CN114513440A

CN114513440A - Message transmission method, sending end and network equipment

Info

Publication number: CN114513440A
Application number: CN202011183729.9A
Authority: CN
Inventors: 徐安民; 王闯; 于德雷
Original assignee: Huawei Technologies Co Ltd
Current assignee: Huawei Technologies Co Ltd
Priority date: 2020-10-29
Filing date: 2020-10-29
Publication date: 2022-05-17

Abstract

The application provides a message transmission method, a sending end and network equipment, wherein the method comprises the following steps: the network equipment receives a first detection message of a first data stream sent by a sending end, wherein the first detection message carries a first identifier corresponding to the first data stream; the network device determines whether to add first information to the first detection packet according to a first quantity, where the first information is used to indicate that the sending end is prohibited from sending data packets of the first data stream, and the first quantity is a quantity of data packets currently stored in the first cache. By sending the detection message corresponding to the data stream, the message transmission method, the sending end and the network device can limit the number of concurrent data streams simultaneously rushing to the output port of one network device before the data stream starts to be transmitted, so that network congestion is avoided, and the transmission efficiency of the data stream is improved.

Description

Message transmission method, sending end and network equipment

Technical Field

The present application relates to the field of computer networks, and more particularly, to a message transmission method, a sending end and a network device.

Background

With the rise of more and more compute-intensive applications, such as Data mining, machine learning, artificial intelligence, and the like, the development of Data Center Networks (DCNs) is receiving more and more attention from people. Different applications of DCNs require data streaming, requiring a large amount of continuous throughput, predictable latency, and low CPU resource consumption by different hosts of DCNs. The main reason affecting DCN host streaming performance is the limited buffers of the switches. When a plurality of service flows simultaneously burst to an egress port of the same switch, the cache of the switch is easily overflowed, which may cause a large amount of packet loss and Retransmission Timeout (RTO), thereby greatly affecting the completion Time of the application.

Data Center Transmission Control Protocol (DCTCP) uses Explicit Congestion Notification (ECN) to detect and respond to network Congestion, i.e. when the egress queue length of a switch exceeds a threshold, a newly arriving packet is marked. In DCTCP, the sender maintains an estimator for each data stream to measure the congestion level of the current network. When the receiving end receives a marked data packet, the receiving end feeds back the information to the sending end in time through the ACK packet, so that the sending end can accurately know the congestion condition in the network, and a large amount of packet loss and RTO (real time offset) caused by the fact that the length of an outlet queue of the switch exceeds a threshold value are avoided.

However, the traditional TCP/IP protocol stack has high CPU overhead, and Remote Direct Memory Access (RDMA) allows a user-mode application program to directly read and write a Remote Memory without kernel intervention and Memory copy, and has low CPU resource consumption and low system delay. An RDMA over switched Ethernet (RoCE) protocol using remote direct memory access through Ethernet is an RDMA protocol operating on Ethernet, and has a very broad application prospect in DCN. RoCE itself has no congestion control algorithm and therefore needs to be coordinated with other schemes. Data Center Quantized Congestion Notification (DCQCN) is a Congestion control algorithm matched with a RoCE, a transmitting end sends an eccn-enabled RoCE message, a switch (Congestion Point, CP) marks the RoCE message through the ECN, and a receiving end (Notification Point, NP) returns Congestion Notification to the transmitting end (reading Point, RP) according to the marked message. At this time, the sending end adjusts the sending rate according to the congestion notification. In order to avoid packet loss, the switch also needs to set Priority-based Flow Control (PFC), and when the queue length reaches the upper limit, the switch sends Pause (Pause) information to all flows in the queue to stop sending them.

The flow control scheme in the prior art can only limit the flow after the transmission starts, and no corresponding detection and management mechanism exists in the stage before the transmission starts, so that an opportunity is generated for the connection of uncontrolled and sudden large flows, and the problem that a plurality of service flows rush to the outlet port of the same switch at the same time due to the large connections sent instantly is caused, and the application efficiency is further influenced.

Disclosure of Invention

The application provides a message transmission method, a sending end and network equipment, which can limit the number of concurrent flows simultaneously rushing to a port, further avoid the problems of RTO, Pause and the like caused by the Incast problem, and improve the transmission efficiency of DCN application.

In a first aspect, a method is provided, comprising: the method is executed by network equipment, wherein the network equipment is provided with a first cache and a second cache, wherein the first cache is used for storing data messages, and the second cache is used for storing detection messages; the network equipment receives a first detection message sent by a sending end, wherein the first detection message carries a first identifier corresponding to a first data stream; the network device determines a first number, where the first number is the number of data packets currently stored in the first cache, and may be understood as the number of data packets stored in the first cache when the network device receives a first probe packet sent by a sending end; the network device determines whether to add first information to the first detection message according to the first quantity, wherein the first information is used for indicating that a sending end is prohibited from sending a data message of the first data flow.

In the above technical solution, the network device determines whether to add the first information to the detection packet according to the first quantity by sending the detection packet before sending the data packet, so as to limit the number of concurrent flows simultaneously rushing to one port.

In a possible implementation manner, the determining, by the network device, whether to add the first information to the first probe packet according to the first quantity includes: the network device updates a first value according to the first number, an initial value of the first value is determined by a second number when the data packet cached by the first cache is 0, the second number is the number of the maximum concurrent data streams supported by the output port of the network device, and the second number is periodically updated. And the network equipment determines whether to add the first information in the first detection message according to the updated first value.

