WO2024082727A1

WO2024082727A1 - Packet scheduling method, network device, storage medium, and computer program product

Info

Publication number: WO2024082727A1
Application number: PCT/CN2023/107728
Authority: WO
Inventors: 彭少富
Original assignee: 中兴通讯股份有限公司
Priority date: 2022-10-20
Filing date: 2023-07-17
Publication date: 2024-04-25
Also published as: CN117955928A

Abstract

The present application provides a packet scheduling method, a network device, a storage medium, and a computer program product. The method comprises: determining an ordering guarantee object to which a target packet belongs, the ordering guarantee object being used for storing forwarding queue information of latest cached packets corresponding to the ordering guarantee object (S110); acquiring an allowed queuing delay of the target packet, the allowed queuing delay being used for representing the maximum queuing duration for the target packet waiting to be sent in a forwarding queue (S120); determining an initially selected queue according to the allowed queuing delay, the initially selected queue corresponding to a first sending countdown time (S130); and scheduling the target packet according to the first sending countdown time and the forwarding queue information stored in the ordering guarantee object (S140).

Description

Message scheduling method, network device, storage medium and computer program product

CROSS-REFERENCE TO RELATED APPLICATIONS

This application is based on the Chinese patent application with application number 202211289451.2 and application date October 20, 2022, and claims the priority of the Chinese patent application. The entire content of the Chinese patent application is hereby introduced into this application as a reference.

Technical Field

The embodiments of the present application relate to but are not limited to the field of network technology, and in particular, to a message scheduling method, network equipment, storage medium, and computer program product.

Background technique

In the related art, the controller will limit the local deadline of each node in the deterministic path, form a local deadline stack with these local deadlines, and carry the local deadline stack in the message, and then require each node to send out the message carrying the local deadline stack of its local deadline when its local deadline is reached. Although the scheme of carrying the local deadline stack in the message in the related art will meet the deterministic delay requirements, carrying the local deadline stack in the message will increase the complexity of message scheduling. For this reason, in the related art, the allocation rules and scheduling algorithms of the forwarding queue (or called the deadline queue) are proposed to reduce the complexity of message scheduling, and at the same time solve the problem of message disorder. However, this method requires the selection of parameters. If the selected parameters are not appropriate, the results may not meet expectations. Therefore, how to reduce the complexity of message scheduling and solve the problem of message disorder without selecting parameters is a technical problem that needs to be solved urgently.

Summary of the invention

Embodiments of the present application provide a message scheduling method, a network device, a storage medium, and a computer program product.

In a first aspect, an embodiment of the present application provides a message scheduling method, including: determining an ordered guarantee object to which a target message belongs, the ordered guarantee object being used to store forwarding queue information of a message corresponding to the ordered guarantee object that was most recently cached; obtaining an allowed queuing delay of the target message, the allowed queuing delay being used to characterize the maximum queuing time that the target message waits to be sent in the forwarding queue; determining a preliminary queue based on the allowed queuing delay, the preliminary queue corresponding to a first sending countdown time; scheduling the target message based on the first sending countdown time and the forwarding queue information stored in the ordered guarantee object.

In a second aspect, an embodiment of the present application further provides a network device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the message scheduling method as described in the first aspect above is implemented.

In a third aspect, an embodiment of the present application further provides a computer-readable storage medium storing computer-executable instructions, wherein the computer-executable instructions are used to execute the message scheduling method described in the first aspect above.

In a fourth aspect, the embodiments of the present application further provide a computer program product, including a computer program or a computer instruction, wherein the computer program or the computer instruction is stored in a computer-readable storage medium, and a processor of a computer device receives the computer program or the computer instruction from the computer. The computer-readable storage medium reads the computer program or the computer instruction, and the processor executes the computer program or the computer instruction, so that the computer device executes the message scheduling method described in the first aspect above.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a schematic diagram of a network topology for executing a message scheduling method provided by an embodiment of the present application;

FIG2 is a schematic diagram of a Deadline queue provided in a specific example of the present application;

FIG3 is a flow chart of a message scheduling method provided by an embodiment of the present application;

FIG4 is a flow chart of a message scheduling method provided by another embodiment of the present application;

FIG5 is a flowchart of a specific method of step S130 in FIG3 ;

FIG6 is a flowchart of a specific method of step S320 in FIG5 ;

FIG. 7 is a flowchart of a specific method of step S110 in FIG. 3 ;

FIG8 is a schematic diagram of determining an ordered guarantee object to which a target message belongs provided by a specific example of the present application;

FIG9 is a schematic diagram of a message queue provided by a specific example of the present application;

FIG. 10 is a schematic diagram of the structure of a network device provided in one embodiment of the present application.

Detailed ways

In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

It should be noted that although a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than that in the flowchart. The terms "first", "second", etc. in the specification, claims and the above drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

The present application provides a message scheduling method, a network device, a storage medium and a computer program product. First, the ordered guarantee object to which the target message belongs is determined. The ordered guarantee object is used to store the forwarding queue information of the message corresponding to the ordered guarantee object that was cached most recently. Then, the allowed queuing delay of the target message is obtained. The allowed queuing delay is used to characterize the maximum queuing time of the target message waiting to be sent in the forwarding queue. The preliminary selection queue is determined according to the allowed queuing delay. The preliminary selection queue corresponds to a first sending countdown time. Therefore, the target message can be scheduled according to the first sending countdown time and the forwarding queue information stored by the ordered guarantee object. There is no need to select parameters, and the node can avoid the situation where the message is out of order when sending the target message in the same data stream. At the same time, there is no need to carry a local deadline stack in the message, thereby reducing the complexity of message scheduling. Therefore, the embodiment of the present application can reduce the complexity of message scheduling and solve the problem of message out of order without selecting parameters.

The embodiments of the present application are further described below in conjunction with the accompanying drawings.

As shown in Figure 1, Figure 1 is a schematic diagram of a network topology for executing a message scheduling method provided by an embodiment of the present application. In the example of Figure 1, the network topology includes a first node 110, a second node 120, and a third node 130. Among them, a deterministic delay path passing through the second node 120 can be established between the first node 110 and the third node 130. The first node 110, the second node 120, and the third node 130 can all be network devices such as routers or switches, which can forward messages.

In this network topology, any node maintains a countdown timer for sending certain specific outbound ports. The forwarding queue (or deadline queue) of the countdown time (CT) and the authorization time (or authorization time (AT)) can be one or more, and can be appropriately set according to the actual application. The deadline queue has the following characteristics:

(1) The sending countdown time of each deadline queue will decrease over time. The smaller the sending countdown time value, the higher the scheduling priority of the corresponding deadline queue. A cyclic timer can be set in the node to decrement the sending countdown time of all deadline queues. That is, every time the cyclic timer times out, the sending countdown time of all deadline queues will be subtracted by the time interval of the cyclic timer. For example, assuming that the time interval of the cyclic timer is 1μs, then every time the cyclic timer times out, the countdown time of all deadline queues will be subtracted by 1μs.

