method shown in Fig. 3. It is assumed that the cache entity periodically sends the available capacity of each sub-cache space in the reserved cache space, the available capacity of the shared cache space, and the available capacity of each sub-cache space in the headroom cache space to the policy management entity. The policy management entity determines the message processing strategy 1 and sends the message processing strategy 1 to the message processing entity. Assume that the message processing strategy 1 is to discard type 1 messages.

As shown in Figure 5, it is assumed that the message processing entity acquires and executes message processing strategy 1 and starts to execute message processing strategy 1 at time t3. In this way, the message processing entity discards the newly received type 1 message from time t3. In this case, since the egress port used to send type 1 has been sending type 1 packets, the type 1 packets stored in the headroom buffer space 111 continue to decrease. At time t5, there are no more type 1 messages in the headroom buffer space 111. And, starting from t5, the type 1 messages stored in the shared buffer space are also decreasing. Assume that at t6, the message processing entity obtains the message processing strategy 2 determined by the policy management entity and starts to execute the message processing strategy 2, where the message processing strategy 2 is to stop executing the message processing strategy 1. Therefore, starting from t6, the message processing entity can stop discarding newly received type 1 messages. In this case, since the speed at which the port 111 receives type 1 messages has not changed, the amount of type 1 messages buffered in the shared buffer space continues to increase.

Compared with FIG. 4, since the newly received type 1 message is discarded according to the state of the cache entity, the type 1 message no longer preempts the sub-buffer space of other types of messages. Therefore, type 2 messages received from time t4 can be cached in the sub-buffer space reserved for buffering type 2 in the reserved buffer space. In this way, it is possible to avoid the situation that the type 1 message preempts the sub-buffer space of other types of messages caused by the configuration error of the cache entity.

Fig. 6 is a schematic flowchart of a method for processing a message in a network device according to an embodiment of the present application. The method shown in FIG. 6 may be executed by the network device shown in FIG. 2. For example, the policy management entity in the method shown in FIG. 6 may be the policy management entity 141 shown in FIG. 2. The cache entity in the method shown in FIG. 6 may be the cache entity 142 shown in FIG. 2. The message processing entity in the method shown in FIG. 6 may be the message processing entity 143 shown in FIG. 2.

601: The policy management entity obtains first cache state information from the cache entity.

The first cache state information is used to indicate the first state of the cache entity at the first moment.

In some embodiments, the first state may be the state of the cache entity. In this case, the first state may be one of a plurality of states. For example, the multiple states include: an abnormal state and a normal state. In this case, the first state may be a normal state or an abnormal state. For another example, the multiple states may include: a normal state, a critical state, and an abnormal state. In this case, the first state may be a normal state, a critical state or an abnormal state.

In other embodiments, the first state may be related information of the cache entity. These related information can reflect the state of the cache entity. For example, the first state may include at least one of the following information: the length of the sending queue, the buffer occupancy of the sending queue, the delay of the sending queue, the available capacity of the buffer entity, the used capacity of the buffer entity, and the size of the buffer entity. The rate of the outgoing port, etc.

Multiple counters are set in the cache entity, and each counter corresponds to an egress port queue. The counter is used to record the number of packets of the corresponding outgoing port. According to the number of packets recorded by the counter, the length of the queue (which can be referred to as the sending queue) of each outgoing port can be obtained. The length of the sending queue can be the queue length of each outgoing port, the queue length of one or more specific outgoing ports (for example, the outgoing port with the longest queue or the outgoing port with the lowest outgoing port rate, etc.), the average queue length, or the total The length of the queue, etc., are not limited in the embodiment of the present application.

Take the network device shown in Figure 2 as an example. In some embodiments, the first state may include the queue length corresponding to port 114 and the queue length of port 115. In other embodiments, the first state may only include the queue length of the egress port with the longest length (for example, it may be the queue length of port 114). In other embodiments, the first state may be the sum of the queue length of port 114 and the queue length of port 115.

