CN113542151A

CN113542151A - Traffic scheduling method and device and computer readable storage medium

Info

Publication number: CN113542151A
Application number: CN202110676498.3A
Authority: CN
Inventors: 吴晓佳
Original assignee: New H3C Big Data Technologies Co Ltd
Current assignee: New H3C Big Data Technologies Co Ltd
Priority date: 2021-06-18
Filing date: 2021-06-18
Publication date: 2021-10-22

Abstract

The present specification provides a traffic scheduling method and apparatus, where the method includes: determining whether a plurality of traffics have a common port of a cross device passing through the common port, wherein the common port is an output port through which the plurality of traffics flow; and judging whether a time conflict exists when the flow is forwarded out from the shared port of the cross device, if so, sending an instruction to the cross device, wherein the instruction is used for indicating the flow passing through the conflict among the flows with the conflict to wait for a preset time. The network management equipment determines the time range of the flow possibly staying at the common port by calculating the earliest time of reaching the common port and the latest time of the flow completely forwarded from the common port, if the time ranges corresponding to different flows are overlapped, the conflict of the flow forwarding on the time exists, and the conflict problem can be solved well in a time delay mode.

Description

Traffic scheduling method and device and computer readable storage medium

Technical Field

The present disclosure relates to the field of communications technologies, and in particular, to a traffic scheduling method and apparatus, and a computer-readable storage medium.

Background

With the constant convergence of Information Technology (IT) and Operation Technology (OT), the need for a unified network architecture becomes urgent. The development of intelligent manufacturing, industrial internet of things and big data all make the integration more urgent. The different demands of IT and OT for communication have also led for a long time to a great obstacle to merging these two areas: data in the internet and information fields requires more bandwidth, and real-time and certainty are critical to the industry. These data cannot usually be transmitted in the same network. Therefore, finding a unified solution has become a necessary requirement for industry convergence.

A Time Sensitive Network (TSN) is a completely new industrial communication technology that is being actively promoted by the international industry at present. The time sensitive network allows periodic and aperiodic data to be transmitted in the same network, so that the standard ethernet has the advantage of deterministic transmission and has become a key technology of wide focus through a vendor independent standardization process.

The network traffic forwarded in the TSN network needs to find an optimal path, which is short in route and less in time consumption, and ensures that all the traffic forwarded in the TSN network has no conflict time on all aggregated devices, where an aggregated device refers to a network device included in multiple paths.

The Dijkstra algorithm is a common algorithm for finding a shortest path, but the algorithm cannot meet the problems of time collision and single point collision of port queue allocation when the forwarded traffic reaches the same device in the same time period in the TSN network.

Disclosure of Invention

In order to overcome the problems in the related art, the present specification provides a traffic scheduling method and apparatus.

According to a first aspect of embodiments herein, there is provided a traffic scheduling method applied to a network management device, the method including:

determining whether a plurality of traffics have a common port of a cross device passing through the common port, wherein the common port is an output port through which the plurality of traffics flow;

and judging whether a time conflict exists when the flow is forwarded out from the shared port of the cross device, if so, sending an instruction to the cross device, wherein the instruction is used for indicating the flow passing through the conflict among the flows with the conflict to wait for a preset time.

Optionally, the method further includes:

determining a constraint value of forwarding delay between a sender and a receiver of each flow;

determining a difference value between the starting forwarding time of the flow on a sender corresponding to a target path and the arrival time of the flow reaching a receiver, wherein the target path is an reachable path between the sender and a target;

if the difference value is larger than the constraint value of the forwarding delay, the target path is abandoned, and other forwarding paths between the sender and the receiver corresponding to the flow are determined again.

Optionally, determining whether there is a common port of the crossing device that is passed through by multiple traffic, including:

determining the shortest path corresponding to each flow for each flow;

comparing according to each path with the shortest flow, and determining whether the shortest path of each flow has a common port of the cross equipment which passes through the shortest path;

if the difference value between the starting forwarding time and the arrival time at the receiver on the sender corresponding to the shortest path of the traffic is larger than the constraint value of the forwarding delay, selecting a secondary short path corresponding to the traffic to compare with the shortest paths of other traffic, and determining whether a shared port of the cross device passes through commonly.

Optionally, determining whether there is a time conflict when traffic is forwarded on the common port of the cross device, and if there is a conflict, sending an instruction to the cross device, including:

determining the earliest time each flow reaches the common port;

determining the latest time when each flow is completely forwarded out of the common port;

determining a time range within which each flow is processed at the common port according to the earliest time and the latest time;

and if the time ranges of the plurality of flows have intersection, sending an instruction to the intersection equipment to indicate that the flows passing through the time ranges after the intersection exist in the flows waiting for presetting.

Optionally, the method further includes:

updating the earliest time of the traffic passing through the output port of the network equipment after the current cross equipment and the latest time of the traffic being completely forwarded out of the output port according to the preset time of the traffic waiting at the common port of the current cross equipment;

and determining whether the time conflict exists at the shared port of the crossing equipment behind the current crossing equipment or not according to the updated earliest time and latest time.