The time interval of the cyclic timer is the step value of the countdown time of the deadline queue. The time interval must be less than or equal to the authorized sending time of the deadline queue, and the authorized sending time must be divisible by the time interval, that is, AT = N*I, where I represents the time interval and N is an integer greater than or equal to 1. In addition, the time interval of the cyclic timer should be fine enough to perceive the difference in delay requirements between messages. For example, if the difference in delay requirements between messages is only a few microseconds, the time interval of the cyclic timer can be 1 microsecond.

Initially, the initial values of the send countdown time of each deadline queue are staggered, that is, at any time, only one deadline queue's send countdown time decreases to 0.

(2) When the send countdown time decreases to 0, the scheduling priority of the corresponding deadline queue will be set to the highest, so that it will immediately get the opportunity to schedule messages, and will be prohibited from caching new messages. The messages cached in the deadline queue will be immediately sent out from the egress port. Among them, the maximum duration allowed for the deadline queue to send messages is the preset authorized sending time. The deadline queue will send all the messages cached in the queue within the authorized sending time. If the messages in the queue are sent and there is still free time for the authorized sending, the messages in other queues with the next highest priority can continue to be scheduled during the free authorized sending time.

(3) For a deadline queue whose sending countdown has been decremented to 0, after the authorized sending time has passed, its sending countdown will be restored to the preset maximum initial value (i.e., MAX_CT). At this time, the scheduling priority of the queue is not the highest, and the deadline queue is allowed to re-cache new messages, so that the deadline queue enters the next round of the operation of decrementing the sending countdown time as time goes by.

(4) For a deadline queue whose sending countdown time has not decreased to 0, the deadline queue is allowed to buffer messages.

Referring to FIG. 2 , FIG. 2 is a schematic diagram of a deadline queue provided by a specific example. In FIG. 2 , the deadline queue includes queues queue1 to queue7, while the other queues are traditional non-deadline queues. The authorized sending time of each deadline queue is 10 μs, and each has its countdown time attribute, and the preset maximum countdown time is 60 μs. Referring to FIG. 2 , at the initial moment (i.e., moment T0), the initial values of the sending countdown time of all deadline queues are staggered from each other, for example, the initial value of the sending countdown time of queue queue1 is 60 μs, the initial value of the sending countdown time of queue queue2 is 50 μs, the initial value of the sending countdown time of queue queue3 is 40 μs, and so on. At this time, only the initial value of the sending countdown time of queue queue7 is 0, that is, queue queue7 has the highest scheduling priority at this time, so queue queue7 will no longer cache new messages and immediately send the messages cached in queue queue7.

Assume that a cyclic timer with a time interval of 1μs is set in the node. Then, every time the cyclic timer times out, the countdown time of all deadline queues will be subtracted from the time interval of the cyclic timer. As shown in Figure 2, at time T0+1μs, the cyclic timer times out, and the countdown time of queue 1 decreases to 59μs, the countdown time of queue 2 decreases to 49μs, the countdown time of queue 3 decreases to 39μs, and so on. At this time, the countdown time of queue 7 decreases to 59μs. The sending countdown time is decremented to -1 (or, during a subsequent period of authorized sending time, the sending countdown time of queue 7 is kept at 0 μs), so the scheduling priority of queue 7 is still the highest level.

At T0+10μs, that is, after a period of authorized sending time, the sending countdown time of queue 1 decreases to 50μs, the sending countdown time of queue 2 decreases to 40μs, the sending countdown time of queue 3 decreases to 30μs, etc. At this time, the sending countdown time of queue 7 is restored to the maximum countdown time (i.e., 60μs), and the scheduling priority of queue 7 is no longer the highest level, that is, queue 7 is allowed to cache new messages again, and the sending countdown time of queue 6 becomes 0. Therefore, queue 6 has the highest scheduling priority, that is, during this period, queue 6 is prohibited from caching new messages, and the messages cached in queue 6 are sent immediately.

It is worth noting that the sending countdown time of each deadline queue is sorted in ascending order. Therefore, the next deadline queue of a deadline queue refers to the next deadline queue with a larger sending countdown time than the deadline queue and adjacent to it. For example, at time T0, the next deadline queue of queue7 is queue6. In addition, the sending countdown time of two deadline queues can be compared, and the deadline queue with a smaller sending countdown time value is called the previous deadline queue, and the deadline queue with a larger sending countdown time value is called the next deadline queue.

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

Those skilled in the art will appreciate that the topological structure shown in FIG. 1 does not constitute a limitation on the embodiments of the present application, and may include more or fewer components than shown in the figure, or a combination of certain components, or a different arrangement of components.

Based on the structure of the above network topology, various embodiments of the message scheduling method of the present application are proposed below.

Referring to Figure 3, Figure 3 is a flowchart of a message scheduling method provided by an embodiment of the present application. The message scheduling method can be applied to nodes in a network, such as the first node 110 or the second node 120 in the network topology shown in Figure 1. The message scheduling method includes but is not limited to step S110, step S120, step S130 and step S140.

Step S110: determining the order guarantee object to which the target message belongs, where the order guarantee object is used to store the forwarding queue information of the message corresponding to the most recently cached order guarantee object.

In this step, the message corresponding to the cached order guarantee object can be any message. The order guarantee object (OGO) is maintained in advance on the egress port where the queue (such as the candidate queue) is located. The number of the order guarantee objects is unlimited, and the number can be 1 or more. When the number of order guarantee objects is more than one, the forwarding queue information corresponding to different order guarantee objects can be different or the same.

In one embodiment, the target messages belonging to the same order guarantee object have some common features, such as coming from the same ingress port and/or having the same planned residence time and/or belonging to the same data flow, etc., which are not specifically limited here. In addition, the forwarding queue information can record the queue's sending countdown time, the queue number or queue name of the queue, etc., which are not specifically limited here.

Step S120: obtaining the allowed queuing delay of the target message, where the allowed queuing delay is used to represent the maximum queuing time length of the target message waiting to be sent in the forwarding queue.

In one embodiment, the allowed queuing delay can be obtained based on the planned residence time of the message in the current node, plus the accumulated delay deviation of the message in all upstream nodes, and then minus the forwarding delay of the message in the current node, where the planned residence time represents the maximum residence time of the message in the local node. It can be understood that if a message is on the current node If the actual queuing delay is greater than the allowed queuing delay, the message may have a smaller allowed queuing delay on the downstream node; conversely, if the actual queuing delay of a message on the current node is smaller than the allowed queuing delay, the message may have a larger allowed queuing delay on the downstream node.

Step S130: determining a preliminary selection queue according to the allowed queuing delay, and the preliminary selection queue corresponds to a first sending countdown time.

In this step, the preliminary selection queue is also a forwarding queue (i.e., a Deadline queue), and the number of preliminary selection queues can be 1. The first sending countdown time is the sending countdown time of the preliminary selection queue. For the specific description of the sending countdown time, please refer to the relevant description in the above embodiment, which will not be repeated here.

Step S140: Schedule the target message according to the first sending countdown time and the forwarding queue information stored in the order guarantee object.

In this step, since the ordered guarantee object to which the target message belongs is determined in step S110, it is possible to determine that the ordered guarantee object is used to store the forwarding queue information of the message corresponding to the most recently cached ordered guarantee object, and the first sending countdown time of the preliminary queue is determined in step S130. Therefore, the target message can be scheduled according to the first sending countdown time and the forwarding queue information stored in the ordered guarantee object. This eliminates the need to select parameters and can avoid message disorder when the node sends the target message in the same data stream. At the same time, there is no need to carry a local deadline stack in the message, thereby reducing the complexity of message scheduling.