The rate of the outgoing port of the cache entity can be the rate of each outgoing port, or the rate of one or more specific outgoing ports (such as the outgoing port with the longest queue or the outgoing port with the lowest outgoing port rate, etc.), or the average Rate, etc., which are not limited in the embodiment of the present application

Take the network device shown in Figure 2 as an example. In some embodiments, the first state may include the rate corresponding to port 114 and the rate of port 115. In other embodiments, the first state may only include the rate of the egress port with the longest length (for example, it may be the queue length of port 114). In other embodiments, the first state may be the average of the port 114 rate and the port 115 rate.

The available capacity of the cache entity may be the total available capacity of the cache entity, or the available capacity of part of the cache space in the cache entity.

In some embodiments, the first state may include the available capacity of the first buffer space. In other embodiments, the first state may include the available capacity of each sub-cache space in the second cache space. In other embodiments, the first state may include the available capacity of one or more sub-cache spaces in the second cache space. For example, a sub-buffer space used to cache messages of the target type. In other embodiments, the first state may include the available capacity of the first cache space and the available capacity of the second cache space.

Similarly, the used capacity of the cache entity may be the total used capacity of the cache entity or the used capacity of part of the cache space in the cache entity.

It is also assumed that the cache space of the cache entity may include a first cache space and a second cache space. In some embodiments, the first state may include the used capacity of the first buffer space. In other embodiments, the first state may include the used capacity of each sub-cache space in the second cache space. In other embodiments, the first state may include the used capacity of one or more sub-cache spaces in the second cache space. For example, a sub-buffer space used to cache messages of the target type. In other embodiments, the first state may include the used capacity of the first cache space and the used capacity of the second cache space.

For example, in the embodiment shown in FIG. 3, the first state is the used capacity of the target cache space at time 1.

In some embodiments, the caching entity may determine the first state, and send first caching state information for indicating the first state to the policy management entity. Correspondingly, the policy management entity receives the first cache state information sent by the cache entity. In other words, the first cache state information is actively determined by the cache entity and sent to the policy management entity.

In some embodiments, the cache entity may periodically determine the state of the cache entity and indicate the determined state of the cache entity to the policy management entity through cache state information.

In other embodiments, the caching entity may send the first caching state information to the policy management entity when it is determined that the first state satisfies the first preset condition. In other words, the cache entity only indicates the state of the cache entity to the network device when the state meets the first preset condition. If the state of the cache entity does not meet the first preset condition, the cache entity may not need to indicate the state of the cache device to the policy management entity.

In some embodiments, the first preset condition is configured at the factory and cannot be modified by the user. In other embodiments, the user can set the first preset condition according to needs.

For example, the first preset condition is that the available capacity of the cache entity is less than or equal to the threshold Th1. In this case, if the cache entity determines that the available capacity of the cache entity is less than or equal to Th1, it may send the first cache state information to the policy management entity.

For another example, the first preset condition may be that the rate of the egress port reaches the maximum rate. In this case, if the cache entity determines that the rate of the outgoing port of the cache entity reaches the maximum rate, it may send the first cache state information to the policy management entity.

For another example, the first preset condition may be that the available capacity of the target sub-cache space is less than the threshold Th2. In this case, if the cache entity determines that the available capacity of the target sub-cache space of the cache entity is less than Th2, it may send the first cache state information to the policy management entity.

For another example, the first preset condition may be that the queue length of the egress port is greater than the threshold Th3. In this case, if the cache entity determines that the queue length of at least one of the port 114 or the port 115 is greater than the threshold Th3, it may send the first cache status information to the policy management entity.

In some embodiments, only one preset condition for triggering the reporting of the first cache state information may be stored in the cache entity.

In other embodiments, the cache entity may store a plurality of preset conditions for triggering the reporting of the first cache status information. As long as the first state satisfies one of the multiple preset conditions, the cache entity wants the policy management entity to send the first cache state information. Among the plurality of preset conditions, the preset condition satisfied by the first state is the first preset condition.