In the above technical solution, the network device determines the first value by calculating the number of the maximum concurrent data flows supported by the output port, and determines whether to add the first information to the first detection packet by determining the first value, so as to limit the number of concurrent flows simultaneously rushing to one port.

In one possible implementation, the network device updating the first value according to the first number includes: the network device determines a current second quantity according to the first quantity, and if the current second quantity is smaller than or equal to a current first value, the network device updates the first value to the current second quantity, where the current second quantity can be understood as the quantity of the maximum concurrent data streams supported by the output port of the network device when the network device receives the first probe packet sent by the sending end.

In the above technical solution, the number of the maximum concurrent data flows supported by the egress port represented by the second number is related to the number of the data packets stored in the first cache represented by the first number, and the second number is periodically updated, and when the second number is updated to be smaller than the first value, the first value is updated to the second number, so that the first value is more accurate.

In a possible implementation manner, after the network device determines whether to add the first information to the first probe packet according to the first value, the network device sends the first probe packet or the first probe packet to which the first information is added to the receiving end, and subtracts 1 from the first value.

In a possible implementation manner, after the network device subtracts 1 from the first value, the method further includes: if the network device determines that the first value is smaller than or equal to a preset first threshold and the first number is smaller than or equal to a second threshold, the network device updates the first value to a second number corresponding to the first time at the first time, and a preset time interval is formed between the first time and the time when the network device receives the first detection message.

In the above technical solution, by determining that the first number is less than or equal to the second threshold, it can be determined whether the current network device is in a light load state, and by updating that the first value is equal to the second number after the first time period, data packet transmission caused by misdetermination of the first value can be avoided, and congestion caused by a network device port is avoided.

In a possible implementation manner, after the network device sends the first probe packet or the first probe packet added with the first information to the receiving end, the method further includes: the network equipment receives a feedback message, wherein the feedback message is used for indicating that the receiving end receives the first detection message or the first detection message added with the first information; if the network equipment sends the first detection message added with the first information to a receiving end, the feedback message carries the first information; and the network equipment sends the feedback message to the sending end.

In the above technical solution, the receiving end notifies the sending end whether the first detection message is added with the first information or not in a form of a feedback message, so that the sending end knows a concurrent flow condition of the output port of the current network device, and the sending end can determine whether to send a data message corresponding to the first detection message or not according to the feedback message.

In a possible implementation manner, after the network device sends the feedback packet carrying the first information to the sending end, the first value is incremented by 1, because at this time, the data packet corresponding to the first detection packet does not start transmission.

In a possible implementation manner, after the first detection packet carries first priority information and the network device sends the feedback packet carrying the first information to the sending end, the method further includes: the network receives a second detection message sent by the sending end, wherein the second detection message carries second priority information and the first identifier corresponding to the first data stream, and the priority indicated by the second priority information is higher than the priority indicated by the first priority information; judging that the second priority information meets a second preset condition, wherein the second preset condition is that the priority indicated by the second priority information is greater than a second threshold; the network device does not judge the first value, and the network device sends the second detection message to a receiving end.

In a second aspect, a method for transmitting a packet is provided, where the method is performed by a sending end, and includes: the sending end sends a first detection message to the network equipment, wherein the first detection message carries first priority information and a first identifier corresponding to a first data stream; the sending end receives a feedback message; judging whether the feedback message carries first information, wherein the first information is used for indicating that the sending end is forbidden to send the data message of the first data stream; if the feedback message does not carry the first information, the sending end sends a data message of the first data stream; if the feedback message carries the first information, the sending end sends a second detection message, the second detection message carries second priority information and the first identifier corresponding to the first data stream, and the priority indicated by the second priority information is higher than the priority indicated by the first priority information.

In the technical scheme, the sending end determines to send the second detection message or the data message corresponding to the first data flow according to the receiving condition of the feedback message, and the sending end can judge whether to start transmission of the data flow by sending the detection message before sending the data message, so that network congestion caused by excessive concurrent flows simultaneously rushing to one port in a network is avoided.

When the detection message of the first data flow is limited for multiple times, the priority of the detection message of the first data flow is increased, and when the priority meets a second preset condition, the network equipment does not judge the first value, directly sends the detection message corresponding to the first data flow, and avoids the limitation of one detection message for multiple times.

In a third aspect, a network device for transmitting a packet is provided, where the network device includes a receiving unit, a determining unit, and a processing unit, where the receiving unit includes a first cache unit and a second cache unit, the first cache unit is used to store a data packet, and the second cache unit is used to store a detection packet; the receiving unit is used for receiving a first detection message sent by a sending end, wherein the first detection message carries a first identifier corresponding to a first data stream; the determining unit is used for determining a first quantity, wherein the first quantity represents the quantity of the data messages stored in the first cache unit; the processing unit is configured to determine whether to add first information to the first probe packet according to the first quantity, where the first information is used to indicate that the sending end is prohibited from sending a data packet of the first data stream.

In a possible implementation manner of the third aspect, the determining unit is specifically configured to update a first value according to the first number, where an initial value of the first value is determined according to a second number when the data packet cached in the first cache is 0, and the second number is the number of the maximum concurrent data streams supported by the egress port of the network device, where the second number is periodically updated; the determining unit determines whether to add the first information in the first detection message according to the updated first value.

In a possible implementation manner of the third aspect, the determining unit is configured to determine the current second number according to the first number; if the current second number is less than or equal to the current first value, it is determined to update the first value to the current second number.