Optionally, the instruction carries opening times corresponding to multiple forwarding queues corresponding to the common port, where, in the first traffic and the second traffic that have conflicts, the opening time of the second forwarding queue corresponding to the later-arriving second traffic is determined according to the latest time when the first traffic is completely forwarded out of the common port.

Optionally, any one of the flows is a flow sent according to a certain period.

According to a second aspect of embodiments herein, there is provided a traffic scheduling method and a traffic scheduling apparatus, including: the device comprises a determining module, a conflict detecting module and a sending module; a determining module, configured to determine whether a plurality of traffic flows have a common port of a cross device that passes through the common port, where the common port is an output port through which the plurality of traffic flows;

and the conflict detection module is used for judging whether a time conflict exists when the flow is forwarded out from the shared port of the cross device, if so, the sending module sends an instruction to the cross device, and the instruction is used for indicating the flow passing through the conflict to wait for a preset time.

Optionally, the determining module is further configured to determine a constraint value of a forwarding delay between a sender and a receiver of each flow; determining a difference value between the starting forwarding time of the flow on a sender corresponding to a target path and the arrival time of the flow reaching a receiver, wherein the target path is an reachable path between the sender and a target; if the difference value is larger than the constraint value of the forwarding delay, the target path is abandoned, and other forwarding paths between the sender and the receiver corresponding to the flow are determined again.

Optionally, when determining whether a plurality of traffic has a common port of a cross device that passes through in common, the determining module is specifically configured to determine, for each piece of traffic, a shortest path corresponding to the piece of traffic; comparing according to each path with the shortest flow, and determining whether the shortest path of each flow has a common port of the cross equipment which passes through the shortest path; if the difference value between the starting forwarding time and the arrival time at the receiver on the sender corresponding to the shortest path of the traffic is larger than the constraint value of the forwarding delay, selecting a secondary short path corresponding to the traffic to compare with the shortest paths of other traffic, and determining whether a shared port of the cross device passes through commonly.

Optionally, the collision detection module is configured to, when determining whether there is a time collision when traffic is forwarded on the common port of the cross device, and if there is a collision, send an instruction to the cross device, specifically:

determining the earliest time each flow reaches the common port;

Optionally, the apparatus further comprises:

an updating module, configured to update, according to preset time for which traffic waits at a common port of a current cross device, earliest time for the traffic to pass through an output port of a network device after the current cross device and latest time for the traffic to be completely forwarded out of the output port;

and the conflict detection module is used for determining whether the flow has time conflict at the shared port of the cross device behind the current cross device according to the updated earliest time and latest time.

Optionally, the instruction carries opening times corresponding to multiple forwarding queues corresponding to the common port, where, in the first traffic and the second traffic which have conflicts, the opening time of the second forwarding queue corresponding to the later-arriving second traffic is determined according to the latest time when the first traffic is completely forwarded out of the common port;

any one of the plurality of flows is a flow sent according to a certain period.

According to a third aspect of embodiments herein, there is provided a computer-readable storage medium storing a computer-executable program which, when invoked by a computer, causes the computer to perform any of the methods as provided by the first aspect.

The technical scheme provided by the embodiment of the specification can have the following beneficial effects: the network management equipment determines the time range of the flow possibly staying at the common port by calculating the earliest time of reaching the common port and the latest time of the flow completely forwarded from the common port, if the time ranges corresponding to different flows are overlapped, the conflict of the flow forwarding on the time exists, and the problem can be well solved by a time delay mode.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the specification.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present specification and together with the description, serve to explain the principles of the specification.

Fig. 1 is a schematic flow chart of a traffic scheduling method provided in the present specification;

FIG. 2 is a schematic diagram of a network architecture provided herein;

fig. 3 is a schematic flow chart of a traffic scheduling method according to another embodiment of the present disclosure;

FIG. 4 is a schematic diagram of internal forwarding delays provided herein;

fig. 5 is a flow chart of another traffic scheduling method provided in the present specification;

fig. 6 is a flowchart illustrating a traffic scheduling method according to still another embodiment of the present disclosure;

fig. 7 is a schematic structural diagram of a traffic scheduling apparatus provided in the present specification;

fig. 8 is a schematic structural diagram of a controller provided in the present specification.

Detailed Description

Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present specification. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the specification, as detailed in the appended claims.

Example one

The embodiment provides a traffic scheduling method, which may be applied to a network management device, and the network management device determines a common port of a cross device through which a plurality of traffics pass in a network topology, determines whether a time conflict exists when the plurality of traffics are forwarded out from the common port, and if the time conflict exists, sends an instruction to the cross device to indicate that a later arriving traffic among the traffics with the conflict waits, and forwards the later traffic after waiting for a preset time. By means of waiting for the traffic arriving later, the problem of discarding caused by conflict when the traffic is forwarded out from the common port is avoided.

Before describing the method of this embodiment, this embodiment first describes device attributes that can be acquired by a network management device.

