Service orchestration method and apparatus, storage medium, and program product

Abstract

A service orchestration method and apparatus, a storage medium, and a program product, relating to the technical field of industrial Internet. The method comprises: acquiring service flow characteristics and delay margins of service flows sent by communication terminals, and wireless scheduling data of a network device in communicational connection with the communication terminals (S100); on the basis of the wireless scheduling data, the service flow characteristics and the delay margins, determining target scheduling time slots for the network device to schedule the service flows (S200); and controlling the network device to schedule, within the target scheduling time slots, corresponding service flows, and orchestrating the gating operation times of the service flows corresponding to the target scheduling time slots (S300).

Description

Method, device, storage medium and program product for service orchestration

[0001] Cross-reference to related applications

[0002] The present application is based on the Chinese patent application No. 202410950620.5, filed on July 15, 2024, and claims priority to the Chinese patent application No. 202410950620.5, the entire contents of which are incorporated herein by reference. TECHNICAL FIELD

[0003] Embodiments of the present application relate to, but are not limited to, the technical field of industrial internet, and in particular to a method, device, storage medium and program product for service orchestration. BACKGROUND

[0004] In the industrial internet, there is often a problem of large air interface delay between the communication terminal and the network device on the base station side. When the service period of the communication terminal is short, and the scheduling time of the network device is long, there may be a delay of more than one service period. In the industrial internet, there are often multiple communication terminals communicating with the same network device, and when multiple communication terminals randomly access, there are two situations: one is that the number of services scheduled by the network device in some time slots is large, causing scheduling congestion and increasing the scheduling time of the network device; the other is that the number of services scheduled by the network device in some time slots is small, and some time slots of the network device are idle. Therefore, the utilization rate of wireless resources on the network device side is low, and the service delay of the communication terminal is longer. Therefore, how to orchestrate the service flows of multiple communication terminals in the industrial internet to improve the utilization rate of wireless resources of the network device while reducing the service delay of the communication terminal is a problem to be solved. SUMMARY

[0005] The following is a summary of the subject matter described in detail herein. This summary is not intended to limit the scope of the claims.

[0006] Embodiments of the present application provide a method, device, storage medium and program product for service orchestration.

[0007] In a first aspect, the method for service orchestration provided by the embodiments of the present application comprises: obtaining service flow characteristics and delay margin of service flows sent by a communication terminal, and wireless scheduling data of a network device in communication connection with the communication terminal; determining target scheduling time slots for the network device to schedule each of the service flows according to the wireless scheduling data, the service flow characteristics and the delay margin; controlling the network device to schedule the corresponding service flow in each of the target scheduling time slots, and orchestrating the gating operation time of the service flow corresponding to the target scheduling time slot.

[0008] Secondly, according to the network device provided in the embodiments of this application, it includes: a latency measurement module configured to acquire the latency margin of a service flow sent by a communication terminal; a service flow feature learning module configured to acquire the service flow features of the service flow; and a cross-layer orchestration module configured to determine the target scheduling time slot for scheduling each of the service flows by the network device based on the wireless scheduling data, the service flow features, and the latency margin; control the network device to schedule the corresponding service flow within each target scheduling time slot; and orchestrate the gating operation time of the service flow corresponding to the target scheduling time slot.

[0009] Thirdly, the network apparatus provided according to the embodiments of this application includes: at least one processor; at least one memory for storing at least one program; and when at least one of the programs is executed by at least one of the processors, implementing the service orchestration method as described in any one of the first aspects.

[0010] Fourthly, the computer-readable storage medium provided according to the embodiments of this application stores computer-executable instructions, which are used to perform the service orchestration method described in any of the first aspects.

[0011] Fifthly, the computer program product provided according to the embodiments of this application includes a computer program or computer instructions, the computer program or computer instructions being stored in a computer-readable storage medium, a processor of a network device reading the computer program or computer instructions from the computer-readable storage medium, and the processor executing the computer program or computer instructions to cause the network device to perform the service orchestration method as described in any of the first aspects. Attached Figure Description

[0012] Figure 1 is a schematic diagram of a framework of an embodiment of the network device provided in this application;

[0013] Figure 2 is a schematic diagram of another embodiment of the network device provided in this application;

[0014] Figure 3 is a flowchart illustrating the business orchestration method provided in this application;

[0015] Figure 4 is a schematic diagram of an embodiment of the communication terminal selection in the service orchestration method provided in this application;

[0016] Figure 5 is a schematic diagram of another embodiment of the communication terminal selection in the service orchestration method provided in this application;

[0017] Figure 6 is a schematic diagram showing the positions of the gating system, communication terminal, and network device scheduler in the service orchestration method provided in this application.