In this embodiment, by adopting the message scheduling method including the above steps S110 to S140, first determine the ordered guarantee object to which the target message belongs, the ordered guarantee object is used to store the forwarding queue information of the message corresponding to the most recent cached ordered guarantee object, then obtain the allowed queuing delay of the target message, the allowed queuing delay is used to characterize the maximum queuing time of the target message waiting to be sent in the forwarding queue, determine the preliminary selection queue according to the allowed queuing delay, the preliminary selection queue corresponds to the first sending countdown time, therefore, the target message can be scheduled according to the first sending countdown time and the forwarding queue information stored by the ordered guarantee object, without selecting parameters, and can also avoid the situation that the node sends the target message in the same data stream out of order, and at the same time, there is no need to carry the local deadline stack in the message, thereby reducing the complexity of message scheduling. Therefore, the embodiment of the present application can reduce the complexity of message scheduling and solve the problem of message out of order without selecting parameters.

In one embodiment of the present application, step S140 is further described. When the forwarding queue information stored in the order guarantee object includes the second sending countdown time, step S140 may include but is not limited to the following steps:

Get the second sending countdown time stored in the order guarantee object.

The second sending countdown time is compared with the first sending countdown time to obtain a first comparison result.

The target message is scheduled according to the first comparison result.

In one embodiment, the second sending countdown time may be a preset value or a sending countdown time of a candidate queue. For example, if the ordered guarantee object to which the target message belongs is in the initial state, that is, the ordered guarantee object does not store the forwarding queue information of the message corresponding to the cached ordered guarantee object, then the second sending countdown time may be a preset value, which may be 0μs, -1μs or other values; when the second sending countdown time is the sending countdown time of the candidate queue, the specific description of the second sending countdown time may refer to the relevant description in the above embodiment, which will not be repeated here.

In this embodiment, the second sending countdown time stored in the order guarantee object can be obtained, and then the second sending countdown time and the first sending countdown time can be compared to obtain a first comparison result, and then the target message is scheduled according to the first comparison result. There is no need to select parameters, and it can also avoid the situation where the node sends the target message in the same data stream out of order. At the same time, there is no need to carry a local deadline stack in the message, thereby reducing the complexity of message scheduling. Therefore, the embodiments of the present application can reduce the complexity of message scheduling and solve the problem of message disorder without selecting parameters.

In another embodiment of the present application, step S140 is further described. When the forwarding queue information stored in the order guarantee object includes at least one of a queue number or a queue name, step S140 may include but is not limited to the following steps:

The second sending countdown time is obtained according to the forwarding queue information stored in the order guarantee object.

The target message is scheduled according to the first comparison result.

In one embodiment, there are many implementation methods for obtaining the second sending countdown time based on the forwarding queue information stored in the order guarantee object, such as obtaining the second sending countdown time based on the queue number; or obtaining the second sending countdown time based on the queue name; or obtaining the second sending countdown time based on the queue number and the queue name, and no specific restrictions are made here.

It can be understood that when the order guarantee object to which the target message belongs is in the initial state, that is, the order guarantee object does not store the forwarding queue information of the message corresponding to the cached order guarantee object, the forwarding queue information stored in the order guarantee object is empty, and the second sending countdown obtained at this time is the preset value.

It is understandable that different queue numbers or queue names correspond to different second sending countdown times. For example, assuming that the second sending countdown time obtained according to queue number queue6 is CT1, and the second sending countdown time obtained according to queue number queue7 is CT2, CT1 and CT2 are not equal.

In this embodiment, when the forwarding queue information stored in the order guarantee object includes at least one of the queue number or the queue name, the second sending countdown time can be obtained according to the forwarding queue information stored in the order guarantee object, and then the second sending countdown time is compared with the first sending countdown time to obtain a first comparison result, and then the target message is scheduled according to the first comparison result, without selecting parameters, and it can also avoid the situation that the node sends the target message in the same data stream out of order, and at the same time, it is not necessary to carry the local deadline stack in the message, thereby reducing the complexity of message scheduling. Therefore, the embodiment of the present application can reduce the complexity of message scheduling and solve the problem of message out of order without selecting parameters.

It is understandable that the sending countdown time of the above-mentioned queue (such as the first sending countdown time, the second sending countdown time) can be an explicitly set attribute of the queue, and the attribute (i.e., the sending countdown time) is presented to the outside during hardware implementation. In a specific application, the sending countdown time of the above-mentioned queue can also be implicitly derived through other information of the queue (such as the queue number, the queue name) combined with the necessary time number (such as the number of time intervals of the cyclic timer experienced within a single authorized sending time), and it is not necessary to present the attribute to the outside during hardware implementation. In addition, whether the sending countdown time is an explicitly set attribute or an attribute implicitly derived through other information of the queue, there is no difference in the first comparison result obtained by comparing the first sending countdown time with the second sending countdown time.

4 , an embodiment of the present application further illustrates the step of “scheduling the target message according to the first comparison result”, which may include but is not limited to step S210 and step S220.

Step S210: Determine a target forwarding queue according to the first comparison result.

In this step, the number of target forwarding queues is not limited, and may be one or more, and may be determined according to actual conditions, and no specific limitation is made here.

Step S220: Cache the target message to the target forwarding queue, and wait for the target forwarding queue to schedule the target message.

It is understandable that the fourth sending countdown time of each target forwarding queue will decrease over time, and the The smaller the value of the fourth sending countdown time is, the higher the scheduling priority of the corresponding target forwarding queue is. In addition, different target forwarding queues have different fourth sending countdown times.

In one embodiment, when the fourth sending countdown time has not decreased to 0, the target forwarding queue corresponding to the fourth sending countdown time fails to schedule the target message immediately, that is, waits for the target forwarding queue to schedule the target message; when the fourth sending countdown time decreases to 0, the scheduling priority of the target forwarding queue corresponding to the fourth sending countdown time will be set to the highest. At this time, the target forwarding queue can schedule the target message, and the embodiment of the present application does not impose any specific restrictions on this.

In one embodiment, it is assumed that there are two target forwarding queues, namely, target forwarding queue A and target forwarding queue B, and the fourth sending countdown time of target forwarding queue A is first decremented to 0. When the fourth sending countdown time of target forwarding queue A is decremented to 0, target forwarding queue A can schedule the target message cached therein. After a period of authorized sending time, the fourth sending countdown time of target forwarding queue B is decremented to 0, and the fourth sending countdown time of target forwarding queue A is restored to the preset maximum value. At this time, target forwarding queue B can schedule the target message cached therein.

In this embodiment, by adopting the message scheduling method including the above steps S210 to S220, the target forwarding queue can be determined according to the first comparison result, the target message is cached in the target forwarding queue, and the target forwarding queue is waited for to schedule the target message, without selecting parameters, and it can also avoid the situation that the node sends the target message in the same data stream out of order, and there is no need to carry the local deadline stack in the message, thereby reducing the complexity of message scheduling. Therefore, the embodiment of the present application can reduce the complexity of message scheduling and solve the problem of message out of order without selecting parameters.