The device attributes may include the following: the identification of the network equipment, the forwarding delay range of the network equipment, the maximum gating interval time, the gating time precision, the identification of the forwarding port, the bandwidth of the forwarding port and the forwarding link delay;

the identifier of the network device may adopt an identifier that can uniquely identify the network device, such as an MAC address, an IP address, or a serial number of the network device;

internal forwarding delay range of network device: the time from the flow going from the input port to the flow arriving at the output port is referred to, and the network management equipment can acquire the internal forwarding delay range of each network equipment from the network equipment; since a series of processing may be performed after the traffic enters from the ingress port, and the processing time for the data stream is not a fixed value, the internal forwarding delay range of the network device is a range interval determined by the minimum value of the forwarding delay of the network device and the maximum value of the forwarding delay of the network device. For example, the time when traffic reaches the ingress port of the network device a is 01:05, after the traffic enters from the ingress port, the network device a may perform processes such as identification and decapsulation on the traffic, and due to device processing performance and the pressure of currently processing other traffic, the earliest time when the network device reaches the egress port after a series of processes is measured in a period of time is 01:10, and the latest time when the network device reaches the egress port is 01:15, and then the minimum value and the maximum value of the forwarding delay of the network device may be determined according to the times when the traffic reaches the ingress port 01:05, 01:10, and 01:15, respectively.

Gating the maximum interval time: in the method provided by the application, the opening time of the forwarding queues can be controlled, and the maximum time from opening to closing of each forwarding queue is called the gated maximum interval time;

the gating time precision is as follows: is the precision of the time length of the forwarding queue opened, for example, the precision is 5us, and then the time length from the forwarding queue opened to the forwarding queue closed is at least an integral multiple of the time precision.

Forwarding delay of the link: the time difference from the output port of the network device to the input port of the next hop device of the network device is the forwarding delay of the link between the network device and the next hop device.

Fig. 1 shows a flowchart of the traffic scheduling method, which may be applied to Software or hardware entities for managing Network devices, such as Network management Software, an SDN (Software Defined Network) controller, and the like.

As shown in fig. 1, the traffic scheduling method includes:

step 101, determining whether a plurality of flows have a common port of a cross device passing through together, wherein the common port is an output port through which the plurality of flows can flow;

there may be many types of traffic flows to forward in the network, some of which may be periodically forwarded, such as in the context of automation control of a plant, some control tasks are periodic. A possible application scenario of the traffic scheduling method provided in this embodiment is that a plurality of traffics are periodically sent traffics.

The present embodiment takes a plurality of traffic as TSN flows as an example for explanation. In the context of TSN traffic, there are roles for each stream of traffic for the sender talker and the receiver listener.

At least one of the sender and the receiver of different traffic is different.

Fig. 2 shows a schematic diagram of a network architecture. In fig. 1, three flows are taken as an example, and the senders of the three flows are talker1, talker2 and talker3 respectively. To simplify the model, the receivers of all three streams are illustrated as listener 4.

The network management device (hereinafter referred to as network management) may obtain a topology map of all network devices managed by the network management device, where the topology map includes connection relationships between the network devices. Meanwhile, the network device can obtain information of a sender and a receiver of each flow, and the network management can search the edge device connected with the sender according to the topological graph by taking the sender of the flow as a starting point, so as to search the network device connected with the edge device until the receiver is searched, and thus, the reachable path of each flow can be searched.

For example, the reachable path of traffic 1 may be determined to include:

talker1->A->D->E->listener4；

talker1->A->B->E->listener4；

talker1->A->C->D->E->listener4。

similarly, it may be determined that the traffic 2 reachable path includes:

Talker2->C->D->E->listener4；

Talker2->C->A->B->E->listener4；

Talker2->C->A->D->E->listener4

Talker2->C->D->A->B->E->listener4；

the traffic 3 reachable path includes:

Talker3->B->E->listener4。

since each flow has a plurality of reachable paths, the shortest paths of the respective flows may be compared first when determining whether there are crossing devices passing through together. Specifically, in implementation, for each flow, the reachable routes of the flow may be arranged according to the time sequence of reaching the receiving party.

In one possible implementation: one of the time when the traffic reaches the receiver may be to test the actual traffic and instruct the forwarding of the subsequent traffic according to the test result.

In another possible implementation: the network manager can estimate the time of each flow reaching the receiver according to the acquired equipment information. Such an implementation will be developed in detail in the following embodiments of the present specification, and will not be described herein again.

If there are multiple shortest paths, one shortest path may be selected randomly or according to other rules, and the method for selecting the shortest path is not limited in this embodiment.

The shortest paths of the three streams are respectively: talker1- > A- > D- > E- > listener4, Talker2- > C- > D- > E- > listener4, and Talker3- > B- > E- > listener4 are taken as examples, then the intersection device between the flow 1 and the flow 2 can be determined to be D, and the intersection device between the flow 1, the flow 2 and the flow 3 can be determined to be E.