[0018] Figure 7 is a schematic diagram illustrating an embodiment of the business orchestration method provided in this application;

[0019] Figure 8 is a schematic diagram illustrating another embodiment of the business orchestration method provided in this application;

[0020] Figure 9 is a schematic diagram of the device hardware structure corresponding to the business orchestration method provided in this application. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0022] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0023] The flowcharts shown in the accompanying drawings are merely illustrative and do not necessarily include all content and operations / steps, nor do they necessarily have to be performed in the described order. For example, some operations / steps can be broken down, while others can be combined or partially combined; therefore, the actual execution order may change depending on the specific circumstances.

[0024] The following is a brief description of the terminology used in the embodiments of this application.

[0025] UE: The full name of UE is User Equipment, which refers to communication terminals in the following text.

[0026] TSN stands for Time-Sensitive Networking.

[0027] DS-TT: The full name of the English term is Device-side TSN translator, which is a TSN converter on the device side.

[0028] QBV stands for Queue-Based VLAN, which is also known as queue-based virtualization.

[0029] In the Industrial Internet, communication terminals and network devices on the base station side often suffer from high air interface latency. When the service cycle of the communication terminal is short, while the scheduling time of the network device is long, there may be a latency exceeding one service cycle. For example, if the packet transmission time of the communication terminal does not match the scheduling time of the network device, it will increase the air interface latency. Furthermore, in the Industrial Internet, there are often multiple communication terminals communicating with the same network device. When multiple communication terminals randomly access the network, there are two scenarios: one where the network device schedules a large number of services in some time slots, causing scheduling congestion and increasing the scheduling time; the other where the network device schedules fewer services in some time slots, leaving some time slots idle. Therefore, the utilization rate of wireless resources on the network device side is low, and the service latency of the communication terminals is even longer. Therefore, how to orchestrate the service flows of multiple communication terminals in the Industrial Internet to improve the utilization rate of wireless resources of the network device while reducing the service latency of the communication terminals is an urgent problem to be solved. Based on this, embodiments of this application provide a service orchestration method, apparatus, storage medium, and program product that can orchestrate the service flows of multiple communication terminals, thereby improving the wireless resource utilization of network devices while reducing the service latency of communication terminals.

[0030] Referring to Figure 1, an embodiment of this application provides a network device, which includes:

[0031] The delay measurement module is configured to acquire the delay margin of the service stream sent by the communication terminal.

[0032] The business flow feature learning module is configured to acquire the business flow features of the business flow.

[0033] The cross-layer orchestration module is configured to determine the target scheduling time slots for network devices to schedule each service flow based on wireless scheduling data, service flow characteristics, and latency margin; control the network devices to schedule the corresponding service flows within each target scheduling time slot; and orchestrate the gating operation time of the service flows corresponding to the target scheduling time slots.

[0034] It should be noted that the network device can be a network device within the base station, such as a single board integrated into a network device, or a network device like a BBU. The network device can also be a device outside the base station, such as an edge computing node with computing capabilities. Therefore, this application does not limit the specific type of network device.

[0035] It should be noted that the gating operation time refers to the gating opening time and the length of the gating opening time of the scheduler that controls the flow of services to the network device. When communication terminals and network devices communicate, time slot alignment is often performed. Therefore, after determining the target scheduling time slot, the gating operation time of the service flow can also be determined based on that target scheduling time slot. The specific determination method will not be elaborated upon in this embodiment.

[0036] Understandably, in practical applications, some communication terminals require service orchestration, while others do not. In some embodiments, each service flow of each communication terminal under the same network device is orchestrated through a cross-layer orchestration module. The target scheduling time slot for terminals requiring no service orchestration is their initial scheduling time slot, requiring no gating control. The target scheduling time slot for terminals requiring service orchestration is re-determined based on wireless scheduling data, service flow characteristics, and latency margin, satisfying the latency margin requirement, and requires gating control based on the gating operation time. In other embodiments, the target scheduling time slot is determined only for the service flows of communication terminals requiring service orchestration.

[0037] It should be noted that in some embodiments, the network device further includes a gating module. The gating module is configured to perform gating control on multiple communication terminals accessing the network device, so that the service flow of each communication terminal is controlled to send the service flow according to the gating operation time determined by the cross-layer orchestration module, so that the network device can receive the corresponding service flow in each target scheduling time slot. As shown in Figure 2, the gating module is set to TSN gating.