In one embodiment of the present application, step S210 is further described. When the forwarding queue information stored in the order guarantee object corresponds to the candidate queue, and the candidate queue corresponds to the forwarding queue (i.e., the deadline queue) of the message corresponding to the most recently cached order guarantee object, step S210 may include but is not limited to the following steps:

When the first comparison result is that the first sending countdown time is less than the second sending countdown time, the candidate queue is determined as the target forwarding queue.

It can be understood that when the forwarding queue information stored in the order guarantee object corresponds to the candidate queue, and the candidate queue corresponds to the forwarding queue (ie, the deadline queue) of the message corresponding to the most recently cached order guarantee object, the second sending countdown time is the sending countdown time of the candidate queue.

In this embodiment, a first comparison result is obtained by comparing the second sending countdown time with the first sending countdown time of the preliminary queue. When the first comparison result is that the first sending countdown time is less than the second sending countdown time, the candidate queue is determined as the target forwarding queue, so that the target message can be cached to the target forwarding queue in the subsequent steps. The target message is scheduled through the target forwarding queue, which does not require parameter selection and can solve the problem of message disorder. At the same time, there is no need to carry a local deadline stack in the message, thereby reducing the complexity of message scheduling.

In another embodiment of the present application, step S210 is further described. Step S210 may include but is not limited to the following steps:

When the first comparison result is that the first sending countdown time is greater than or equal to the second sending countdown time, the preliminary selection queue is determined as the target forwarding queue.

In this embodiment, a first comparison result is obtained by comparing the second sending countdown time with the first sending countdown time of the preliminary selection queue. When the first comparison result is that the first sending countdown time is greater than or equal to the second sending countdown time, the candidate queue is determined as the target forwarding queue, so that the target message is cached to the target forwarding queue in the subsequent steps, and the target message is scheduled by the target forwarding queue. This does not require parameter selection, and can solve the problem of message disorder. There is no need to carry a local deadline stack in the message, thereby reducing the complexity of message scheduling.

5 , an embodiment of the present application further illustrates step S130 , which may include but is not limited to step S310 and step S320 .

Step S310: Obtain the authorized sending time and the third sending countdown time of the forwarding queue.

The authorized sending time is used to indicate the maximum duration of the forwarding queue to send the message. For the specific description of the authorized sending time, reference may be made to the relevant description in the above embodiment, which will not be repeated here.

In this step, there are multiple forwarding queues (i.e., Deadline queues), and different forwarding queues correspond to different third sending countdown times. For a specific description of the third sending countdown time (i.e., the sending countdown time of the Deadline queue), reference can be made to the relevant description in the above embodiment, which will not be repeated here.

Step S320: Determine the preliminary selection queue according to the authorized sending time, the third sending countdown time and the allowed queuing delay.

In this embodiment, multiple forwarding queues (i.e., deadline queues) may be provided in the node. Therefore, after executing step S120 to obtain the allowed queuing delay of the target message, and executing step S310 to obtain the authorized sending time and the third sending countdown times of multiple forwarding queues, a preliminary queue may be determined from the multiple forwarding queues according to the authorized sending time, the multiple third sending countdown times, and the allowed queuing delay, so that in subsequent steps, the target forwarding queue may be determined according to the first sending countdown time and the second sending countdown time of the preliminary queue. The target message may be scheduled through the target forwarding queue, without the need to select parameters and the problem of message disorder may be solved. At the same time, there is no need to carry a local deadline stack in the message, thereby reducing the complexity of message scheduling.

6 , an embodiment of the present application further illustrates step S320 , which may include but is not limited to step S410 and step S420 .

Step S410: Compare the third sending countdown time with the sum of the authorized sending time, the allowed queuing delay, and the third sending countdown time to obtain a second comparison result.

In this step, there can be multiple forwarding queues (i.e., Deadline queues), and the third sending countdown times of multiple forwarding queues can be compared with the sum of the authorized sending time, the allowed queuing delay, and the third sending countdown times of multiple forwarding queues at the same time; or, the third sending countdown times of multiple forwarding queues can be compared with the sum of the authorized sending time, the allowed queuing delay, and the third sending countdown times of multiple forwarding queues in an orderly manner. For example, multiple forwarding queues are sorted according to the third sending countdown times of the forwarding queues, and the third sending countdown times of multiple forwarding queues are compared with the sum of the authorized sending time, the allowed queuing delay, and the third sending countdown times of multiple forwarding queues in turn, and so on. No specific restrictions are made here.

Step S420: Determine a preliminary selection queue according to the second comparison result.

In this embodiment, after executing step S310 to obtain the authorized sending time and the third sending countdown time of the forwarding queue, and executing step S120 to obtain the allowed queuing delay of the target message, the third sending countdown time can be compared with the sum of the authorized sending time, the allowed queuing delay, and the third sending countdown time to obtain a second comparison result, and then a preliminary queue is determined based on the second comparison result, so that in subsequent steps, the target forwarding queue can be determined based on the first sending countdown time and the second sending countdown time of the preliminary queue. The target message is scheduled through the target forwarding queue, which does not require the selection of parameters and can solve the problem of message disorder. At the same time, there is no need to carry a local deadline stack in the message to reduce the complexity of message scheduling.

In one embodiment of the present application, step S410 is further described. There are multiple forwarding queues, different forwarding queues have different third sending countdown times, and different forwarding queues have the same authorized sending time. In this case, step S410 may include but is not limited to the following steps:

The allowed queuing delay is compared with the third sending countdown time of each forwarding queue, and the sum of the third sending countdown time corresponding to each forwarding queue and the authorized sending time.

When there is a forwarding queue among multiple forwarding queues that makes the allowed queuing delay greater than or equal to the third sending countdown time and less than the sum of the third sending countdown time and the authorized sending time, a second comparison result is obtained that the allowed queuing delay is greater than or equal to the third sending countdown time and less than the sum of the third sending countdown time and the authorized sending time.

It can be understood that when there are multiple forwarding queues (i.e., Deadline queues), and different forwarding queues correspond to different third sending countdown times, and different forwarding queues correspond to the same authorized sending time, the sum of the third sending countdown time and the authorized sending time corresponding to different forwarding queues is different.

In one embodiment, when there are multiple forwarding queues (i.e., deadline queues), the third sending countdown time corresponding to each forwarding queue can be compared with the sum of the authorized sending time, the allowed queuing delay, and the third sending countdown time of each forwarding queue at the same time; or the third sending countdown time of each forwarding queue can be compared with the sum of the authorized sending time, the allowed queuing delay, and the third sending countdown time of each forwarding queue in an orderly manner. For example, multiple forwarding queues are sorted according to the third sending countdown time of the forwarding queue, and the third sending countdown time of multiple forwarding queues are compared with the sum of the authorized sending time, the allowed queuing delay, and the third sending countdown time of multiple forwarding queues in turn, and so on. No specific restrictions are made here.