In a possible implementation manner of the third aspect, after the determining unit determines whether to add the first information to the first probe packet according to the first value, the sending unit is further configured to send the first probe packet or the first probe packet after adding the first information to the receiving end; the sending unit is configured to send a first probe packet or the first probe packet added with the first information to a receiving end, and the processing unit is further configured to subtract 1 from the first value.

In a possible implementation manner of the third aspect, the processing unit is further configured to, after subtracting 1 from the first value, determine that the first value is smaller than a first threshold and the first number is smaller than or equal to a second threshold; if the determining unit determines that the first value is smaller than the first threshold and the first number is smaller than or equal to the second threshold, the processing unit is further configured to update the first value to the second number corresponding to the first time at the first time, where a preset time interval is provided between the first time and the time when the network device receives the first detection packet.

In a possible implementation manner of the third aspect, the receiving unit is further configured to: receiving a feedback message; if the sending unit sends the first detection message added with the first information, the feedback message carries the first information; if the sending unit sends the first detection message, the feedback message is used for indicating the receiving end to receive the first detection message or the first detection message added with the first information; the sending unit is specifically configured to send the feedback packet to the sending end.

In a possible implementation manner of the third aspect, after the receiving unit receives the feedback packet carrying the first information sent by the receiving end, the processing unit is specifically configured to update the first value plus 1;

in a possible implementation manner of the third aspect, the determining unit is further configured to: determining that the first value meets a first preset condition; when the first number is smaller than a first threshold, the determining unit determines that the first value is equal to the second number after a first period of time.

In a possible implementation manner of the third aspect, the sending unit sends the feedback packet to the sending end, where the feedback packet carries the first information, and the receiving unit is further configured to: receiving a second detection message sent by the sending end, where the second detection message carries second priority information and the first identifier corresponding to the first data stream, and the priority indicated by the second priority information is higher than the priority indicated by the first priority information; judging that the second priority information meets a second preset condition, wherein the second preset condition is that the priority indicated by the second priority information is greater than a second threshold; the sending unit is specifically configured to send the second probe packet.

In a fourth aspect, a sending end for transmitting a packet is provided, where the sending end includes: a sending unit, configured to send a first detection packet, where the first detection packet carries first priority information and a first identifier corresponding to a first data stream; a receiving unit, configured to receive a feedback packet; a determining unit, configured to determine whether the feedback packet carries first information, where the first information is used to indicate that the sending end is prohibited from sending a data packet of the first data stream; if the feedback message does not carry the first information, the sending unit is specifically configured to send a data message of the first data stream; if the feedback packet carries the first information, the sending unit is specifically configured to send a second detection packet, where the second detection packet carries second priority information and the first identifier corresponding to the first data stream, and a priority indicated by the second priority information is higher than a priority indicated by the first priority information.

Drawings

Fig. 1 is a schematic diagram of a network.

Fig. 2 is a schematic flow chart of a method for transmitting a message according to an embodiment of the present application.

Fig. 3 is a schematic flow chart diagram of a specific embodiment of a method for transmitting messages according to the present application.

Fig. 4 is a schematic flow chart diagram of yet another embodiment of a method for transmitting messages according to the present application.

Fig. 5 is a schematic block diagram of a transmitting end according to an embodiment of the present application.

Fig. 6 is a schematic block diagram of a network device according to an embodiment of the present application.

Fig. 7 is a schematic structural diagram of a transmitting end according to an embodiment of the present application.

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

Fig. 9 is an IP datagram header format.

Detailed Description

The technical solution in the present application will be described below with reference to the accompanying drawings.

The network device in the embodiment of the present application may be a network device (e.g., a router) having a routing function or a network device (e.g., a switch) having a switching function. The network device in the embodiment of the present application may be a network device in a wired communication network, and may also be a core network device in a wireless communication network (for example, a Global System of Mobile communication (GSM) System, a Code Division Multiple Access (CDMA) System, a Long Term Evolution (LTE) System, a future 5G network, and the like). The sending end and the receiving end of the application can be terminal equipment, such as a computer, a mobile phone, a tablet computer and the like.

In the embodiments of the present application, the words "exemplary," "for example," and the like are used to mean serving as an example, instance, or illustration. Any embodiment or design described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments or designs. Rather, the term using examples is intended to present concepts in a concrete fashion.

In the embodiments of the present application, "corresponding" and "corresponding" may be sometimes used in a mixed manner, and it should be noted that the intended meaning is consistent when the difference is not emphasized.

In the examples of the present application, the subscripts are sometimes as W₁It may be mistaken for a non-subscripted form such as W1, whose intended meaning is consistent when the distinction is de-emphasized.

The network architecture and the service scenario described in the embodiment of the present application are for more clearly illustrating the technical solution of the embodiment of the present application, and do not form a limitation on the technical solution provided in the embodiment of the present application, and as a person of ordinary skill in the art knows that along with the evolution of the network architecture and the appearance of a new service scenario, the technical solution provided in the embodiment of the present application is also applicable to similar technical problems.

Reference throughout this specification to "one embodiment" or "some embodiments," or the like, means that a particular feature, structure, or characteristic described in connection with the embodiment is included in one or more embodiments of the present application. Thus, appearances of the phrases "in one embodiment," "in some embodiments," "in other embodiments," or the like, in various places throughout this specification are not necessarily all referring to the same embodiment, but rather "one or more but not all embodiments" unless specifically stated otherwise. The terms "comprising," "including," "having," and variations thereof mean "including, but not limited to," unless expressly specified otherwise.