It is assumed that there is a common port for traffic 1 and traffic 2 on the cross device D, that is, both traffic 1 and traffic 2 are forwarded out through the same output port on the cross device D; the traffic 1, the traffic 2, and the traffic 3 have a common port on the cross device E, that is, the traffic 1, the traffic 2, and the traffic 3 are forwarded out through the same output port on the cross device E.

Step 103, determining whether there is a time conflict when the traffic is forwarded out at the common port of the cross device, and if there is a conflict, sending an instruction to the cross device, where the instruction is used to instruct the traffic that passes after the conflict to wait for a preset time.

Step 103 can be specifically implemented by the implementation provided in steps 1031 to 1033, as shown in fig. 3.

Step 1031, determining the earliest time each traffic arrives at the common port.

Fig. 4 is a schematic diagram illustrating traffic forwarding, and the processing for the traffic in fig. 4 includes two parts, one part is other processing performed after the traffic enters from an ingress port, where the type of the other processing is not limited, and the traffic arrives at an egress port after the other processing is performed, and waits for forwarding. The time for the network device to perform other processing on the traffic, i.e., the internal forwarding delay, may be 1s or 5 s. Therefore, the earliest time to reach the common port of the crossing device can be determined according to the minimum value of the internal forwarding delay, the link delay and the time when the traffic is processed in the forwarding queue of each network device.

The network manager can obtain the minimum value and the maximum value of the internal forwarding delay of the network equipment from the network equipment, and can determine the range of the internal forwarding delay of the network equipment according to the minimum value and the maximum value;

time ta (i) when traffic is processed in the forwarding queue is the size of traffic/bandwidth of the egress port;

where i is a device identifier, for example, ta (a) is a time when a forwarding queue of traffic in network device a is processed.

Then the earliest time to reach the egress port of network device a from talker1 is t (a) min:

the time when the traffic 1 reaches the entrance of the talk1 device + the minimum value of the internal forwarding delay of the network device a;

the earliest time for Talk1 to reach the egress port of network device B is t (B) min:

the time when the traffic 1 reaches the entry of the talk1 device + the minimum value of the internal forwarding delay of the network device a + the time when the traffic is processed in the forwarding queue of the output port of the network device a + the link delay between the output port of the network device a and the input port of the network device B;

by analogy, the earliest time traffic 1 reaches an egress port of each network device may be determined.

Step 1032, determining the latest time when each traffic is completely forwarded out of the common port;

the latest time when the traffic is completely forwarded out of the common port is determined according to the maximum value of the internal forwarding delay of the network equipment. (the earliest leaving time of the traffic completely forwarded out of the shared port is determined according to the minimum value of the internal forwarding delay of the network device, and in this embodiment, the earliest leaving time of the traffic completely forwarded out of the shared port is temporarily not used).

Specifically, following the above example, the latest time t (a) max when the traffic is completely forwarded out of the network device a is:

the time when the traffic 1 reaches the entry of the talk1 device + the maximum value of the internal forwarding delay of the network device a + the time when the traffic is processed in the forwarding queue of the egress port of the network device a;

by analogy, the latest departure time of the traffic 1 that is completely forwarded on each network device may be calculated.

Step 1033, determine a time range for the flow to dwell at the common port based on the earliest time and the latest time.

A time range, referred to as a dwell time at the shared port in this embodiment, may be determined based on the earliest time traffic arrives at the shared port and the latest time it leaves the device that is forwarded out of the shared port.

Step 1034, if there is an intersection in the time ranges of the multiple flows, an instruction is sent to the intersection device.

The case where the time ranges of the plurality of traffic volumes intersect is described in tables 1 to 5.

Table 1 shows the earliest time that traffic 1 arrives at an egress port of each network device on the flow path, and the latest time that it leaves the egress port. Illustratively, table 1 shows that the earliest time for traffic 1 to reach an egress port of network device a is 01:00, and the latest time for traffic 1 to leave the egress port is 01: 10; the earliest time for traffic 1 to reach the egress port of network device D is 01:11, the latest time to leave the exit port is 01: 21; the earliest time for traffic 1 to reach the egress port of network device E is 01: 22, the latest time to leave the egress port of network device E is 01: 32.

TABLE 1

Flow identifier	Network device identification	Earliest time to arrive at an egress port	Latest time of exit from exit port
				1	A	01:00	01:10
1	D	01:11	01:21
				1	E	01:22	01:32

Table 2 shows the earliest time for traffic 2 to reach an egress port, and the latest time to leave an egress port, of each network device on the flow path.

TABLE 2

Flow identifier	Network device identification	Earliest time to arrive at an egress port	Latest time of exit from exit port
				2	C	01:00	01:10
2	D	01:12	01:23
				2	E	01:23	01:33

Table 3 shows the earliest time for traffic 3 to reach an egress port, and the latest time to leave an egress port, of each network device on the flow path.