In this embodiment, a plurality of forwarding queues (i.e., Deadline queues) may be provided in the node. Therefore, after executing step S310 to obtain the authorized sending time and the third sending countdown time of the forwarding queue, and executing step S120 to obtain the allowed queuing delay of the target message, the allowed queuing delay may be compared with the third sending countdown time of each forwarding queue and the sum of the third sending countdown time corresponding to each forwarding queue and the authorized sending time. When there is a forwarding queue among the plurality of forwarding queues for which the allowed queuing delay is greater than or equal to the third sending countdown time and less than the sum of the third sending countdown time and the authorized sending time, a second comparison result is obtained in which the allowed queuing delay is greater than or equal to the third sending countdown time and less than the sum of the third sending countdown time and the authorized sending time, so as to facilitate determining the preliminary selection queue according to the second comparison result in the subsequent steps.

In one embodiment, step S420 is further described. When there are multiple forwarding queues, different forwarding queues correspond to different third sending countdown times, and different forwarding queues correspond to the same authorized sending time, step S420 may include but is not limited to the following steps:

When the second comparison result is that the allowed queuing delay is greater than or equal to the third sending countdown time and less than the sum of the third sending countdown time and the authorized sending time, the forwarding queue whose allowed queuing delay is greater than or equal to the third sending countdown time and less than the sum of the third sending countdown time and the authorized sending time is determined as the preliminary queue.

In this embodiment, multiple forwarding queues (i.e., deadline queues) may be provided in the node. Therefore, when the second comparison result is that the allowed queuing delay is greater than or equal to the third sending countdown time and less than the sum of the third sending countdown time and the authorized sending time, the forwarding queue whose allowed queuing delay is greater than or equal to the third sending countdown time and less than the sum of the third sending countdown time and the authorized sending time is determined as the preliminary queue, so that in the subsequent steps, the target forwarding queue is determined according to the first sending countdown time and the second sending countdown time of the preliminary queue. The target message is scheduled through the target forwarding queue, which does not require the selection of parameters and can solve the problem of message disorder. At the same time, there is no need to carry a local deadline stack in the message to reduce the complexity of message scheduling.

Specifically, among the multiple forwarding queues, if there is a forwarding queue whose third sending countdown time CT satisfies CT<=Q<CT+AT, the forwarding queue is determined as the primary selection queue, where Q is the allowed queuing delay of the target message. Assume that the preliminary queue is queue-X and the candidate queue corresponding to OGO is queue-Y. If the first sending countdown time of the preliminary queue queue-X is less than the second sending countdown time stored by OGO, the preliminary queue queue-X is abandoned, and the candidate queue queue-Y corresponding to OGO is used as the target forwarding queue, and the fourth sending countdown time of the target forwarding queue = MAX{the first sending countdown time of the preliminary queue, the second sending countdown time stored by OGO}, that is, the fourth sending countdown time of the target forwarding queue is equal to the largest sending countdown time between the first sending countdown time of the preliminary queue and the second sending countdown time stored by OGO, where "=" does not indicate assignment, but indicates a specific value; MAX indicates an operator for taking the maximum value.

It is understandable that the messages in the same queue are scheduled according to the first-in-first-out rule of the queue. When the target message is not at the head of the target forwarding queue, the scheduling countdown time of the target forwarding queue is not equal to the scheduling countdown time of the target message, that is, the target message needs to wait longer than the head message to be scheduled. In order to eliminate this effect, in some specific implementations, the preliminary selection queue can be selected after deducting a certain margin based on the allowed queuing delay Q of the target message, such as selecting the preliminary selection queue according to CT<=Q-k*AT<CT+AT, where 0<=k<=1. However, it should be noted that in other cases, the target message may also be near the head of the target queue, and the allowed queuing delay Q of the target message is much greater than the remaining countdown time of the target queue, that is, the target message only needs to wait for a shorter time to be scheduled. In summary, and for the sake of simplicity in the implementation of the solution, the embodiment of the present application recommends selecting the primary queue based on CT<=Q<CT+AT. In fact, the scheduling mechanism of the Deadline queue has the ability to compensate for delays, and the delay deviation generated at the current node will be compensated at the downstream node.

7 , an embodiment of the present application further illustrates step S110 , which may include but is not limited to step S510 and step S520 .

Step S510: classify the target message according to a preset granularity to obtain the object to which the target message belongs.

Among them, the preset granularity can be per "ingress port + planned residence time", per "flow", per "ingress port", per "ingress port + flow", or per "planned residence time", etc., wherein "flow" refers to data flow. Specifically, the target messages are classified according to each "ingress port + planned residence time", that is, the target messages from the same ingress port and with the same planned residence time are classified into the same category; similarly, the target messages are classified according to each "flow", that is, the target messages belonging to the same data flow are classified into the same category; the target messages are classified according to each "ingress port", that is, the target messages from the same ingress port are classified into the same category; the target messages are classified according to each "ingress port + flow", that is, the target messages from the same ingress port and belonging to the same data flow are classified into the same category; the target messages are classified according to each "planned residence time", which means that the target messages with the same planned residence time are classified into the same category. The preset granularity can be selected according to the actual situation, and no specific restrictions are made here.

Step S520: Determine the belonging object as the order-guaranteed object to which the target message belongs.

It is understandable that there are multiple order guarantee objects, and there can be multiple target messages belonging to the same order guarantee object, and the target messages belonging to the same order guarantee object have some common points, such as coming from the same ingress port and/or having the same planned residence time and/or belonging to the same data flow, etc., and the common points can be known by preset granularity. Among them, when the preset granularity is per "ingress port + planned residence time", the order guarantee object can be used to implement the maintenance of the ingress port value and the planned residence time attribute value.

In this embodiment, by adopting the message scheduling method including the above-mentioned steps S510 to S520, the target message can be classified and processed according to the preset granularity to obtain the belonging object of the target message, and then the belonging object is determined as the ordered guarantee object to which the target message belongs, paving the way for determining the target forwarding queue in the subsequent steps.

In one embodiment of the present application, when the forwarding queue information stored in the order guarantee object corresponds to the candidate queue, and the candidate queue corresponds to the forwarding queue of the message corresponding to the most recently cached order guarantee object, the message scheduling method may also include but is not limited to the following steps:

When the target message is cached in the target forwarding queue, the forwarding queue information stored in the order guarantee object is updated to the queue information of the target forwarding queue.

In one embodiment, after the target message A is cached to the target forwarding queue a and the forwarding queue information stored in the ordered guarantee object is updated to the target forwarding queue a, when the target message B needs to be scheduled, the ordered guarantee object to which the target message B belongs can be determined. If the ordered guarantee object to which the target message B belongs is the same as the ordered guarantee object to which the target message A belongs, at this time, the candidate queue corresponding to the ordered guarantee object to which the target message B belongs is the target forwarding queue a, that is, the target message B is scheduled according to the second sending countdown time of the ordered guarantee object (before it is reduced to a preset value, it is equal to the fourth sending countdown time corresponding to the target forwarding queue a), the first sending countdown time corresponding to the preliminary queue, and the allowed queuing delay of the target message B.

In one embodiment, when the forwarding queue information stored in the order guarantee object includes at least one of a queue number or a queue name, after the forwarding queue information stored in the order guarantee object is updated to the queue information of the target forwarding queue, the fourth sending countdown time corresponding to the target forwarding queue can be decremented using a time interval. When the fourth sending countdown time corresponding to the target forwarding queue is decremented to a preset value, the forwarding queue information stored in the order guarantee object is cleared, that is, the queue information of the target forwarding queue stored in the order guarantee object is cleared, wherein the queue information includes at least one of a queue number or a queue name.