[0038] It should be noted that the service flow characteristics reflect the original characteristics of the service flow, such as the size of the data packets generated by the service flow, the data generation period, and the time it takes for the data to arrive at the network device without gating processing.

[0039] It should be noted that wireless scheduling data is used to record data related to wireless resource scheduling by network devices. In some embodiments, wireless scheduling data includes scheduler parameters, wireless frame structure data, system time and air interface time synchronization information, etc. The scheduler parameters correspond one-to-one with the wireless frame structure and represent the scheduling capability of each time slot. In some embodiments, the scheduler parameters record the number of scheduling slots and the amount of scheduling data. It should be noted that the allocation of uplink and downlink time slots, the supported number of scheduling slots, and the amount of scheduling data vary under different wireless frame structures.

[0040] It should be noted that the embodiments of this application do not limit the number of communication terminals connected to a network device. When the network device and the network equipment of the base station are two independent devices, the embodiments of this application also do not limit the number of network equipment corresponding to a network device. Those skilled in the art can selectively set the number of network devices according to actual needs.

[0041] In some embodiments, as shown in Figure 2, the network device includes a latency measurement module, a service flow feature learning module, a cross-layer orchestration module, and a gating module, wherein the gating module is configured as a TSN gating. The network device and the base station's network equipment are two independent devices. The specific description is as follows.

[0042] The cross-layer orchestration module and gating module can determine the target scheduling time slot based on service flow characteristics, air interface scheduling frame structure, and scheduler parameters, and then orchestrate the service flow based on radio resources according to the target scheduling time slot. The cross-layer orchestration module simultaneously sends the orchestration data obtained based on the target scheduling time slot to the network equipment of TSN gating and base station.

[0043] The service flow feature learning module acquires the service flow features of each service flow. These features can be configured through an interface, or in some embodiments, they can be learned in real time based on the service flows sent by the communication terminal (e.g., using machine learning methods, statistical models, etc.). Service flow features include packet size, packet period, and packet arrival time.

[0044] The latency measurement module is configured to determine the latency margin for each service flow. This latency margin can be determined based on end-to-end latency measurements at the user or service flow level. This embodiment of the application does not limit the specific measurement method used for latency measurement. By utilizing the latency margin, during service orchestration, when scheduling resources for a certain time slot are limited, service flows with latency margins can be shifted in the time domain, thereby achieving peak-shifting scheduling when multiple communication terminals access the service simultaneously.

[0045] TSN gating manages the DS-TT QBV gating of multiple access communication terminals, ensuring that the gating of each service flow of each communication terminal is performed according to the scheduled time points. In other words, service flow transmission is based on the gating operation time of each service flow.

[0046] The network equipment of the base station provides wireless scheduling data to the cross-layer orchestration module, including but not limited to wireless frame structure data, synchronization information of system time and air interface time, cell information, scheduler parameters, etc.

[0047] The cross-layer orchestration module, based on the acquired service flow characteristics {packet size, packet period, packet arrival time}, converts the service flow arrival time (i.e., system time, such as hh; mm:ss:us) into air interface time (SFN & slot) based on the synchronization information of the system time and air interface time reported by the base station. It then combines the air interface time with the radio frame structure data in the radio scheduling data (e.g., if the network device uses a DS frame structure, it combines the downlink and uplink time slot data of the DS frame structure; if the network device uses a 5ms frame structure, it combines the downlink and uplink time slot data of the 5ms frame structure; if the network device uses a 2.5ms single-period frame structure, it combines the downlink and uplink time slot data of the 2.5ms single-period frame structure; if the network device uses a 2.5ms double-period frame structure, it combines the downlink and uplink time slot data of the 2.5ms double-period frame structure; if the network device uses a 1D3U frame structure, it combines the downlink and uplink time slot data of the 1D3U frame structure) to determine the initial scheduling time slot for the service flow on the radio frame. Then, based on the service flow cycle and the initial scheduling time slot, the service flow is orchestrated onto appropriate time slots in the radio frame structure. If multiple service flows conflict in a certain time slot of the frame structure, based on the delay margin measured by the service layer, and within the allowable delay range, the service flows with delay margins are shifted backward from the initial scheduling time slot of the frame structure to find suitable time slots. This achieves staggered orchestration of multiple service flows, thus orchestrating the service flow at the service layer onto appropriate target scheduling time slots in the frame structure. Finally, the orchestration data obtained based on the target scheduling time slot is sent to the TSN gating and the base station scheduler, thereby achieving coordinated transmission and reception of service flows. In this way, network devices can prepare time-domain and frequency-domain resources in advance to directly schedule service flows, shorten scheduling waiting latency, and fully utilize air interface resources.