TABLE 3

Flow identifier	Network device identification	Earliest time to arrive at an egress port	Latest time of exit from exit port
				3	B	01:10	01:20
3	E	01:21	01:29

Table 1 to table 3 show only the shortest paths corresponding to the respective flows. In practical cases, each flow may correspond to multiple paths, which are not shown one by one. The crossover equipment for each flow can be determined from tables 1 to 3.

Assuming that the output ports of the respective flows on the crossing device are the same, the common port of the crossing device D exists on the flow paths of the flow 1 and the flow 2, the time range determined by the latest time of the flow 1 arriving at and leaving the common port according to table 1 is (01:11, 01:21), and the time range determined by the latest time of the flow 2 arriving at and leaving the common port according to table 2 is (01:12, 01: 23). It can be seen that there is a traffic collision between the two flows when the two flows are forwarded on the common port. The network control device may instruct the crossing device D that the later traffic 2 waits for a predetermined time, for example, to have the traffic 2 forwarded on the common port after 01: 21. The preset time may be determined according to the latest time of the first arriving traffic leaving the common port.

Table 4 shows the earliest time that traffic 1 and traffic 2 arrive at the common port on the crossbar device and the latest time that traffic 2 leaves the common port after waiting for the preset time after issuing the command.

TABLE 4

Flow identifier	Network device identification	Earliest time to arrive at an egress port	Latest time of exit from exit port
				1	D	01:11	01:21
2	D	01:21	01:32

In another possible case, there may be more than one cross device for multiple flows, and if there is a common port of multiple cross devices in the network for multiple traffic flows, on the basis of the foregoing embodiment, the method provided in this embodiment further includes:

step 1035, updating the earliest time of the traffic passing through the output port of the network device after the current cross device and the latest time of the traffic being completely forwarded out of the output port according to the preset time of the traffic waiting at the common port of the current cross device.

Step 1036, determining whether the traffic has a time conflict at the common port of the crossing device after the current crossing device according to the updated earliest time and latest time.

Following the above embodiment, the traffic 2 is forwarded after 9s (from the original 01:12 to 01:21) after the cross device D arrives at the common port.

After waiting for the preset time (9s) for the flow 2, the corresponding time to reach the next hop device is delayed by 9 s.

In this embodiment, since another intersection apparatus E exists in the traffic 1, the traffic 2, and the traffic 3, the earliest time when the traffic 2 arrives at the common port on E and the latest time when the traffic 2 leaves the common port on E after waiting for the preset time can be achieved. Table 5 shows the earliest time when the traffic 2 arrives at the common port of the intersection apparatus E and the latest time when it leaves the common port after waiting for the preset time.

TABLE 5

In this case, steps 1035 and 1036 are actually repeated on the crossing equipment E to perform steps 101 to 1034.

It is determined from table 5 that there is an intersection in the time ranges corresponding to flow 3 and flow 1, and therefore, it is necessary to delay both flow 1 and flow 2 which reach the common port later. After the delay, the earliest time that each flow arrives at the shared port on network device E and the latest time that each flow leaves the shared port on network device E are shown in table 6.

TABLE 6

Flow identifier	Network device identification	Earliest time to arrive at an egress port	Latest time of exit from exit port
				3	E	01:21	01:29
1	E	01:29	01:39
				2	E	01:39	01:49

In the method provided by this embodiment, the network management device determines the time range in which the traffic may stay at the common port by calculating the earliest time when the traffic arrives at the common port and the latest time when the traffic is completely forwarded from the common port, and if the time ranges corresponding to different traffic overlap, it indicates that there is a collision in time when the traffic is forwarded, and this problem can be solved well by a delay manner.

Example two

On the basis of the above embodiments, the present embodiment mainly explains the case of the instruction issued to the crossing device.

In an implementation manner, the instruction carries the opening times corresponding to the multiple forwarding queues corresponding to the common port. And aiming at one forwarding queue, when the opening time is up, allowing the traffic corresponding to the forwarding queue to forward.

In one example, a user may specify a corresponding forwarding queue for traffic. For example, forward queue 7 is designated for traffic 1, forward queue 6 is designated for traffic 2, and forward queue 4 is designated for traffic 3. Of course, a plurality of queues may be assigned to one traffic, and in this embodiment, traffic and forwarding queues are described as one-to-one correspondence. In an alternative embodiment, a fixed queue may be assigned to each flow on each network device, for example, traffic 1 is forwarded on network device a through forwarding queue 7, and traffic 1 is also forwarded on network device D through forwarding queue 7, in this way, when a problem occurs, a specific queue listing the problem may be traced to the source.

And the network management device may further carry, in the instruction, the opening time set for each forwarding queue corresponding to the common port of the cross device, where the opening time of the forwarding queue corresponding to the later-arriving traffic is determined according to the latest time at which the earlier-arriving first traffic is completely forwarded out of the common port. Fig. 7 shows information that may be carried in an instruction issued to the crossing device D. As shown in fig. 7, the open time of the queue corresponding to the later arriving traffic 2 may be determined according to the latest time 01:21 when the earlier arriving traffic 1 is forwarded out of the common port (determined according to table 4).