In the present application, "at least one" means one or more, "a plurality" means two or more. "and/or" describes the association relationship of the associated objects, meaning that there may be three relationships, e.g., a and/or B, which may mean: a exists alone, A and B exist simultaneously, and B exists alone, wherein A and B can be singular or plural. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship. "at least one of the following" or similar expressions refer to any combination of these items, including any combination of the singular or plural items. For example, at least one (one) of a, b, or c, may represent: a, b, c, a-b, a-c, b-c, or a-b-c, wherein a, b, c may be single or multiple.

Fig. 1 is a schematic diagram of a network. As shown in fig. 1, network 100 includes network device 110, upstream device 121, upstream device 122, upstream device 123, upstream device 124, upstream device 125, downstream device 131, downstream device 132, downstream device 133, downstream device 134, and downstream device 135. The upstream device may be a sender or a network device. Similarly, the downstream device may be a receiving end or a network device.

As shown in fig. 1, in the network 100, packets (packets) from the upstream device 121, the upstream device 122, and the upstream device 123 may be received through the port 111 of the network device 110. Similarly, messages from upstream devices 124 and 115 are received through

ports

112 and 113. The message is sent to downstream device 131 and downstream device 132 through port 114 of network device 110. Similarly, messages are sent through port 115 to downstream device 133, downstream device 134, and downstream device 135.

It is to be understood that fig. 1 is only a schematic diagram of a network for helping those skilled in the art understand the method of the present application, and is not a limitation to the network to which the technical solution of the present application can be applied. For example, in some embodiments, a packet sent from one or more upstream devices other than upstream device 121 and upstream device 122 may also be received through port 111 of network device 110. For another example, a message sent from one or more upstream devices may also be received through another port of the network device 110. As another example, the message may also be sent to one or more downstream devices other than downstream device 131 and downstream device 132 through port 114 of network device 110. As another example, the message may also be sent to one or more downstream devices through another port of the network device 110.

Ports (e.g., port 111, port 112, and port 113) of the network device for receiving messages from an upstream device may be referred to as ingress ports (or ingress ports) of the network device, and ports (e.g., port 114 and port 115) for sending messages to a downstream device may be referred to as egress ports (or egress ports) of the network device.

In a network, such as network 100 shown in fig. 1, a packet sent by an upstream device is received through an ingress port of a network device (e.g., network device 110), and is sent to a corresponding downstream device through an egress port. In some cases, when messages of multiple input ports simultaneously rush to one output port, congestion is easily caused at the output port, which causes a large amount of packet loss and RTO, and further greatly affects the completion time of the application.

The present application provides a message transmission method, which can be applied to a network as shown in fig. 1. The method can limit the number of concurrent data streams simultaneously rushing to an output port of the network equipment before starting message transmission, further avoid the problems of RTO, Pause and the like caused by the fact that messages of a plurality of input ports simultaneously rush to an output port, and improve the transmission efficiency of DCN application.

Fig. 2 is a schematic flow chart of a message transmission method according to an embodiment of the present application. FIG. 2 includes four steps S210-S240, which are described in detail below.

S210, the sending end sends a detection message a to the network equipment, and the network equipment receives the detection message a.

Specifically, when a data stream of a sending end needs to be transmitted, the sending end may select a packet in the data stream and mark the packet as a detection packet a, where the detection packet a carries an identifier corresponding to the data stream. For example, for TCP, the probe packet a may be 1 Syn packet, a flag field is set at a head of the Syn packet, a characteristic of the flag packet is the probe packet, and the network device can determine, according to a received number of the probe packets, a number of data streams that need to be currently started to be transmitted.

In a possible implementation manner, as shown in fig. 9, a sending end of an IP datagram header format sets a value of a field of an IP packet header to indicate identification information of a probe packet. For example, the sending end sets the value of the option field as a special value to determine that the type of the IP packet is a probe packet.

It should be understood that when the sending end sends the probe packet a, a timer needs to be set in order to handle packet loss. And after the timer is overtime, the sending end sends the detection message b corresponding to the data flow, and the time of the timer can be set to be 1.5-2 RTTs.

It should be understood that the embodiment of the present application does not limit the specific content and form of the probe packet. For example, the measurement packet may be a packet predefined by the sending end, the network device, and the receiving end and having a specific form or content, and the network device may determine that the received packet is a probe packet according to the form or content of the received packet.

S220, the network device is configured with a first cache and a second cache, where the first cache is used to receive the data packet sent by the sending end, the second cache is used to receive the detection packet, after receiving the detection packet a, the network device stores the detection packet a in the second cache and determines whether to add the first information to the detection packet a according to the number (first number) of the data packets stored in the first cache, and the network device sends the detection packet a or the detection packet a to which the first information is added to the receiving end.

Specifically, after receiving the probe packet a, the network device stores the probe packet a in the second cache. Before forwarding the detection message a, the network device determines whether to add first information in the detection message a by judging the size of the first value, wherein the first information is used for indicating that a sending end prohibits sending a data message of a data stream corresponding to the detection message a. When the network equipment judges that the first value is less than or equal to 0, adding first information into the detection message a, and sending the detection message a added with the first information to a receiving end by the network equipment; if the first value is larger than 0, the network equipment sends the detection message a to a receiving end. After receiving the detection message a, the receiving end determines to send different feedback messages by judging whether the detection message a carries the first information, so that the sending end can know whether to send the data message corresponding to the data stream. The detection message is processed by the network equipment and the processing result is fed back to the sending end, so that the detection effect in the stage before the transmission of the data message is started is achieved, and the problem of time delay caused by packet loss or data stream pause due to the fact that a large amount of data streams rush to an output port of the network equipment in the data stream starting stage is solved.