In one embodiment of the present application, when the order guarantee object includes a countdown time field for storing the second sending countdown time, after executing the step of “updating the forwarding queue information stored in the order guarantee object to the queue information of the target forwarding queue”, the message scheduling method may also include but is not limited to the following steps:

The second sending countdown time in the countdown time field is updated to the fourth sending countdown time corresponding to the target forwarding queue.

The value in the countdown time field is decremented using a time interval until the value in the countdown time field reaches a preset value, wherein the time interval is calculated based on the authorized sending time.

In this step, the value in the countdown time field is decremented using the time interval, that is, whenever the cyclic timer times out, the values stored in all order guarantee objects will be subtracted from the time interval of the cyclic timer.

In one embodiment, a cyclic timer can be set in the node, and the time interval is the step value used by the cyclic timer to decrease the sending countdown time (such as the first sending countdown time, the second sending countdown time, the third sending countdown time, and the fourth sending countdown time) of all deadline queues (i.e., forwarding queues), and the time interval must be less than or equal to the authorized sending time, and the time interval of the cyclic timer should be set finely enough to perceive the delay requirement difference between messages. For example, if the delay requirement difference between messages is only a few microseconds, the time interval of the cyclic timer can be 1 microsecond.

In a feasible implementation manner, the preset value may be other values of the fourth sending countdown time that is less than or equal to the minimum initial value of all target forwarding queues, such as 0 microseconds or -1 microsecond, etc. No specific limitation is made here.

In this embodiment, when the order guarantee object includes a countdown time field for storing the second sending countdown time, after the forwarding queue information stored in the order guarantee object is updated to the queue information of the target forwarding queue, the second sending countdown time in the countdown time field can be updated to the fourth sending countdown time corresponding to the target forwarding queue, and the value in the countdown time field can be decremented using a time interval until the value in the countdown time field is a preset value. That is, the sending countdown time (ie, the second sending countdown time) recorded by the order guarantee object may decrease synchronously with the sending countdown times of other forwarding queues (eg, the third sending countdown time) as time goes by.

It is understandable that before the forwarding queue information stored in the order guarantee object is updated to the queue information of the target forwarding queue, the second sending countdown time in the countdown time field corresponding to the order guarantee object can also decrease synchronously with the sending countdown time of other forwarding queues over time.

In one embodiment of the present application, the message scheduling method may also include but is not limited to the following steps:

When the value in the countdown time field of the order guarantee object decreases to a preset value, the value in the countdown time field is kept unchanged until the forwarding queue information stored in the order guarantee object is updated again.

In one embodiment, assuming that the preset value is 0μs, the time interval is 1μs, and the value in the countdown time field is 3μs, when the value in the countdown time field decreases to 0μs at a time interval of 1μs, the value in the countdown time field will remain at 0μs. After a period of time, the forwarding queue information stored in the order guarantee object is updated again. If the fourth sending countdown time corresponding to the new target forwarding queue is 4μs, then at this time, the value 0μs in the countdown time field of the order guarantee object will be updated to 4μs. No specific restrictions are made here.

In one embodiment, when the second sending countdown time in the countdown time field corresponding to the order guarantee object (such as the sending countdown time of the candidate queue) decreases to a preset value, all records of the queue can be cleared in the order guarantee object.

The following is a specific example to explain in detail how to determine the order guarantee object to which the target message belongs.

Referring to FIG8 , FIG8 is a schematic diagram of determining the ordered guarantee object to which the target message belongs, provided by a specific example of the present application. In FIG8 , it is assumed that there are m ingress ports, namely, ingress port 1, ingress port 2, ..., ingress port m, where m is a positive integer, and there is one egress port, wherein the egress port creates different ordered guarantee objects according to the preset granularity of each "ingress port + planned residence time". If a target message with a corresponding planned residence time is received from a certain ingress port, the target message can be classified and processed according to the ingress port and the planned residence time to obtain the object to which the target message belongs, and the belonging object is determined as the ordered guarantee object to which the target message belongs.

It is understandable that each of the above-mentioned order guarantee objects (ie, OGOs) may be a simple register or memory unit allocated on the hardware, and this embodiment does not impose any limitation on this.

In order to more clearly illustrate the processing flow of the message scheduling method, a specific example is given below.

Example 1:

In the network topology, node R receives a sequence of messages from one of its inbound ports, among which messages P1 to P5 are five messages belonging to the same data stream, and these five messages are sent almost simultaneously at the source end. However, in the process of message forwarding, due to the different interference delays faced by each message, the time when node R receives message P2 is 1μs later than the time when it receives message P1, the time when node R receives message P3 is 1μs later than the time when it receives message P2, the time when node R receives message P4 is 1μs later than the time when it receives message P3, and the time when node R receives message P5 is 1μs later than the time when it receives message P4. That is to say, node R receives messages P1, P2, P3, P4 and P5 in sequence, and the delay between each adjacent message received is 1μs. Therefore, at node R, the later received message will have a smaller allowed queuing delay Q. As shown in Figure 9, the allowed queuing delay Q of P1 is 25μs, that is, P1(25), while the allowed queuing delay Q of P2 is 24μs, that is, P2(24), the allowed queuing delay Q of P3 is 23μs, that is, P3(23), the allowed queuing delay Q of P4 is 22μs, that is, P4(22), and the allowed queuing delay Q of P5 is 21μs, that is, P5(21).

Assume that the authorized sending time AT of the Deadline queue (i.e., the forwarding queue) maintained by node R is 10 μs, and the time interval I used to decrement the sending countdown time CT of the Deadline queue is 1 μs. Also, assume that node R is at T0 Message P1 is received at time T0+1μs, message P2 is received at time T0+1μs, and so on.

If the deadline queues set by node R include queue-A, queue-B, and queue-C, and it is assumed that in the time interval from T0 to T0+1μs, the CT of queue-A, the CT of queue-B, and the CT of queue-C are 25μs, 15μs, and 5μs respectively. In the time interval from T0+1μs to T0+2μs, the CT of queue-A, the CT of queue-B, and the CT of queue-C decrease to 24μs, 14μs, and 4μs respectively.

Since P1 and P2 are received at the same time within the time interval I from T0 to T0+1μs, according to relevant technologies, for example, the message is queued according to the rule of Q>=CT, that is, when the allowed queuing delay of the message is greater than or equal to the sending countdown time, the message can be cached in the queue corresponding to the sending countdown time, then message P1 will be inserted into queue-A, and P2 will be inserted into queue-B. Because the sending countdown time of queue-B is less than the sending countdown time of queue-A (that is, compared with queue-A, queue-B schedules message P2 first), node R will send message P2 first, and then send message P1. Therefore, message P2 and message P1 will be out of order. Similarly, when the message is queued according to the rule of Q<=CT, there will be a problem of message out of order. In order to avoid repetition and redundancy, it will not be repeated here.