[0048] It is understood that, referring to FIG3, the service orchestration method provided according to the embodiments of this application includes steps S100 to S300.

[0049] Step S100: Obtain the service flow characteristics and latency margin of the service flow sent by the communication terminal, as well as the wireless scheduling data of the network device connected to the communication terminal.

[0050] Step S200: Based on wireless scheduling data, service flow characteristics, and latency margin, determine the target scheduling time slot for each service flow to be scheduled by the network device.

[0051] Step S300: Control the network device to schedule the corresponding service flow in each target scheduling time slot, and arrange the gating operation time of the service flow corresponding to the target scheduling time slot.

[0052] Therefore, by using the service flow characteristics and latency margins of each service flow of each communication terminal, as well as the wireless scheduling data of the network device, the target scheduling time slots for scheduling each service flow of the network device can be determined. Based on the target scheduling time slots, service orchestration is achieved by controlling the network device to schedule the corresponding service flow and controlling the gating operation time of the service flow. At this time, the transmission of the service flow at the service layer is associated with the reception and processing at the wireless layer of the network device, so that the network device can control the transmission timing of the service flow based on wireless resources. Furthermore, the introduction of latency margins can further ensure that the latency of the service flow meets the requirements during the service orchestration process. Therefore, the embodiments of this application can perform service orchestration for each service flow of multiple communication terminals, thereby improving the utilization rate of wireless resources of the network device while reducing the service latency of the communication terminals.

[0053] Understandably, the service orchestration method is applied to computing power devices or network devices. For example, a network device can execute steps S100 to S300 to determine the target scheduling time slots for the service flows of each communication terminal connected to it, then schedule the corresponding service flows within each target scheduling slot, and orchestrate the gating operation time of the service flows corresponding to the target scheduling time slots.

[0054] In this application, both computing power equipment and network equipment are types of network devices.

[0055] Understandably, service flow characteristics are the raw data characteristics of a service flow, where a service flow represents a data stream formed by multiple consecutive data packets. In some embodiments, service flow characteristics include at least packet arrival time. In some embodiments, service flow characteristics include packet arrival time and packet size, thereby allowing the number of available scheduling slots in the initial scheduling time slots to be determined based on packet size and packet arrival time. In other embodiments, service flow characteristics include packet arrival time, packet size, and packet period, thereby allowing the number of delayed time slots to be determined based on packet period and packet arrival time, and further allowing the target scheduling time slot to be determined based on service flow characteristics. Service flow characteristics can be configured or dynamically learned. In some embodiments, the packet size and packet period of the service flow of the communication terminal are fixed, and the network environment is relatively stable, so service flow characteristics can be directly configured. In some embodiments, the packet size and packet period of the service flow are dynamically changing, or the network environment is complex, so a real-time learning method can be used. In other embodiments, any of the above-mentioned methods of obtaining service flow characteristics can be enabled through configuration.

[0056] It should be noted that the latency margin represents the allowable delay time for the service flow. The latency margin can be represented by a duration value or a time slot. In this application embodiment, there is no restriction on the specific way the latency margin is represented. This application embodiment does not restrict the method of obtaining the latency margin; it can be manually configured or calculated based on a preset latency threshold and the latency result after real-time measurement of the end-to-end latency result.

[0057] In some embodiments, it is assumed that communication terminals A, B, and C all communicate with network equipment D of the base station. Communication terminal A has service flow configuration S1; communication terminal B has service flow configurations S2 and S3; and communication terminal C has service flow configurations S4, S5, and S6. Taking the example that communication terminals A, B, and C all participate in service orchestration, the service flow characteristics and delay margins of S1, S2, S3, S4, S5, and S6 are obtained respectively. The radio scheduling data of D is obtained, thereby determining the target scheduling time slots for each of S1, S2, S3, S4, S5, and S6. Therefore, S1 can be scheduled in the target scheduling time slot corresponding to S1, S2 in the target scheduling time slot corresponding to S2, S3 in the target scheduling time slot corresponding to S3, S4 in the target scheduling time slot corresponding to S4, S5 in the target scheduling time slot corresponding to S5, and S6 in the target scheduling time slot corresponding to S6. Furthermore, based on the respective target scheduling time slots of S1, S2, S3, S4, S5, and S6, the gating operation times sent by S1, S2, S3, S4, S5, and S6 to D are arranged to ensure that S1, S2, S3, S4, S5, and S6 arrive at the network device within their respective target scheduling time slots; S4, S5, and S6 arrive at the network device within their respective target scheduling time slots.