In other embodiments, the policy management entity or the message processing entity may send a status request to the cache entity. The caching entity may determine the first state and send the first caching state information used to indicate the first state to the policy management entity upon receiving the state request. The manner in which the policy management entity or the message processing entity sends the status request to the cache entity may be periodic or aperiodic. For example, the message processing entity may send the status request to the cache entity when the rate of the ingress port is greater than a threshold. For another example, the message processing entity may send the status request to the cache entity when a message with a specific mark or level is received.

For example, in the embodiment shown in FIG. 3, the first cache status information is sent by the cache entity to the policy management entity.

In other embodiments, the policy management entity may actively read the first cache state information in the cache entity. The cache entity may obtain the state of the cache entity, determine the cache state information according to the obtained state of the cache entity, and save the determined cache state information. For example, the cache entity may obtain the first state of the cache entity at the first moment, determine the first cache state information, and save the first cache state information. The policy management entity can read the first cache state information stored by the cache entity.

In other embodiments, the buffer entity may adjust the length of the sending queue, the buffer occupancy of the sending queue, the delay of the sending queue, the available capacity of the buffer entity, the used capacity of the buffer entity, the rate of the outgoing port of the buffer entity, etc. At least one type of information in is sent to the policy management entity. The policy management entity may determine the first cache state information according to the received at least one type of information. In this case, the first cache state information is the state of the cache entity.

In other embodiments, the buffer entity can obtain the length of the sending queue, the buffer occupancy of the sending queue, the delay of the sending queue, the available capacity of the buffer entity, the used capacity of the buffer entity, the rate of the outgoing port of the buffer entity, etc. At least one type of information in and save the acquired at least one type of information. The policy management entity may read at least one type of information stored in the cache entity, and determine the first cache state information according to the at least one type of information. In this case, the first cache state information is the state of the cache entity.

602. The policy management entity determines a first packet processing policy according to the acquired first cache state information.

Taking the method shown in FIG. 3 as an example, in the method shown in FIG. 3, if the cache state information 1 meets the following preset conditions: the first cache space cannot continue to cache the target message and the target sub-cache space cannot continue By caching the target message, the policy management entity can determine the message processing strategy, for example, discarding the newly received target message.

If the cache status information 1 does not meet the foregoing preset conditions, the policy management entity may not be sure of a new message processing policy.

In other embodiments, the policy management entity may maintain multiple message processing policies. Different message processing strategies can correspond to different cache status information. For example, message processing strategy 11 corresponds to the queue length of the egress port being less than the threshold Th5; message processing strategy 12 corresponds to the queue length of the egress port being greater than or equal to Th5 and less than the threshold Th6; message processing strategy 13 corresponds to the queue of the egress port The length is greater than or equal to Th6. In this case, if the length of the outbound port queue included in the first buffer status information is less than Th5, the policy management entity may determine that the first packet processing strategy is the packet processing strategy 11.

603. The policy management entity sends the first packet processing policy to the packet processing entity

604: The message processing entity executes the first message processing strategy, and processes the newly received message according to the first message processing strategy.

In some embodiments, the first message processing strategy may be to discard all newly received messages. In this case, the message processing entity can discard all newly received messages.

In other embodiments, the first message processing strategy may be to modify the message information of all newly received messages. Modifying the message information can be modifying the priority of the message, modifying the discarding priority of the message, modifying the port number of the message, modifying the explicit congestion notification (ECN) bit of the message, or modifying the discarding of the message One or more of the enable bits, etc. In this case, the message processing entity can modify the message information of all newly received messages.

In other embodiments, the first message processing strategy may be to discard some newly received messages. In this case, the message processing entity can discard some of the newly received messages. For example, in the embodiment shown in FIG. 3, only the newly received target message may be discarded. For another example, the message processing entity can randomly discard newly received messages.