TABLE 7

Flow identification	Forwarding queue identification	Queue open time
			1	7	01:11
2	6	01:21

In another possible implementation, assigning a forwarding queue to traffic may be implemented by other instructions. That is, table 7 may only carry the traffic identifier and the open time of the forwarding queue, or only carry the identifier of the forwarding queue or the open time of the forwarding queue.

The network management device may determine the open time of the forwarding queue on each network device according to the following formula:

(T (i) max + TA (i) -T (i) min) + { precision- [ ((T (i) max + TA (i) -T (i) min)% precision ] }, wherein T (i) min is the earliest time to reach the egress port of network device i;

t (i) max is the latest time of being completely forwarded out of the output port of the network device i;

ta (i) is the time when traffic is processed in the forwarding queue of network device i;

precision is the gate control time Precision;

% is the remainder symbol.

In the above formula, (t (i) max + ta (i) -t (i) min) is calculated by the latest time when the flow arrives at the egress port on the i-th device, and adding the time when the flow is processed in the forwarding queue of the egress port, the time when the flow is completely forwarded from the egress port at the latest is obtained, and then the earliest time when the flow arrives at the i-egress port of the network device is subtracted, and finally the time when the queue gate control of the flow needs to be opened is obtained.

And (c) performing precision residue on the time calculated according to (T (i) max + TA (i) -T (i) min), wherein the precision residue represents that a plurality of times are left after the time is subjected to integral multiple precision, and the residue is used for calculating a number to be compensated for reducing the precision. Thus, it can be calculated how much the initial gating time is added to be the integral multiple of the precision value.

In the process of forwarding the traffic, in another possible implementation manner: the network management device may instruct the traffic to forward the packets in the order from high priority to low priority according to the time sequence of reaching the common port. Illustratively, if forward queue 7 is the highest priority, forward queue 6 times, and forward queue 4 is the next highest. If the flow 1 arrives at the earliest, the flow 1 is put into a forwarding queue 7 for forwarding, and if the flow 1 is not forwarded completely and the flow 2 arrives, the flow 2 is put into a forwarding queue 6 for waiting for forwarding. Thereby achieving the goal that the later arriving traffic 2 waits for traffic 1.

EXAMPLE III

On the basis of the above embodiment, the method provided by this embodiment can further optimize the way of selecting the path.

On the basis of the first embodiment or the second embodiment, fig. 5 is a schematic flow chart of another traffic scheduling method provided in this specification, and as shown in fig. 5, the method includes:

step 501, determining a constraint value of forwarding delay between a sender and a receiver of each flow.

The network management device may determine topology maps of all network devices managed by the network management device, and specifically, may use an identifier of the network device managed by the network management device as a key, and store the device attribute as a value in a dictionary, so as to become an undirected graph with the device attribute. The content specifically included in the device attribute is introduced in the embodiment, and is not described in detail in this embodiment.

The network management device may receive an identifier of a sender, an identifier of a receiver, an identifier of an edge device connected to the sender, an identifier of an edge device connected to the receiver, a time when the traffic starts to be forwarded, a size of the traffic, a flow identifier, and a constraint value of a forwarding delay of the traffic between the sender and the receiver (e.g., a maximum forwarding delay of the traffic between the sender and the receiver), which correspond to the traffic input by the user.

Step 503, determining a difference value between the forwarding start time of the traffic on the sender corresponding to the target path and the arrival time of the traffic on the receiver, where the target path is an reachable path between the sender and the destination.

The network management device may traverse the dictionary mentioned in step 501 with the sender as a starting point according to the identifiers of the sender and the receiver input by the user, find the edge device connected to the sender, then find the device ID connected to the edge device until the receiver device is found to stop, and store the path to the reachable path list. Repeating this step can obtain all reachable paths of the traffic between the sender and the receiver.

Traversing the network devices on the reachable path in the reachable path list, looking up the value corresponding to the current network device in the dictionary, and determining the time range of the flow staying at the shared port through steps 103-1033. And meanwhile, the forwarding time delay of the link can be determined from the value, the time range of the flow reaching the input port of the next hop device of the current device can be determined according to the forwarding time delay of the link, the time range is used as the time range of the next hop device for starting the flow forwarding, the process is repeated until the calculation is stopped at the receiving party, and the time ranges corresponding to all network devices on the reachable path can be obtained through calculation.

And 505, if the difference is greater than the constraint value of the forwarding delay, discarding the target path, and re-determining other forwarding paths between the sender and the receiver corresponding to the traffic.

For each reachable path, comparing the difference between the obtained time for starting forwarding on the sender and the calculated arrival time for reaching the receiver with the maximum forwarding delay mentioned in step 501, and if the difference exceeds the maximum forwarding delay, discarding the reachable path, that is, the reachable path does not meet the constraint of the maximum forwarding delay.

Paths which do not accord with the time delay constraint condition are removed, and the calculated amount can be reduced when whether cross equipment exists in the flow.