[0058] It should be noted that when the executing entity of steps S100 to S400 is a network device, the gating operation time in step S400 represents the time when the module caching the service flow sends the data to the scheduler of the network device. When the executing entity of steps S100 to S400 is another computing power device, such as an edge computing power node, the gating operation time in step S400 represents the time when the computing power device sends the data to the network device.

[0059] Understandably, step S200, based on wireless scheduling data, service flow characteristics, and latency margin, determines the target scheduling time slots for each service flow to be scheduled by the network device, including:

[0060] - Determine the initial scheduling slots for each service flow;

[0061] -Based on the scheduler parameters and service flow characteristics in the wireless scheduling data, determine the first service flow to be adjusted in the initial scheduling time slot and the remaining service flows from each service flow;

[0062] -Based on the latency margin of the first service flow, determine the target scheduling time slot of the first service flow;

[0063] - Use the initial scheduling slot of the remaining service flow as the target scheduling slot of the remaining service flow.

[0064] Understandably, the target scheduling slot is a slot following the initial scheduling slot, and the difference in duration between the target and initial scheduling slots is less than or equal to the delay margin.

[0065] It should be noted that in some embodiments, the service flow characteristics include at least the packet arrival time (i.e., the system time when the first data packet in the service flow arrives). The radio scheduling data includes synchronization information of system time and air interface time. Therefore, the packet arrival time can be converted into air interface time based on the synchronization information of system time and air interface time, thereby matching the time slot of the radio frame with the closest time to that air interface time. In this case, the time slot is the initial scheduling time slot. In other embodiments, the service flow characteristics also include packet period. Since different packet periods will lead to different distributions of service flows in each radio frame, for example, due to a short packet period, multiple time slots of a radio frame may need to process the same service flow. Therefore, the impact of packet period must also be considered when confirming the target scheduling time slot; for example, the time slots of multiple radio frames can be co-arranged, etc.

[0066] It should be noted that scheduler parameters are used to characterize the scheduling capability of a single time slot, such as the number of supported schedules and the amount of scheduling data.

[0067] It should be noted that the scheduling capacity of the initial scheduling slot corresponding to the first service flow is insufficient, resulting in multiple service flows sharing the same initial scheduling slot, which leads to situations where some service flows cannot be scheduled. If the scheduler parameters include the number of scheduling flows, and the scheduling data volume shows that the initial scheduling slot of service flow 1 only supports two-flow scheduling, and within the current orchestration cycle, the initial scheduling slots of service flows 2, 3, and 4 are all the same as that of service flow 1 (slot 1), then two service flows need to be selected as the first service flow for target scheduling slot confirmation. Assuming that service flows 1 and 2 are the first service flows, and service flows 3 and 4 are the remaining service flows, then the scheduling slots for service flows 1 and 2 will be adjusted to slots after the initial scheduling slots, assuming service flow 1 is adjusted to slot 2 and service flow 2 is adjusted to slot 3. Then, the target scheduling time slot for service flow 1 is time slot 2, the target scheduling time slot for service flow 2 is time slot 3, and the target scheduling time slots for service flows 3 and 4 remain time slot 1. In some other embodiments, the scheduler parameters also include the amount of scheduling data. For example, if the scheduling quantity for time slot 1 is 2, and two service flows initially have time slot 1, but the total data volume of the two service flows is greater than the scheduling data volume of time slot 1, then the scheduling time slot of one of the service flows needs to be adjusted.

[0068] Understandably, the first service flow can also be determined based on service priority. Therefore, based on the scheduler parameters and service flow characteristics in the radio scheduling data, the first service flow to be adjusted in the initial scheduling slot, as well as the remaining service flows, are determined from each service flow, including:

[0069] - Sort each business flow according to its business priority and latency margin to obtain the traversal sort of each business flow;

[0070] - Based on the traversal sorting, each service flow is traversed. During the traversal process, the first service flow is determined according to the scheduler parameters and the service flow characteristics of the service flows that share the same initial scheduling time slot with the traversed service flows.

[0071] - Treat all business flows other than the first business flow as the remaining business flows.

[0072] It should be noted that the target scheduling time slot of the first service flow can be determined based on the delay margin of the first service flow after each first service flow is determined, or the target scheduling time slot of each first service flow can be determined separately after all the first service flows are determined. This application does not limit the implementation of this method.