In other embodiments, the first message processing strategy may be to modify the message information of a part of the newly received message. In this case, the message processing entity can modify the message information of some newly received messages. For example, in the embodiment shown in FIG. 3, only the message information of the target message may be modified. For another example, the message processing entity can randomly select a part of the messages and modify the message information of the selected messages.

In other embodiments, the first message processing strategy may be to discard part of the message while modifying the message information of another part of the message.

In other embodiments, the first message processing strategy may be to discard a part of the message, while modifying the message information of a part of the message, while not performing any processing on a part of the message.

605: The policy management entity obtains the second cache status information.

The second cache state information is used to indicate the second state of the cache entity at the second moment.

In some embodiments, the manner in which the policy management entity obtains the second cache state information is the same as the manner in which the first cache state information is obtained. For example, both the first cache state information and the second cache state information are actively sent by the cache entity to the policy management entity.

In other embodiments, the method of obtaining the second cache state information by the policy management entity may be different from the method of obtaining the first cache state information. For example, the first cache state information is actively sent by the cache entity to the policy management entity, and the second cache state information is read from the cache entity by the policy management entity.

The specific method for the policy management entity to obtain the second cache state information is similar to the specific method for the policy management entity to obtain the first cache state information. For the sake of brevity, I won't repeat them here.

The meaning of the second state is similar to that of the first state. For the sake of brevity, it will not be repeated here.

606. The policy management entity determines a second packet processing policy according to the second cache status information

607. The policy management entity sends the second packet processing policy to the packet processing entity.

608. The message processing entity executes the second message processing strategy, and processes the newly received message according to the second message processing strategy.

In some embodiments, the second packet processing strategy is used to instruct the packet processing entity to no longer execute the first packet processing strategy. For example, assuming that the first packet processing strategy is to discard packets of the target type, the packet processing strategy may stop discarding packets of the target type after receiving the second packet processing strategy.

In some embodiments, the second message processing strategy is also used to instruct the message processing entity to use the specified message processing strategy to process the newly received message.

For example, the designated message processing strategy may be a default message processing strategy. For example, the default message processing strategy is not to perform any additional processing (such as discarding or modifying message information) on the target type of message. In other words, the default message processing strategy is to process the target type messages and non-target type messages in the same way.

In another example, the designated message processing strategy may be a message processing strategy used before the first message processing strategy is used. For example, the message processing strategy used before the first message processing strategy is used is the third message processing strategy. The third message processing strategy is to modify the priority of the target type of message. In this way, after receiving the second message processing strategy, the message processing entity no longer discards the target type of message, but modifies the priority of the target type of message.

In some embodiments, the message in the method shown in FIG. 6 is a lossless traffic message. In other embodiments, the message in the method shown in FIG. 6 may be a lossless traffic message or a lossy traffic message.

According to the method shown in FIG. 6, the message processing entity can execute the corresponding message processing strategy according to the state of the cache entity. In this way, the cache entity can be better managed to reduce the occurrence of cache management exceptions in the cache entity.

For example, due to misconfiguration, the resource management rule in the cache entity is that the message input to the cache entity is discarded only when the queue length of port 115 is greater than 25535. However, the queue length of the port 115 in the resource management rules is greater than the queue length that the port 115 can actually handle (assuming that the queue length that the port 115 can actually handle is 1023). However, using the method shown in Figure 6, the policy management entity can obtain the queue length of port 115 reported by the cache entity, and determine the message processing strategy when the queue length of port 115 is greater than 1000 and send the determined message processing strategy to Message processing entity. It is assumed that the message processing strategy can be to modify the outgoing port of the message to be port 114. In this way, the message processing entity can modify the outgoing port of the packet from 115 to 114. In this way, it is possible to avoid the occurrence of an abnormal cache management caused by a configuration error of the cache entity.