It should be understood that the execution sequence of the steps 501-505 and the steps 101-103, 101-1034 in the first and second embodiments is not limited. For example, the

steps

501 and 505 are executed before the step 101, and may also be executed after the step 505. Executed after step 505, the calculated arrival time of the traffic at the receiver is the time after the delay.

Example four

On the basis of the foregoing embodiments, in order to improve the execution efficiency, on the basis of any one of the first to third embodiments, fig. 6 shows that the shared port of the intersection device that is provided in this specification and that is commonly passed through when determining whether multiple pieces of traffic exist, which may be implemented by adopting the following implementation manners:

step 601, determining the shortest path corresponding to each flow for each flow.

Step 603, comparing according to each path with the shortest flow, and determining whether a common port of the cross device which passes through the shortest path of each flow is available.

An example of finding a common port according to a shortest path is shown in an embodiment, and details are not described in this embodiment.

Step 605, if the difference between the forwarding start time on the sender corresponding to the shortest path of the traffic and the arrival time at the receiver is greater than the constraint value of the forwarding delay, selecting the secondary short path corresponding to the traffic to compare with the shortest paths of other traffic, and determining whether there is a common port of the cross device that passes through together.

In one example, it may be determined that the traffic 2 reachable path comprises:

Talker2->C->D->E->listener4；

Talker2->C->A->B->E->listener4；

Talker2->C->A->D->E->listener4

Talker2->C->D->A->B->E->listener4；

if the difference between the time of arrival at the receiver listener4 and the time of starting forwarding on the sender Talker1 is greater than the constraint value of the forwarding delay of traffic 2 for the shortest path Talker2- > C- > D- > E- > listener4 after the traffic 2 is delayed, the second shortest path Talker2- > C- > a- > B- > E- > listener4 can be selected for comparison with the shortest paths of other traffic.

And finally, after the delay, recording all reachable paths meeting the requirements into a selectable path table, wherein the selectable path table can record sending instructions corresponding to each network device on each reachable path.

In the method provided in this embodiment, in addition to sending the instruction to the crossing device, the network device through which the traffic passes after the crossing device also adaptively issues the instruction to instruct the network device after the crossing device to perform the same delay.

By the method, the shortest and optimal path can be found to forward each flow, and the problem of flow forwarding conflict of the flow on the cross equipment does not exist.

It should be noted that the flow diagram of fig. 6 is further implemented on the basis of the third embodiment, but it should be understood that the method of the present embodiment may not be implemented in steps 501 to 503.

EXAMPLE five

This embodiment provides a traffic scheduling apparatus, which may be used to execute the traffic scheduling method provided in any of the foregoing embodiments, and fig. 7 shows a schematic structural diagram of the traffic scheduling apparatus, as shown in fig. 7, the apparatus includes: a determining module 701, a collision detecting module 702, and a sending module 703;

a determining module 701, configured to determine whether there is a common port of a cross device that passes through a plurality of traffic together, where the common port is an output port through which the plurality of traffic all flows;

a conflict detection module 702, configured to determine whether there is a time conflict when traffic is forwarded out from a common port of the cross device, and if there is a conflict, send a command to the cross device by using a sending module 703, where the command is used to indicate that a traffic waiting time that passes after the conflicting traffic exists.

Optionally, the determining module 701 is further configured to determine a constraint value of a forwarding delay between a sender and a receiver of each flow; determining a difference value between the starting forwarding time of the flow on a sender corresponding to a target path and the arrival time of the flow reaching a receiver, wherein the target path is an reachable path between the sender and a target; if the difference value is larger than the constraint value of the forwarding delay, the target path is abandoned, and other forwarding paths between the sender and the receiver corresponding to the flow are determined again.

Optionally, when determining whether a plurality of traffic has a common port of a cross device that passes through in common, the determining module 701 is specifically configured to determine, for each piece of traffic, a shortest path corresponding to the piece of traffic; comparing according to each path with the shortest flow, and determining whether the shortest path of each flow has a common port of the cross equipment which passes through the shortest path; if the difference value between the starting forwarding time and the arrival time at the receiver on the sender corresponding to the shortest path of the traffic is larger than the constraint value of the forwarding delay, selecting a secondary short path corresponding to the traffic to compare with the shortest paths of other traffic, and determining whether a shared port of the cross device passes through commonly.

Optionally, the collision detection module 702 is specifically configured to, when determining whether there is a time collision when traffic is forwarded on the common port of the cross device, and if there is a collision, send an instruction to the cross device, to:

determining the earliest time each flow reaches the common port;

Optionally, the apparatus further comprises:

an updating module (not shown in the figure), configured to update, according to a preset time that a traffic waits at a common port of a current cross device, an earliest time of an output port of a network device that the traffic passes through after the current cross device and a latest time of the traffic being completely forwarded out of the output port;

the conflict detection module 702 is configured to determine whether a time conflict exists on a common port of a cross device behind the current cross device for the traffic according to the updated earliest time and latest time.