[0073] It is understandable that, firstly, the business flows can be sorted based on latency margin, and then the business flows with the same latency margin can be sorted by business priority. Alternatively, they can be sorted based on business priority first, and then the business flows with the same business priority can be sorted by latency margin. In this regard, the embodiments of this application do not restrict how the business flows are sorted.

[0074] In some embodiments, if the process iterates to service 1, and the service flows that share the same initial scheduling time slot with service 1 are service 2 and service 3, then it can be determined whether service 1 is the first service flow based on the service flow characteristics of service 1, service 2, and service 3 and the scheduler parameters.

[0075] Understandably, traversal can start from high-priority business flows or low-priority business flows. The method for determining whether business flow 1 is the first business flow can be exemplified as follows: If business flow 1 has a latency margin and its corresponding initial scheduling slot has business flows 2 and 3 scheduled, and the total number of business flows (1, 2, and 3) is greater than the scheduling capacity supported by the initial scheduling slot, and the total data volume of business flows 1, 2, and 3 is greater than the scheduling data volume supported by the initial scheduling slot, then business flow 1 is determined to be the first business flow.

[0076] Understandably, the scheduler parameters include the number of schedulers; based on the latency margin of the first service flow, the target scheduling time slot for the first service flow is determined, including:

[0077] -The initial scheduling time slot and the time slots in the same radio frame as the initial scheduling time slot with an interval of less than or equal to the delay margin are all considered as candidate time slots;

[0078] - Determine the available scheduling quantity of candidate time slots based on the scheduling quantity of candidate time slots and the initial scheduling time slots of each service flow;

[0079] - Select one candidate time slot with available scheduling slots as the target scheduling time slot for the first service flow.

[0080] It should be noted that in some embodiments, the amount of scheduling data exceeded by multiple service flows may occur after aggregation. Therefore, the candidate time slot must also satisfy the requirement that the remaining available scheduling data is greater than the packet size of the first service flow. In other embodiments, there is no situation where the scheduling data amount is exceeded, so it is not necessary to determine the scheduling data amount. Those skilled in the art can selectively configure this according to the actual situation.

[0081] Taking the allocation of target scheduling time slots based on service flow priority as an example, the service flow characteristics and priorities are shown in Table 1 below.

[0082] Table 1

[0083] Assuming that orchestration is only performed on communication terminals that meet preset orchestration conditions, including support for deterministic scheduling cells and support for guaranteed services, as shown in Figure 4, the communication terminals in deterministic scheduling cells are {UE1, UE2, UE3, UE4, UE5, UE6}, which means the terminals that meet the orchestration conditions are {UE1, UE2, UE3, UE4, UE5, UE6}. The frame structure adopts the DS structure (i.e., slot type 0 indicates downlink 0, slot type 1 indicates uplink), and each slot is 0.5ms; the scheduler parameters are such that each slot can only schedule 2 communication terminals, and each slot can only schedule 10Kbytes of data. Then, for UE1, UE2, UE3, UE4, UE5, and UE6, without orchestration, the situation is as shown in Table 2 below (where arrival time is represented by arrival slot, period is in ms, packet size is in bytes, and delay margin is in ms).

[0084] Table 2

[0085] As shown in Table 2 above, without orchestration, since time slot 0 can only schedule two service flows (i.e., the scheduling quantity is 2), the network device can ultimately only schedule 4 service flows: two service flows in time slot 0, one service flow in time slot 2, and one service flow in time slot 4.

[0086] Referring to Table 1 for scheduling, the target scheduling time slots can be determined as shown in Table 3 below.

[0087] Table 3

[0088] At this point, as shown in Table 3 above, the scheduling time slot for f1 is adjusted from the initial scheduling time slot 0 to time slot 4; the scheduling time slot for f2 is adjusted from the initial scheduling time slot 0 to time slot 2. Now, the network device can schedule six service flows.

[0089] At this point, for example, UE3 and UE4 open gating at the system time corresponding to time slot 0, allowing the service flow to be sent out. The network device prepares scheduling resources for 200 bytes and two service flows in time slot 0. Specifically, time slots 4, 2, and 0 can be converted to system time as gating operation times. At this time, service orchestration for f1, f2, f3, f4, f5, and f6 can be completed based on each gating operation time.

[0090] Understandably, step S300, which controls the network device to schedule the corresponding service flows within each target scheduling time slot, includes:

[0091] - Generate orchestration data for network devices based on the target scheduling time slots;

[0092] -Based on the orchestration data of the network devices, control the network devices to schedule the service flows corresponding to each target scheduling time slot within the preset orchestration period.