It should be understood that the first buffer may be a first queue, the second buffer may be a second queue, and the second queue may be a priority queue, where the priority queue implements buffering of probe packets with priority information, and the first-come probe packets are stored in the queue and dequeued in order of priority from high to low.

In one possible implementation, at t₀At the moment, after the network equipment receives the detection message a, a first value is determined, the calculation and the update of the first value are related to a second number, and the second number is t₀The Number of Maximum Concurrent Streams (MNCS) supported by an egress port of the network device at a time, the second Number being determined inversely related to a first Number, the first Number being indicated at t₀The number of data packets stored in the first buffer of the network device at that time, that is, the length of the first queue. The second number may be calculated according to the following formula:

MNCS＝C×(q_max-que)

wherein C is a gain coefficient, and 0<C<1，q_maxDenotes the maximum value of the preset first queue length, i.e. the maximum value of the first number, que denotes t₀A first number of moments, and que<q_max。

When t is₀First number of times que>q_maxWhen the MNCS value is 0.

In embodiments of the application, the value of the MNCS is updated periodically, for example, the update period of the MNCS may be 30 μ s. The MNCS calculation formula can show that the number of the maximum concurrent data flows supported by the network equipment output port changes along with the change of the number of the data messages stored in the first cache of the network equipment, and meanwhile, the number of the maximum concurrent data flows supported by the network equipment output port with the first value is calculated, so that the calculation of the first value can be more reasonable, the network equipment can process the detection message a more accurately, the sending end is informed whether to send the data message of the data flow corresponding to the detection message a, and a large number of data flows are prevented from rushing to one output port of the network equipment at the data flow starting stage.

Specifically, determining the first value from the MNCS includes three cases: the first case is to determine an initial value of the first value; when the data stream is just started, the network device only stores the detection message in the second cache, or the first quantity in the current first cache is 0, and the first value is equal to the value of the second quantity; in the second case, when the MNCS is updated after a period, if the value of the MNCS calculated according to the formula is smaller than the first value, the first value is made equal to the value of the MNCS; in a third case, when the first number is smaller than the first threshold, the first threshold represents a minimum value of the preset first number, which represents that the output port of the current network device is under light load, and the first value needs to be updated immediately to be equal to the value of the MNCS, thereby avoiding waste of network resources. After receiving the detection message a sent by the sending end, the network device judges whether to add the first information in the first detection message according to the first value obtained by the condition, and sends the first detection message or the first detection message added with the first information to the receiving end.

In a possible implementation manner, when the network device forwards the probe packet a or the probe packet a added with the aforementioned first information to the receiving end, the first value is decreased by 1. And indicating the change of the number of the data streams allowed to be started by the current network equipment outlet port through the change of the first value. When the network device forwards the detection message a or the detection message a added with the first information and the first value is reduced to meet the first threshold, the network device needs to wait for a first time period and then update the first value according to the current value of the MNCS. When the first value is reduced to meet the first preset condition, the feedback message is not fed back to the sending end, at the moment, the value of the MNCS is large, and if the threshold value is updated according to the value of the MNCS, the network equipment can easily misjudge the first value when the first data flow is just started. For example, the network device sets a timer, and waits for the timer to count for a first period of time. The first Time period is 3 rounds of average Round-Trip Time (RTT).

In another possible implementation manner, after receiving the detection message a, the network device needs to determine the priority of the detection message a, and determines to directly forward the detection message a or determine the first value according to the priority of the detection message a. The priority of the message may be determined according to the Class of Service (CoS) of the message. CoS is defined by the Institute of Electrical and Electronics Engineers (IEEE) 802.1 p. Ieee802.1p specifies 8 CoS in total, 0 to 7, respectively. The CoS value is 0 and the priority is the lowest, and the CoS value is 7 and the priority is the highest. The Network device may determine a CoS of the received packet according to a CoS field and an identification (identification, ID) of a Virtual Local Area Network (Virtual Local Area Network) in a header of a two-layer packet. For example, if the packet is a packet carrying a CoS value and a VLAN ID, the CoS of the packet is the CoS value carried by the packet; if the message only carries the CoS value (at this time, the VLAN ID is 0), the CoS of the message is the CoS value carried by the message; if the message is a message carrying a label, the CoS of the message is defaulted to 0 or a new CoS is designated by changing. The priority of the message may correspond to the CoS of the message one-to-one. For example, the correspondence between the type of the packet and the CoS of the packet may be as shown in table 1.

TABLE 1

Priority level	CoS
		0	0
1	1
		2	2
3	3
		4	4
5	5
		6	6
7	7

As shown in table 1, if the CoS of a message is 0, the priority of the message is 0; if the CoS of the message is 1, the priority of the message is 1, and so on. Taking the first-time sending of the detection message by the sending end as an example, after the network device receives the detection message for the first time, it is determined that CoS of the detection message is 0, that is, the priority of the detection message is 0.