Optionally, any one of the flows is a flow sent according to a certain period.

The present specification also provides a computer-readable storage medium storing a computer-executable program, which, when called by a computer, causes the computer to execute the traffic scheduling method performed by the network management device in any of the above embodiments.

The present disclosure further provides a controller 80, and fig. 8 is a schematic structural diagram of a controller according to another embodiment of the present disclosure, as shown in fig. 8, the controller 80 includes a processor 801 and a memory 802,

the memory 802 is configured to store program instructions, the processor 801 is configured to call the program instructions stored in the memory, and when the processor 801 executes the program instructions stored in the memory 802, the processor is configured to execute any of the traffic scheduling methods provided in the first to fourth embodiments.

The flow scheduling device, the computer-readable storage medium, the process of executing the flow scheduling method by the controller, and the technical effects that can be achieved in this embodiment may refer to the fourth embodiment of the first to fourth embodiments and are not described in detail again.

Furthermore, in the following detailed description, numerous specific details are set forth in order to provide a better understanding of the present disclosure. It will be understood by those skilled in the art that the present disclosure may be practiced without some of these specific details. In some instances, methods, means, elements and circuits that are well known to those skilled in the art have not been described in detail so as not to obscure the present disclosure.

In the embodiments provided in the present disclosure, it should be understood that the disclosed apparatus and method may be implemented in other ways. The apparatus embodiments described above are merely illustrative, and for example, the flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

In addition, functional modules in the embodiments of the present disclosure may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a readable storage medium. Based on such understanding, the technical solution of the present disclosure or portions thereof that contribute to the prior art in essence can be embodied in the form of a software product, which is stored in a readable storage medium and includes several 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 described in the embodiments of the present disclosure. And the aforementioned readable storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.

Claims

1. A traffic scheduling method is applied to a network management device, and comprises the following steps:

2. The method of claim 1, further comprising:

3. The method of claim 1, wherein determining whether there is a common port of a crossroad device that is commonly traversed by multiple traffic comprises:

determining the shortest path corresponding to each flow for each flow;

4. The method of claim 1, wherein determining whether there is a time conflict when traffic is forwarded on a common port of the crossing device, and if there is a conflict, sending an instruction to the crossing device comprises:

determining the earliest time each flow reaches the common port;

5. The method of claim 4, further comprising:

6. The method according to claim 1, wherein the instruction carries opening times corresponding to a plurality of forwarding queues corresponding to the common port, and wherein, among the first traffic and the second traffic which have conflicts, the opening time of the second forwarding queue corresponding to the later-arriving second traffic is determined according to a latest time when the first traffic is completely forwarded out of the common port.

7. The method according to any one of claims 1 to 6, wherein any one of the plurality of traffic is traffic transmitted at a certain period.

8. A traffic scheduling apparatus, comprising: the device comprises a determining module, a conflict detecting module and a sending module; a determining module, configured to determine whether a plurality of traffic flows have a common port of a cross device that passes through the common port, where the common port is an output port through which the plurality of traffic flows;

9. The apparatus of claim 8, wherein the determining module is further configured to determine a constraint value of a forwarding delay between a sender and a receiver of each traffic; determining a difference value between the starting forwarding time of the flow on a sender corresponding to a target path and the arrival time of the flow reaching a receiver, wherein the target path is an reachable path between the sender and a target; if the difference value is larger than the constraint value of the forwarding delay, the target path is abandoned, and other forwarding paths between the sender and the receiver corresponding to the flow are determined again.

10. The apparatus according to claim 8, wherein the determining module, when determining whether there is a common port of a cross device that is passed through by multiple traffic flows in common, is specifically configured to determine, for each traffic flow, a shortest path corresponding to the traffic flow; comparing according to each path with the shortest flow, and determining whether the shortest path of each flow has a common port of the cross equipment which passes through the shortest path; if the difference value between the starting forwarding time and the arrival time at the receiver on the sender corresponding to the shortest path of the traffic is larger than the constraint value of the forwarding delay, selecting a secondary short path corresponding to the traffic to compare with the shortest paths of other traffic, and determining whether a shared port of the cross device passes through commonly.

11. The apparatus of claim 8, wherein the collision detection module, when determining whether there is a time collision when traffic is forwarded on the common port of the cross device, and if there is a collision, sends an instruction to the cross device, is specifically configured to:

determining the earliest time each flow reaches the common port;

12. The apparatus of claim 11, further comprising:

13. The apparatus according to claim 8, wherein the instruction carries opening times corresponding to a plurality of forwarding queues corresponding to the common port, wherein, among the first traffic and the second traffic which have conflicts, the opening time of the second forwarding queue corresponding to the later-arriving second traffic is determined according to a latest time when the first traffic is completely forwarded out of the common port;

any one of the plurality of flows is a flow sent according to a certain period.

14. A computer-readable storage medium, characterized in that it stores a computer-executable program which, when invoked by a computer, causes the computer to perform the method according to any one of claims 1 to 7.