Based on this, the target message can be scheduled according to the first sending countdown time of the preliminary selection queue and the forwarding queue information stored in the order guarantee object. The specific implementation process is as follows:

First, since messages P1 to P5 are sent almost simultaneously at the source end, the planned residence time of messages P1 to P5 is the same. Assuming that the target messages are classified and processed according to the preset granularity of each "inbound port + planned residence time", the above message sequences belong to the same order guarantee object, then:

At time T0, the target message P1 is received, wherein the allowed queuing delay Q of the target message P1 is equal to 25, therefore, the target message P1 can be cached to queue-A (at this time, the third sending countdown time of queue-A is equal to 25). Specifically, the preliminary selection queue is determined according to the authorized sending time, the third sending countdown time and the allowed queuing delay. Since the third sending countdown time of queue-A satisfies 25<=25<35, wherein 35 is the sum of the third sending countdown time of queue-A and the authorized sending time, that is, the allowed queuing delay is greater than or equal to the third sending countdown time of queue-A and less than the sum of the third sending countdown time of queue-A and the authorized sending time, therefore, queue-A is determined as the preliminary selection queue; next, assuming that the second sending countdown time recorded by the ordered guarantee object to which the target message P1 belongs is the preset value 0, that is, the first sending countdown time (i.e. 25) of the selected preliminary selection queue queue-A is greater than the second sending countdown time recorded by the ordered guarantee object, therefore, the preliminary selection queue queue-A is determined as the target forwarding queue. When the target message P1 is cached to the target forwarding queue queue-A, the second sending countdown time (i.e. 0) recorded by the order guarantee object is updated to the fourth sending countdown time (i.e. 25) of the target forwarding queue queue-A. At this time, the candidate queue corresponding to the order guarantee object is queue-A.

At a time close to T0+1μs (at this time, the sending countdown time has not yet jumped), the target message P2 is received, wherein the allowed queuing delay Q of the target message P2 is equal to 24. Therefore, the target message P2 can be cached to queue-A (at this time, the third sending countdown time of queue-A is still equal to 25, which is about to jump to 24). Specifically, the preliminary queue is determined according to the authorized sending time, the third sending countdown time and the allowed queuing delay. Since the third sending countdown time of queue-B satisfies 15<=24<25, where 25 is the sum of the third sending countdown time of queue-B and the authorized sending time, that is, the allowed queuing delay is greater than or equal to the third sending countdown time of queue-B and less than the sum of the third sending countdown time of queue-B and the authorized sending time, therefore, queue-B is determined as the preliminary queue; then, since the second sending countdown time recorded by the order guarantee object is 25 (not yet changed to 24), therefore, the first sending countdown time (i.e. 15) of the selected preliminary queue queue-B is less than the second sending countdown time recorded by the order guarantee object (i.e. the sending countdown time of the candidate queue queue-A, 25), therefore, the candidate queue queue-A is determined as the target forwarding queue. When the target message P2 is cached to the target forwarding queue queue-A Afterwards, accordingly, the second sending countdown time (i.e. 25) recorded by the order guarantee object is updated to the fourth sending countdown time of the target forwarding queue queue-A (i.e. 25, the same as the second sending countdown time recorded by the previous order guarantee object). At this time, the candidate queue corresponding to the order guarantee object is queue-A.

Similarly, at a time close to T0+2μs, the target message P3 is received, wherein the allowed queuing delay Q of the target message P3 is equal to 23, therefore, the target message P3 can be cached to queue-A (at this time, the third sending countdown time of queue-A is still equal to 24, which is about to jump to 23). Specifically, the preliminary selection queue is determined according to the authorized sending time, the third sending countdown time and the allowed queuing delay. Since the third sending countdown time of queue-B is 14 at this time, it satisfies 14<=23<24, therefore, queue-B is determined as the preliminary selection queue; then, since the second sending countdown time recorded by the order guarantee object is decreased to 24 (not yet jumped to 23), therefore, the first sending countdown time (i.e. 14) of the selected preliminary selection queue queue-B is less than the second sending countdown time recorded by the order guarantee object (i.e. the sending countdown time of the candidate queue queue-A is 24), therefore, the candidate queue queue-A is determined as the target forwarding queue. When the target message P3 is cached to the target forwarding queue queue-A, the second sending countdown time (i.e. 24) recorded by the order guarantee object is updated to the fourth sending countdown time of the target forwarding queue queue-A (i.e. 24, the same as the second sending countdown time recorded by the previous order guarantee object). At this time, the candidate queue corresponding to the order guarantee object is queue-A.

Similarly, at the time close to T0+3μs, the target message P4 will also be cached in queue-A; at the time close to T0+4μs, the target message P5 will also be cached in queue-A. In order to avoid redundancy, it will not be repeated here. That is, node R can send messages P1 to P5 belonging to the same data flow in order.

It can be seen that the ordered guarantee object and message scheduling method used in this example (for example, the target message is queued according to the allowed queuing delay of the target message and the rules of the ordered guarantee object) can solve the message disorder problem caused by delay compensation in the Deadline forwarding mechanism in the related technology (for example, the message is queued according to the rule of Q>=CT, or the message is queued according to the rule of Q<=CT, etc.), which is conducive to meeting the needs of deterministic business.

In addition, referring to FIG. 10 , an embodiment of the present application further provides a network device, wherein the network device 200 includes a memory 202 , a processor 201 , and a computer program stored in the memory 202 and executable on the processor 201 .

The processor 201 and the memory 202 may be connected via a bus or other means.

The memory 202, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer executable programs. In addition, the memory 202 may include a high-speed random access memory, and may also include a non-transitory memory, such as at least one disk storage device, a flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory 202 may optionally include a memory remotely arranged relative to the processor 201, and these remote memories may be connected to the processor 201 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

It should be noted that the network device 200 in this embodiment can be applied as any node in the embodiment shown in Figure 1. The network device 200 in this embodiment can constitute a part of the network topology in the embodiment shown in Figure 1. These embodiments all belong to the same inventive concept, so these embodiments have the same implementation principles and technical effects, which will not be described in detail here.

The non-transient software programs and instructions required to implement the message scheduling method of the above embodiment are stored in the memory 202. When executed by the processor 201, the message scheduling method in the above embodiment is executed, for example, the method steps S110 to S140 in Figure 3, the method steps S210 to S220 in Figure 4, the method steps S310 to S320 in Figure 5, the method steps S410 to S420 in Figure 6, and the method steps S510 to S520 in Figure 7 described above are executed.

The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, an embodiment of the present application also provides a computer-readable storage medium, which stores computer-executable instructions, and the computer-executable instructions are executed by a processor or controller, for example, by a processor in the above-mentioned device embodiment, so that the above-mentioned processor can execute the message scheduling method in the above-mentioned embodiment, for example, execute the method steps S110 to S140 in Figure 3 described above, the method steps S210 to S220 in Figure 4, the method steps S310 to S320 in Figure 5, the method steps S410 to S420 in Figure 6, and the method steps S510 to S520 in Figure 7.

In addition, an embodiment of the present application also provides a computer program product, including a computer program or computer instructions, the computer program or computer instructions are stored in a computer-readable storage medium, the processor of a computer device reads the computer program or computer instructions from the computer-readable storage medium, and the processor executes the computer program or computer instructions, so that the computer device executes the message scheduling method in the above embodiment, for example, executes the method steps S110 to S140 in Figure 3, the method steps S210 to S220 in Figure 4, the method steps S310 to S320 in Figure 5, the method steps S410 to S420 in Figure 6, and the method steps S510 to S520 in Figure 7 described above.