[0093] It should be noted that in some embodiments, an orchestration period is set, and no business orchestration or gating control is performed on the business flows outside the orchestration period. For example, from 00:00 to 05:00 every day, periodic orchestration is stopped, and a learning period for the original business model is carried out to obtain the business flow characteristics of each business flow. During the original business model learning period, the historical data of {UE1,UE2,UE3,UE4,UE5,UE6,UE7} is cleared, and business flow data is collected again to learn the business characteristics of {UE1,UE2,UE3,UE4,UE5,UE6,UE7}; UE9 does not have gating, so its business model remains the original business model and therefore does not require processing.

[0094] Understandably, the gating operation time of the service flow corresponding to the orchestration target scheduling time slot in step S300 includes:

[0095] - Generate orchestration data for communication terminals based on the target scheduling time slots;

[0096] - Within the preset orchestration cycle, the orchestration data is forwarded to the preset gating module to determine the gating operation time of the business flow.

[0097] It should be noted that by adding a separate gating module to control the gating operation time, the convenience of management and control can be improved. Furthermore, by setting an orchestration cycle, business orchestration can be made more flexible.

[0098] Understandably, before obtaining the service flow characteristics and latency margin of the service flow sent by the communication terminal, the method also includes:

[0099] - Determine whether the communication terminal supports service orchestration processing;

[0100] - Determine if the communication terminal has the capability to guarantee service transmission;

[0101] - Determine whether the communication terminal supports deterministic scheduling cells.

[0102] Based on wireless scheduling data, service flow characteristics, and latency margin, the target scheduling time slots for each service flow to be scheduled by network devices are determined, including:

[0103] When the communication terminal supports service orchestration processing, there is guaranteed service transmission, and cells that support deterministic scheduling, the target scheduling time slots for network equipment to schedule each service flow are determined based on wireless scheduling data, service flow characteristics, and latency margin.

[0104] It should be noted that supporting service orchestration processing means that the service flow of the communication terminal will be uniformly managed and sent to the network device with other communication terminals. For example, if the service flow of the communication terminal is sent to the TSN gate, it means that service orchestration processing is supported.

[0105] It should be noted that the guarantee service is a service with high requirements for the accuracy of service loss. If it is necessary to ensure that the service flow is 100% scheduled, it can be regarded as a guarantee service.

[0106] It should be noted that deterministic scheduling of cells means that fixed cells can be used for access.

[0107] Therefore, by judging the orchestration conditions of the communication terminals that orchestrate the service flows, the orchestration effect can be further ensured.

[0108] In some embodiments, as shown in Figure 5, the UEs initiating the service guarantee are: {UE1, UE2, UE3}; the UEs in cells supporting deterministic scheduling are: {UE1, UE2, UE4}; and the UEs supporting TSN gating are: {UE1, UE3, UE4}. Therefore, the UE that meets the orchestration conditions is: {UE1}. Only the target scheduling time slot needs to be determined for the service flow of UE1. The rest remain unchanged.

[0109] Understandably, when the timestamp carried by the business flow represents the time when the business flow was generated, the methods include:

[0110] - Receive service streams sent by communication terminals;

[0111] - Extract features from the business flow to obtain business flow features, including packet size, packet period, and packet arrival time.

[0112] It should be noted that the timestamp carried by the service flow, representing the time when the service flow was generated, indicates that the timestamp of the service flow does not include the time of gating processing. When the service flow characteristics are learned in real time, service orchestration and the learning of service flow characteristics are independent and can be performed simultaneously without affecting each other. Taking the example of communication terminals all using TSN gating for gating management, as shown in Figure 6, a DS-TT is deployed on the terminal device side to interact with the TSN gating of the network device. The service flow of the communication terminal first enters its own DS-TT before being sent to the network device, where gating operations are performed. When the timestamp carried by the service flow represents the time when the service flow was generated, it means that the timestamp is the timestamp set before entering the DS-TT.

[0113] In some embodiments, the module diagram shown in Figure 2 is used as the execution subject of the orchestration method. Taking the periodic operation of business orchestration as an example, the business orchestration method is shown in Figure 7.

[0114] S1: The cross-layer orchestration module subscribes to the terminal information of the communication terminals that support the gated TSN gate. The terminal information is such as: {UE ID}.