For another example, in a lossless network, for lossless traffic packets, if the network device is congested on the outgoing port, it requests the upstream device to stop sending the packets sent through the outgoing port. However, the upstream device may not normally respond to the request sent by the network device and continue to send packets to the network device. According to the current industry processing method, since the received message is a lossless traffic message, the network device cannot discard the newly received message. However, network devices cannot send lossless traffic packets in time. In this way, there may not be enough buffer space to buffer the newly received message. Using the method shown in FIG. 6, the policy management entity can obtain the available capacity of the cache space reported by the cache entity. The policy management entity may determine a message processing strategy and send the determined message processing strategy to the message processing entity when the available capacity of the buffer space is insufficient. It is assumed that the message processing strategy can be to discard all newly received messages. In this way, the message processing entity can discard the newly received message. In this way, the message capacity that needs to be cached by the cache entity will not exceed the available capacity of the cache entity.

The cache entity 142 may be realized by a memory.

In some embodiments, different types of cache spaces (ie, the first cache space and the second cache space) in the cache entity 142 may be implemented by the same memory.

In other embodiments, different types of cache spaces in the cache entity 142 may be implemented by different memories. For example, the cache entity 142 may be realized by four memories. Memory 1 can be used as a first buffer space, memory 2 can be used as a second buffer space corresponding to port 111, memory 3 can be used as a second buffer space corresponding to port 112, and memory 4 can be used as a corresponding The second buffer space at port 113.

It can be understood that the memory in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memory. Among them, the non-volatile memory can be read-only memory (ROM), programmable read-only memory (programmable ROM, PROM), erasable programmable read-only memory (erasable PROM, EPROM), and electrically available Erase programmable read-only memory (electrically EPROM, EEPROM) or flash memory. The volatile memory may be random access memory (RAM), 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 (static RAM, SRAM), dynamic random access memory (dynamic RAM, DRAM), and synchronous dynamic random access memory (synchronous DRAM, SDRAM), double data rate synchronous dynamic random access memory (double data rate SDRAM, DDR SDRAM), enhanced synchronous dynamic random access memory (enhanced SDRAM, ESDRAM), synchronous connection dynamic random access memory (synchlink DRAM, SLDRAM) ) And direct memory bus random access memory (direct rambus RAM, DR RAM). It should be noted that the memories of the systems and methods described herein are intended to include, but are not limited to, these and any other suitable types of memories.

In some embodiments, the policy management entity 141 and the message processing entity 143 may be implemented by the same processor or system on chip (SoC). The processor may be a central processing unit (central processing until, CPU) or a network processor (network processor, NP).

In other embodiments, the policy management entity 141 and the message processing entity 143 may be implemented by different processors. For example, the message processing entity 143 may be a central processing until (CPU) or a network processor (NP), and the policy management entity 141 may be an application specific integrated circuit (ASIC). For another example, the message processing entity 143 may be implemented by an ASIC, and the policy management entity 141 may be implemented by another ASIC.

Fig. 7 is a structural block diagram of a network device provided according to an embodiment of the present application. The network device 700 shown in FIG. 7 includes a processor 701, a memory 702, a receiver 703, and a transmitter 704.

The processor 701, the memory 702, the receiver 703, and the transmitter 704 can communicate through a bus 705.

The processor 701 is the control center of the network device 700, and provides sequencing and processing facilities for executing instructions, executing interrupt actions, providing timing functions, and other functions. Optionally, the processor 701 includes one or more central processing units (CPUs). . Optionally, the network device 700 includes multiple processors. The processor 701 may be a single-core (single CPU) processor or a multi-core (multi-CPU) processor. The processor 701 may also be an application specific integrated circuit (ASIC), a system on chip (SoC), a network processor (NP), or a digital signal processing Circuit (digital signal processor, DSP), microcontroller (microcontroller unit, MCU), programmable logic device (PLD), other programmable logic devices, discrete gates, or transistor logic devices , Discrete hardware components, or other integrated chips, etc.