In an embodiment of the application, if the priority of a detection message b of one data stream sent by a sending end does not satisfy a first preset condition, a network device determines whether to add first information to the detection message b according to a first value, and forwards the detection message b, and the network device receives a feedback message. And the sending end judges whether to continuously send the detection message c corresponding to the data stream according to the receiving condition of the feedback message, wherein the priority of the detection message c is higher than that of the detection message b. And when the network equipment judges that the priority of the detection message c is higher than the preset value, the detection message c is directly forwarded, and the network equipment receives a feedback message sent by a receiving end without judging a threshold value and informs the data flow of starting. The priority of the detection message c is higher than the preset value, which indicates that the detection message corresponding to the data flow is limited for multiple times. For example, the preset value of the priority of the probe message in the network device is 2, the priority of the probe message b sent by the sending end is 0, the network device notifies the sending end to send the probe message c after judging that the probe message b is limited by the first value, at this time, the priority of the probe message c is 1, the network device continues to notify the sending end after judging that the probe message c is limited by the updated first value, the sending end sends the probe message d, at this time, the priority of the probe message d is 2, the network device judges that the priority of the probe message d meets the preset value, and at this time, the network device directly forwards the probe message d. The network device can avoid the detection message corresponding to one data stream from being limited for many times by judging the priority information of the detection message.

In another embodiment of the present application, a network device receives a detection message b corresponding to a data stream sent by a sending end, where the detection message b is a first detection message corresponding to the data stream sent by the sending end, the network device determines whether to mark the detection message according to a first value, and simultaneously feeds back a marking result to the sending end, the sending end sends a detection message c corresponding to the data stream, and after receiving the detection message c, the network device determines that the detection message c is not a first detection message corresponding to the data stream, and then the network device directly forwards the detection message.

S230, a receiving end receives a detection message a sent by the network equipment or the detection message a added with the first information;

the network device receives a first detection message sent by a sending end, and if the forwarding of the first detection message meets the first value condition, or the priority of the first detection message meets a second preset condition, the network device directly forwards the first detection message. The receiving end receives the detection message a or the detection message a added with the first information.

In a possible implementation manner, as shown in fig. 3, in a case where the network device adds the first information in the embodiment of the present application, or as shown in fig. 4, in a case where the network device does not add the first information in the embodiment of the present application, when the network device forwards the probe packet a or the probe packet a to which the foregoing first information is added to the receiving end, the first value is decreased by 1.

S240, the receiving end sends the feedback message to the sending end through the network equipment.

Specifically, the receiving end may send the feedback packet in various situations. By way of example, the following three cases:

the first condition is as follows: the receiving end receives a detection message a sent by the network equipment, the detection message a does not carry first information, and the receiving end sends a feedback message to the sending end, wherein the feedback message is an ACK message. And the sending end starts the first data stream after receiving the feedback message, namely, sends a data message corresponding to the first data stream.

Case two: the receiving end receives a detection message a sent by the network equipment, wherein the detection message a carries first information, and the receiving end sends a feedback message to the sending end, and the feedback message carries the first information.

In a possible implementation manner, as shown in fig. 4, the receiving end sends a feedback packet to the network device, where the feedback packet carries first information, and the network device forwards the feedback packet, where the first value is added by 1, because at this time, the data packet corresponding to the first detection packet does not start transmission.

And the sending end determines to continue to send the detection message b corresponding to the first data stream according to the receiving condition of the reflection message until the sending end receives a normal ACK message and confirms that the first data stream is started.

Case three: and the sending end does not receive the feedback message when the feedback timer expires, and then resends the detection message a until receiving the feedback message of the detection message a, and confirms that the first data flow is started. For congestion control after normal startup of the first data flow, reference is made to the prior art, and details are not described here.

The above describes a method for transmitting a packet according to an embodiment of the present application with reference to fig. 2 to 4. The following describes a transmitting end and a network device according to an embodiment of the present application with reference to fig. 5 to 8.

Fig. 5 is a schematic block diagram of a transmitting end 500 according to an embodiment of the present application. As shown in fig. 5, the transmitting end 500 includes: a transmitting unit 510, a receiving unit 520 and a determining unit 530.

A sending unit 510, configured to send a detection packet to a network device, where the detection packet is used for the network device to determine the number of data streams that need to be started to be transmitted currently;

a receiving unit 520, configured to receive a feedback packet sent by the network device, where the feedback packet is a feedback of the measurement packet by the receiving end.

The determining unit 530 is configured to determine to send a data packet or a detection packet according to a receiving result of the feedback packet.

It should be understood that each unit in the transmitting end 500 may be used to perform each action or process in the transmitting end in the method shown in fig. 2 or the transmitting end in the methods shown in fig. 3 and 4, respectively. Here, detailed description thereof is omitted in order to avoid redundancy.

Fig. 6 is a schematic block diagram of a network device 600 according to an embodiment of the present application. As shown in fig. 6, the receiving end 600 includes: a receiving unit 610, a determining unit 620, a processing unit 630 and a transmitting unit 640.

A receiving unit 610, configured to receive a detection packet sent by a sending end;

a determining unit 620 for determining a first value;

a processing unit 630, configured to process the probe packet according to the first value, and add first information to the probe packet when the first value is less than or equal to 0, where the first information is used to indicate that a sending end is prohibited from sending a data packet of the first data stream;

a sending unit 640, configured to send the probe packet or the probe packet with the first information added.

It should be understood that the units in the network device 600 may be respectively used for executing the actions or processes in the network device in the methods shown in fig. 2, fig. 3 and fig. 4. Here, detailed description thereof is omitted in order to avoid redundancy.