The embodiment of the present application includes: firstly determining the ordered guarantee object to which the target message belongs, the ordered guarantee object is used to store the forwarding queue information of the message corresponding to the ordered guarantee object cached most recently, then obtaining the allowed queuing delay of the target message, the allowed queuing delay is used to characterize the maximum queuing time of the target message waiting to be sent in the forwarding queue, determining the preliminary selection queue according to the allowed queuing delay, the preliminary selection queue corresponds to a first sending countdown time, therefore, the target message can be scheduled according to the first sending countdown time and the forwarding queue information stored by the ordered guarantee object, without selecting parameters, and can also avoid the situation that the node sends the target message in the same data stream out of order, and at the same time, there is no need to carry the local deadline stack in the message, thereby reducing the complexity of message scheduling. Therefore, the embodiment of the present application can reduce the complexity of message scheduling and solve the problem of message out of order without selecting parameters.

It will be appreciated by those skilled in the art that all or some of the steps and systems in the disclosed method above may be implemented as software, firmware, hardware and appropriate combinations thereof. Some physical components or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor or a microprocessor, or may be implemented as hardware, or may be implemented as an integrated circuit, such as an application specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or a non-transitory medium) and a communication medium (or a temporary medium). As known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, disk storage or other magnetic storage devices, or any other medium that may be used to store desired information and may be accessed by a computer. Furthermore, it is well known to those skilled in the art that communication media typically embodies computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

Claims

A message scheduling method, comprising:

Determine the order guarantee object to which the target message belongs, the order guarantee object being used to store the forwarding queue information of the message corresponding to the order guarantee object that was cached most recently;

Acquire an allowed queuing delay of the target message, where the allowed queuing delay is used to represent a maximum queuing time length of the target message waiting to be sent in a forwarding queue;

Determine a preliminary selection queue according to the allowed queuing delay, wherein the preliminary selection queue corresponds to a first sending countdown time;

The target message is scheduled according to the first sending countdown time and the forwarding queue information stored in the order guarantee object.
The message scheduling method according to claim 1, wherein the forwarding queue information stored in the order guarantee object includes a second sending countdown time;

The scheduling of the target message according to the first sending countdown time and the forwarding queue information stored in the order guarantee object includes:

Obtaining the second sending countdown time stored in the order guarantee object;

Comparing the second sending countdown time with the first sending countdown time to obtain a first comparison result;

The target message is scheduled according to the first comparison result.
The message scheduling method according to claim 1, wherein the forwarding queue information stored in the order guarantee object includes at least one of a queue number or a queue name;

The scheduling of the target message according to the first sending countdown time and the forwarding queue information stored in the order guarantee object includes:

Obtaining a second sending countdown time according to the forwarding queue information stored in the order guarantee object;

Comparing the second sending countdown time with the first sending countdown time to obtain a first comparison result;

The target message is scheduled according to the first comparison result.
The message scheduling method according to claim 2 or 3, wherein scheduling the target message according to the first comparison result comprises:

Determine a target forwarding queue according to the first comparison result;

The target message is cached in the target forwarding queue, and the target forwarding queue is waited for scheduling the target message.
The message scheduling method according to claim 4, wherein the forwarding queue information stored in the order guarantee object corresponds to a candidate queue, and the candidate queue corresponds to a forwarding queue that most recently cached the message corresponding to the order guarantee object;

The determining the target forwarding queue according to the first comparison result includes:

When the first comparison result is that the first sending countdown time is less than the second sending countdown time, the candidate queue is determined as the target forwarding queue.
The message scheduling method according to claim 4, wherein determining the target forwarding queue according to the first comparison result comprises:

When the first comparison result is that the first sending countdown time is greater than or equal to the second sending countdown time, the preliminary selection queue is determined as the target forwarding queue.
The message scheduling method according to claim 1, wherein the determining the preliminary queue according to the allowed queuing delay comprises:

Obtaining an authorized sending time and a third sending countdown time of the forwarding queue, wherein the authorized sending time is used to represent the maximum duration for which the forwarding queue is allowed to send a message;

A preliminary queue is determined according to the authorized sending time, the third sending countdown time and the allowed queuing delay.
The message scheduling method according to claim 7, wherein the determining the preliminary queue according to the authorized sending time, the third sending countdown time and the allowed queuing delay comprises:

Compare the third sending countdown time with the sum of the authorized sending time, the allowed queuing delay, and the third sending countdown time to obtain a second comparison result;

A preliminary selection queue is determined according to the second comparison result.
The message scheduling method according to claim 8, wherein there are multiple forwarding queues, different forwarding queues correspond to different third sending countdown times, and different forwarding queues correspond to the same authorized sending time;

The determining of a preliminary selection queue according to the second comparison result includes:

When the second comparison result is that the allowed queuing delay is greater than or equal to the third sending countdown time and less than the sum of the third sending countdown time and the authorized sending time, the forwarding queue whose allowed queuing delay is greater than or equal to the third sending countdown time and less than the sum of the third sending countdown time and the authorized sending time is determined as the preliminary queue.
The message scheduling method according to claim 1, wherein the step of determining the order guarantee object to which the target message belongs comprises:

Classify the target message according to a preset granularity to obtain the object to which the target message belongs;

The belonging object is determined as the order guarantee object to which the target message belongs.
The message scheduling method according to claim 4, wherein the forwarding queue information stored in the order guarantee object corresponds to a candidate queue, and the candidate queue corresponds to a forwarding queue that most recently cached the message corresponding to the order guarantee object;

The message scheduling method also includes:

When the target message is cached in the target forwarding queue, the forwarding queue information stored in the order guarantee object is updated to the queue information of the target forwarding queue.
The message scheduling method according to claim 11, further comprising:

After the forwarding queue information stored in the order guarantee object is updated to the queue information of the target forwarding queue, when the fourth sending countdown time corresponding to the target forwarding queue decreases to a preset value, the forwarding queue information stored in the order guarantee object is cleared.
The message scheduling method according to claim 11, wherein the order guarantee object includes a countdown time field for storing the second sending countdown time;

After the forwarding queue information stored in the order guarantee object is updated to the queue information of the target forwarding queue, the message scheduling method further includes:

Updating the second sending countdown time in the countdown time field to the fourth sending countdown time corresponding to the target forwarding queue;

The value in the countdown time field is decremented using a time interval until the value in the countdown time field is is a preset value, wherein the time interval is calculated according to the authorized sending time.
The message scheduling method according to claim 13, further comprising:

When the value in the countdown time field decreases to the preset value, the value in the countdown time field is kept unchanged until the forwarding queue information stored in the order guarantee object is updated again.
A network device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the message scheduling method as claimed in any one of claims 1 to 14 when executing the computer program.
A computer-readable storage medium storing computer-executable instructions, wherein the computer-executable instructions are used to execute the message scheduling method described in any one of claims 1 to 14.
A computer program product, comprising a computer program or computer instructions, wherein the computer program or the computer instructions are stored in a computer-readable storage medium, a processor of a computer device reads the computer program or the computer instructions from the computer-readable storage medium, and the processor executes the computer program or the computer instructions, so that the computer device executes the message scheduling method as described in any one of claims 1 to 14.