The program code executed by the processor 701 may be stored in the memory 702. The processor 701 controls the communication with peripheral devices by controlling the execution of other programs or processes, thereby controlling the operation of the network device 700, so as to implement the operation steps of the foregoing method.

The memory 702 may also be used to store messages from upstream devices.

The receiver 703 is used to receive a message from an upstream device. The transmitter 704 is configured to send the message stored in the memory to the downstream device.

In addition to the data bus, the bus 705 may also include a power bus, a control bus, a status signal bus, and the like. However, for clear description, various buses are marked as bus 705 in the figure.

Fig. 8 is a structural block diagram of a network device provided according to an embodiment of the present application. The network device 800 shown in FIG. 8 includes: a first processor 801, a second processor 802, a first memory 803, a second memory 804, a receiver 805, and a transmitter 806.

The receiver 805 and the transmitter 806 have the same implementation manners and functions as the receiver 703 and the transmitter 704 in the network device 700 shown in FIG. 7, and will not be repeated here for brevity.

The first processor 801 may be used to implement the function of the policy management entity in the foregoing embodiment.

The second processor 802 may be used to implement the function of the message processing entity in the foregoing embodiment.

The first memory 803 may be used to store program codes executed by the first processor 801 and the second processor 802.

The second memory 804 may be used to buffer messages from the upstream device.

The embodiment of the present application also provides a chip, which includes a logic circuit and an input/output interface. The logic circuit may be coupled with the memory to execute instructions and/or codes in the memory to realize the functions performed by the policy management entity in the foregoing embodiments.

The embodiment of the present application also provides a chip, which includes a logic circuit and an input/output interface. The logic circuit may be coupled with the memory for executing instructions and/or codes in the memory, and the chip may perform the functions performed by the message processing entity in the foregoing embodiments.

The embodiment of the present application also provides a chip, which includes a logic circuit and an input/output interface. The logic circuit may be coupled with the memory to execute instructions and/or codes in the memory, and the chip may perform the functions performed by the policy management entity and the message processing entity in the foregoing embodiments.

The aforementioned chip can be a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on chip (SoC), or a central processing unit. The central processor unit (CPU) can also be a network processor (NP), a digital signal processing circuit (digital signal processor, DSP), or a microcontroller (microcontroller unit, MCU) It may also be a programmable logic device (PLD), other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, or other integrated chips.

The embodiment of the present application also provides a computer-readable storage medium on which an instruction is stored, and the method in the foregoing method embodiment is executed when the instruction is executed.

An embodiment of the present application also provides a computer-readable storage medium on which an instruction is stored, and when the instruction is executed, the steps performed by the policy management entity in the foregoing method embodiment are executed.

The embodiment of the present application also provides a computer-readable storage medium on which an instruction is stored, and when the instruction is executed, the steps performed by the message processing entity in the foregoing method embodiment are executed.

As a form of this embodiment, a computer program product containing instructions is provided, and when the instructions are executed, the method in the foregoing method embodiment is executed.

As a form of this embodiment, a computer program product containing instructions is provided, and when the instructions are executed, the steps performed by the policy management entity in the foregoing method embodiments are executed.

As a form of this embodiment, a computer program product containing instructions is provided, and when the instructions are executed, the steps performed by the message processing entity in the foregoing method embodiments are executed.

A person of ordinary skill in the art may realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be implemented by electronic hardware or a combination of computer software and electronic hardware. Whether these functions are executed by hardware or software depends on the specific application and design constraint conditions of the technical solution. Professionals and technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered beyond the scope of this application.

Those skilled in the art can clearly understand that, for the convenience and conciseness of description, the specific working process of the system, device and unit described above can refer to the corresponding process in the foregoing method embodiment, which will not be repeated here.