Fig. 7 shows a schematic block diagram of a transmitting end 700 according to an embodiment of the present application. As shown in fig. 7, the transmitting end 700 includes: transceiver 710, processor 720, and memory 730. The transceiver 710, the processor 720 and the memory 730 communicate with each other via internal connection paths to transfer control and/or data signals.

A transceiver 710, configured to send a probe packet to a network device, where the probe packet is used for the network device to determine the number of data streams that need to start transmission currently;

and a processor 720, configured to determine, according to a reception result of a feedback packet, to send a data packet or a detection packet, where the feedback packet is a feedback of the measurement packet by the receiving end.

It will be appreciated that the processor 720 may be adapted to perform the above-described method and to implement the functionality of the execution body of the method, e.g. the sender, when the processor 720 invokes and runs the computer program from the memory.

Fig. 8 shows a schematic block diagram of a network device 800 according to an embodiment of the application. As shown in fig. 8, the network device 800 includes: a transceiver 810, a processor 820, and a memory 830. The transceiver 810, the processor 820 and the memory 830 communicate with each other via internal connection paths to transmit control and/or data signals.

A transceiver 810, configured to receive a measurement packet sent by a sending end;

the transceiver 810 is further configured to send a feedback message for feeding back the detection message to the sending end according to the feedback message, where the feedback message is a feedback of the measurement message by the receiving end.

It is to be understood that the processor 820 may be adapted to perform the above-described method and implement the function of an executing body of the method, such as a network device, when the processor 820 invokes and runs the computer program from the memory.

The embodiment of the application can be applied to or realized by a processor. The processor may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the above method embodiments may be performed by integrated logic circuits of hardware in a processor or instructions in the form of software. The processor may be a Central Processing Unit (CPU), other general purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic device, or discrete hardware component. The various methods, steps, and logic blocks disclosed in the embodiments of the present application may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of the method disclosed in connection with the embodiments of the present application may be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software in the decoding processor. The software may be in ram, flash, rom, prom, or eprom, registers, among other storage media as is well known in the art. The storage medium is located in a memory, and a processor reads information in the memory and completes the steps of the method in combination with hardware of the processor.

It will also be appreciated that the memory in the embodiments of the subject application can be either volatile memory or nonvolatile memory, or can include both volatile and nonvolatile memory. The non-volatile memory may be a read-only memory (ROM), a Programmable ROM (PROM), an Erasable PROM (EPROM), an electrically Erasable EPROM (EEPROM), or a flash memory. Volatile memory can be Random Access Memory (RAM), which acts as external cache memory. By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), Synchronous Dynamic Random Access Memory (SDRAM), double data rate SDRAM, enhanced SDRAM, SLDRAM, Synchronous Link DRAM (SLDRAM), and Direct Rambus RAM (DRRAM). It should be noted that the memory of the systems and methods described herein is intended to comprise, without being limited to, these and any other suitable types of memory.

It should be understood that, in the various embodiments of the present application, the sequence numbers of the above-mentioned processes do not mean the execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.

It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.

In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.

The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

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

The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a read-only memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.

The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present application, and shall be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A method for message transmission, wherein the method is performed by a network device, and the network device is configured with a first cache and a second cache, and the first cache is used for storing data messages and the second cache is used for storing probe messages, and the method comprises:

the network equipment receives a first detection message sent by a sending end, wherein the first detection message carries a first identifier corresponding to a first data stream;

the network device determines whether to add first information to the first detection packet according to a first quantity, where the first information is used to indicate that the sending end is prohibited from sending data packets of the first data stream, and the first quantity is a quantity of data packets currently stored in the first cache.

2. The method of claim 1, wherein the network device determining whether to add the first information to the first probe packet according to the first number comprises:

the network device updates a first value according to the first quantity, an initial value of the first value is determined according to a second quantity when the data packet stored in the first cache is 0, the second quantity is the quantity of the maximum concurrent data streams supported by an egress port of the network device, and the second quantity is periodically updated;

and the network equipment determines whether to add first information in the first detection message according to the updated first value.

3. The method of claim 2, wherein the network device updating the first value based on the first number comprises:

the network equipment determines a current second quantity according to the first quantity;

if the current second number is less than or equal to a current first value, the network device updates the first value to the current second number.

4. The method according to any of claims 1 to 3, wherein after the network device determines whether to add the first information in the first probe packet according to the first number, the method further comprises:

and the network equipment sends the first detection message or the first detection message added with the first information to a receiving end, and subtracts 1 from the first value.

5. The method of claim 4, wherein after subtracting 1 from the first value, the method further comprises:

if the network device determines that the first value is smaller than or equal to a preset first threshold and the first number is smaller than or equal to a second threshold, the network device updates the first value to a second number corresponding to the first time at a first time, and a preset time interval is formed between the first time and the time when the network device receives the first detection message.

6. The method of claim 4, wherein after subtracting 1 from the first value, the method further comprises:

the network equipment receives a feedback message sent by a receiving end, wherein the feedback message is used for indicating that the receiving end receives the first detection message or the first detection message added with the first information;

the network device sends the feedback message to the sending end, wherein,

if the feedback message is used to indicate that the receiving end receives the first detection message, the feedback message carries the first information,

and if the feedback message is used for indicating the receiving end to receive the first detection message, the feedback message does not carry the first information.