In the several embodiments provided in this application, it should be understood that the disclosed system, device, and method can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division, and there may be other divisions in actual implementation, for example, multiple units or components may be combined or It can be integrated into another system, or some features can be ignored or not implemented. In addition, the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, and may be in electrical, mechanical or other forms.

The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the objectives of the solutions of the embodiments.

In addition, the functional units in the various embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit.

If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solution of the present application essentially or the part that contributes to the existing technology or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including Several instructions are used to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk and other media that can store program code .

The above are only specific implementations of this application, but the protection scope of this application is not limited to this. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in this application. Should be covered within the scope of protection of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

Claims

A method for processing packets in a network device, wherein the network device includes a policy management entity, a cache entity, and a packet processing entity, and the method includes:

Acquiring, by the policy management entity, first cache state information, where the first cache state information indicates the first state of the cache entity at the first moment;

Determining, by the policy management entity, a first packet processing strategy according to the first cache status information, and sending the first packet processing strategy to the packet processing entity;

The message processing entity processes the newly received message according to the first message processing policy.
The method according to claim 1, wherein before the policy management entity obtains the first cache status information, the method further comprises:

The policy management entity obtains at least one of the following information of the buffer entity at the first moment: the length of the sending queue, the buffer occupancy of the sending queue, the delay of the sending queue, the length of the buffer entity Available capacity, the used capacity of the cache entity, and the rate of the outgoing port of the cache entity;

The policy management entity determines the first cache state information according to the acquired at least one type of information.
The method of claim 1, wherein the method further comprises:

Determining the first cache state information by the cache entity;

The obtaining of the first cache state information by the policy management entity includes:

The policy management entity obtains the first cache state information from the cache entity.
The method of claim 3, wherein the policy management entity acquiring the first cache state information from the cache entity comprises:

The policy management entity receives the first cache state information sent by the cache entity.
The method according to claim 4, wherein before the cache entity sends the first cache status information to the policy management entity, the method further comprises:

The caching entity determines that the first state satisfies a first preset condition.
The method according to any one of claims 1 to 5, wherein the first state includes normal or abnormal.
The method according to any one of claims 1 to 5, wherein the first state includes at least one of the following: the length of the sending queue, the buffer occupancy of the sending queue, and the time of the sending queue. Extension, the available capacity of the cache entity, the used capacity of the cache entity, and the rate of the outgoing port of the cache entity.
The method according to any one of claims 1 to 7, characterized in that, before the policy management entity determines a first packet processing policy according to the first cache status information, the method further comprises:

The policy management entity determines that the first cache state information satisfies a second preset condition.
The method according to any one of claims 1 to 5, wherein the cache entity includes a target cache space, and the first state includes a usage situation of the target cache space at the first moment;

The policy management entity determines the first message processing policy according to the first cache status information, including:

The policy management entity determines the first message processing policy according to the usage of the target cache space at the first moment.
The method according to any one of claims 1 to 9, wherein the method further comprises:

Acquiring, by the policy management entity, second cache state information, where the second cache state information indicates the second state of the cache entity at the second moment;

Determining, by the policy management entity, a second packet processing strategy according to the second cache state information, and sending the second packet processing strategy to the packet processing entity;

The message processing entity processes the newly received message according to the second message processing strategy.
The method according to claim 10, wherein the second message processing strategy is used to instruct the message processing entity to no longer execute the first message processing strategy.
The method according to claim 11, wherein the second message processing strategy is further used to instruct the message processing entity to use a specified message processing strategy to process newly received messages.
The method according to claim 12, wherein the designated message processing strategy is a default message processing strategy or a message processing strategy used before the first message processing strategy is used.
The method according to any one of claims 1 to 13, wherein the newly received message comprises a lossless traffic message.
The method according to any one of claims 1 to 14, wherein the message processing entity processes the newly received message according to the first message processing strategy, comprising:

The message processing entity adopts one or more of the following processing methods:

Discard all the newly received messages;

Discarding part of the newly received message;

Modify the message information of all the newly received messages; or

Modify some of the message information of the newly received message.
The method according to claim 15, wherein modifying the message information of the newly received message comprises:

Modify one or more of the following information of the newly received message:

The priority of the newly received message;

The discard enable bit of the newly received message;

The discarding priority of the newly received message;

The port number of the newly received message; or

The display congestion flag of the newly received message.
A network device, characterized in that the network device includes a policy management entity, a cache entity, and a message processing entity, and the method includes:

The policy management entity is configured to obtain first cache state information, where the first cache state information indicates the first state of the cache entity at a first moment;

The policy management entity is configured to determine a first packet processing strategy according to the first cache status information, and send the first packet processing strategy to the packet processing entity;

The message processing entity is configured to process the newly received message according to the first message processing strategy.
The network device of claim 17, wherein:

The policy management entity is further configured to obtain at least one of the following information of the buffer entity at the first moment: the length of the sending queue, the buffer occupation of the sending queue, the delay of the sending queue, the The available capacity of the cache entity, the used capacity of the cache entity, and the rate of the outgoing port of the cache entity;

The policy management entity is specifically configured to determine the first cache state information according to the acquired at least one type of information.
The network device according to claim 17, wherein the cache entity is further used to determine the first cache state information;

The policy management entity is specifically configured to obtain the first cache state information from the cache entity.
The network device according to claim 19, wherein the policy management entity is specifically configured to receive the first cache state information sent by the cache entity.
The network device according to claim 20, wherein the cache entity is further configured to determine that the first state satisfies a first preset before sending the first cache state information to the policy management entity condition.
The network device according to any one of claims 17 to 21, wherein the first state includes normal or abnormal.
The network device according to any one of claims 17 to 21, wherein the first state includes at least one of the following: the length of the sending queue, the buffer occupancy of the sending queue, and the size of the sending queue. Delay, the available capacity of the cache entity, the used capacity of the cache entity, and the rate of the outgoing port of the cache entity.
The network device according to any one of claims 17 to 23, wherein the policy management entity is further configured to determine before determining the first packet processing policy according to the first cache status information The first cache state information satisfies a second preset condition.
The network device according to any one of claims 17 to 21, wherein the cache entity includes a target cache space, and the first state includes a usage situation of the target cache space at the first moment;

The policy management entity is specifically configured to determine the first packet processing policy according to the usage of the target cache space at the first moment.
The network device according to any one of claims 17 to 25, wherein:

The policy management entity is further configured to obtain second cache state information, where the second cache state information indicates the second state of the cache entity at the second moment;

The policy management entity is further configured to determine a second packet processing strategy according to the second cache state information, and send the second packet processing strategy to the packet processing entity;

The message processing entity is further configured to process the newly received message according to the second message processing strategy.
The network device according to claim 26, wherein the second packet processing strategy is used to instruct the packet processing entity to no longer execute the first packet processing strategy.
The network device according to claim 27, wherein the second message processing strategy is further used to instruct the message processing entity to use a specified message processing strategy to process newly received messages.
The network device according to claim 28, wherein the designated message processing strategy is a default message processing strategy or a message processing strategy used before the first message processing strategy is used.
The network device according to any one of claims 17 to 29, wherein the newly received message comprises a lossless traffic message.
The network device according to any one of claims 17 to 30, wherein the message processing entity is specifically configured to use one or more of the following processing methods to process a newly received message:

Discard all the newly received messages;

Discarding part of the newly received message;

Modify the message information of all the newly received messages; or

Modify some of the message information of the newly received message.
The network device according to claim 31, wherein the message processing entity is specifically configured to modify one or more of the following information of the newly received message:

The priority of the newly received message;

The discard enable bit of the newly received message;

The discarding priority of the newly received message;

The port number of the newly received message; or

The display congestion flag of the newly received message.
A computer-readable storage medium, wherein the computer-readable storage medium stores instructions for the method according to any one of claims 1 to 16.