7. The method according to claim 6, wherein if a feedback packet sent by a network device to the sending end carries the first information, the method further comprises:

the network device adds 1 to the first value.

8. The method according to claim 6 or 7, wherein the first probe packet further carries first priority information, and

after the network device sends the feedback packet carrying the first information to the sending end, the method further includes:

the network device receives a second detection message sent by the sending end, wherein the second detection message carries second priority information and the first identifier, and the priority indicated by the second priority information is higher than the priority indicated by the first priority information;

and if the priority indicated by the second priority information is greater than a third threshold, the network equipment sends the second detection message to the receiving end.

9. A method for transmitting a packet, the method being performed by a transmitting end and comprising:

the sending end sends a first detection message to the network equipment, wherein the first detection message carries first priority information and a first identifier corresponding to a first data stream;

the sending end receives a feedback message, wherein the feedback message is the feedback of the receiving result of the first detection message by the receiving end;

judging whether the feedback message carries first information, wherein the first information is used for indicating that the sending end is forbidden to send the data message of the first data stream;

if the feedback message does not carry the first information, the sending end sends the data message of the first data flow,

if the feedback message carries the first information, the sending end sends a second detection message, the second detection message carries second priority information and the first identifier corresponding to the first data stream, and the priority represented by the second priority information is higher than the priority represented by the first priority information.

10. The network equipment is characterized by comprising a receiving unit, a determining unit and a processing unit, wherein the receiving unit comprises a first cache unit and a second cache unit, the first cache unit is used for storing data messages, and the second cache unit is used for storing detection messages;

the receiving unit is configured to receive a first detection packet sent by a sending end, where the first detection packet carries a first identifier corresponding to a first data stream;

the determining unit is configured to determine a first number, where the first number indicates the number of the data packets stored in the first cache unit;

the processing unit is configured to determine whether to add first information to the first probe packet according to the first quantity, where the first information is used to indicate that the sending end is prohibited from sending the data packet of the first data flow.

11. The network device of claim 10, wherein the determining unit is further configured to:

updating a first value according to the first number, where an initial value of the first value is determined according to a second number when the data packets stored in the first cache are 0, where the second number is the number of maximum concurrent data streams supported by an egress port of the network device, and the second number is periodically updated;

the determining unit determines whether to add first information to the first detection packet according to the updated first value.

12. The network device according to claim 11, wherein the determining unit is configured to, according to the first number, specifically:

the determining unit determines the current second number according to the first number;

the determining updates the first value to the current second number if the current second number is less than or equal to a current first value.

13. The network device according to any of claims 10 to 12, wherein after the processing unit is configured to determine whether to add first information in the first probe packet according to the first number, the sending unit is further configured to:

sending the first detection message or the first detection message added with the first information to a receiving end;

the sending unit is configured to send the first probe packet to a receiving end or subtract 1 from the first value after the first probe packet is added with the first information.

14. The network device of claim 13, wherein the processing unit is further configured to, after subtracting 1 from the first value, the determining unit is further configured to:

the determining unit determines that the first value is less than a first threshold and the first number is less than or equal to a second threshold;

if the determining unit determines that the first value is smaller than a first threshold and the first number is smaller than or equal to a second threshold, the processing unit is further configured to update the first value to a second number corresponding to a first time at the first time, where a preset time interval is provided between the first time and a time when the network device receives the first detection packet.

15. The network device of claim 14, wherein the processing unit is configured to subtract 1 from the first value, and wherein the receiving unit is further configured to:

receiving a feedback message sent by a receiving end, wherein the feedback message is used for indicating that the receiving end receives the first detection message or the first detection message added with the first information;

after the receiving unit receives the feedback message sent by the receiving end, the sending unit is further configured to send the feedback message to the sending end, wherein,

if the feedback message is used to indicate that the receiving end receives the first detection message added, the feedback message carries the first information,

and if the feedback message is used for indicating that the receiving end receives the first detection message, the feedback message does not carry the first information.

16. The network device according to claim 15, wherein the sending unit is configured to send the feedback packet carrying the first information, and after the sending unit is configured to send the feedback packet carrying the first information, the processing unit is further configured to add 1 to the first value.

17. The network device according to claim 15 or 16, wherein the first probe packet further carries first priority information, and

the sending unit is configured to send the feedback packet to the sending end, where after the feedback packet carries the first information:

the receiving unit is further configured to receive a second detection packet sent by the sending end, where the second detection packet carries second priority information and the first identifier corresponding to the first data stream, and a priority indicated by the second priority information is higher than a priority indicated by the first priority information;

the determining unit is further configured to determine that the priority indicated by the second priority information is greater than a third threshold, and the network device sends the second probe packet to a receiving end.

18. A transmitting end, comprising:

a sending unit, configured to send a first detection packet, where the first detection packet carries first priority information and a first identifier corresponding to a first data stream;

a receiving unit, configured to receive a feedback packet, where the feedback packet is a feedback of a receiving result of the first detection packet by the receiving end;

a determining unit, configured to determine whether the feedback packet carries first information, where the first information is used to indicate that the sending end is prohibited from sending a data packet of the first data stream;

if the feedback packet does not carry the first information, the sending unit is specifically configured to send a data packet of the first data flow,

if the feedback packet carries the first information, the sending unit is specifically configured to send a second detection packet, where the second detection packet carries second priority information and the first identifier corresponding to the first data flow, and a priority indicated by the second priority information is higher than a priority indicated by the first priority information.