[0115] S2: The cross-layer orchestration module subscribes to the service flow characteristics of the communication terminal from the service flow characteristic learning module. Service flow characteristics include: {packet size, packet period, packet arrival time}.

[0116] S3: The cross-layer orchestration module subscribes to radio scheduling data from the base station's network equipment. The radio scheduling data includes: {frame structure information, synchronization information of system time and air interface time, cell information, scheduler parameters}.

[0117] S4: The cross-layer orchestration module subscribes to the delay measurement module for delay measurement results, such as: {RTTdelay}.

[0118] As shown in Figure 7, service orchestration is performed periodically. When the period ends, the cross-layer orchestration module starts service orchestration based on the information collected by S1 to S4. Service orchestration is used to ensure the service flows that need to be guaranteed. Combining the scheduler's capabilities (how many users can be scheduled in one time slot) and frame structure information (such as DS frame structure, 5ms frame structure, 2.5ms single-cycle frame structure, 2.5ms double-cycle frame structure, 1D3U frame structure, etc.), it first determines the time slot where the first packet is located based on the arrival time of the service flow. Then, it orchestrates the services on appropriate time slots according to the packet period (e.g., all uplink services need to be placed in the uplink time slot). If a multi-user conflict occurs in a time slot (e.g., due to limited frequency domain resources), then based on the delay margin, within the allowable delay range, services with delay margins are postponed in the time slots of the frame structure to find suitable time slots (with sufficient frequency domain resources) to obtain the target scheduling time slots for each service flow.

[0119] The orchestration data obtained based on the scheduling time slots of each target is simultaneously sent to the TSN gating system to ensure that each communication terminal sends service flows according to the orchestration data; and also sent to the scheduler of the network equipment at the base station, which prepares scheduling resources according to the orchestration data. For example, the following is sent to the TSN: {base time, orchestration period, opening time of each data packet, data packet size}; and to the base station: {base time, orchestration period, time offset of each data packet, data packet size}.

[0120] Understandably, when the timestamp carried by the business flow includes the gating processing duration, the method also includes:

[0121] -Continuously preset the duration to stop the gating operation time of the business flow corresponding to the target scheduling time slot;

[0122] - Acquire the service flow sent by the communication terminal within a preset time period;

[0123] - Extract features from the received service flow to obtain service flow features;

[0124] -When the preset duration is reached, the target scheduling time slots for scheduling each service flow of the network device are re-determined based on wireless scheduling data, service flow characteristics and latency margin, and the network device is controlled to schedule the corresponding service flow within each target scheduling time slot, as well as to arrange the gating operation time of the service flow corresponding to the target scheduling time slot.

[0125] Understandably, when the gating operation time corresponding to the target scheduling time slot is stopped, the gating operation fails.

[0126] In some embodiments, as shown in Figure 8, the timestamp carried in the business flow is after gating; that is, the timestamp carried in the business flow is the time after gating, including the gating processing time, and not the original time of business flow generation. In this scenario, the processing of TSN gating and the learning of business flow features are interdependent, therefore business flow features and orchestration data cannot be effective simultaneously. This scenario is divided into an orchestration effective period and an orchestration stop period. During the orchestration stop period, the original business model of the business flow is learned.

[0127] The orchestration start / stop methods can be configured in several ways, including: a) Manual shutdown, via a function switch; when the function switch is turned on again, it starts learning the original business model for one orchestration cycle duration. b) Automatic shutdown: the shutdown duration can be configured as a fixed time period or set as a multiple of the orchestration cycle. Among them, the "Orchestration Stop" shown in Figure 8 is the automatic shutdown mechanism.

[0128] As shown in Figure 8, the following steps S1 to S5 are executed during the orchestration and execution phase.

[0129] S1: The cross-layer orchestration module subscribes to the terminal information of the communication terminals that support the gated TSN gate. The terminal information is such as: {UE ID}.

[0130] S2: The cross-layer orchestration module subscribes to business flow features from the business flow feature learning module. Business flow features include: {packet size, packet period, packet arrival time}.

[0131] S3: The cross-layer orchestration module subscribes to radio scheduling data from the base station's network equipment. The radio scheduling data includes: {frame structure information, synchronization information of system time and air interface time, cell information, scheduler parameters}.

[0132] S4: The cross-layer orchestration module subscribes to the latency measurement results of the communication terminals participating in the orchestration from the latency measurement module, such as: {RTTdelay}.

[0133] S5: When the service orchestration cycle is over, the cross-layer orchestration module performs time-domain and frequency-domain orchestration on the service flows that need to be guaranteed based on the information obtained from S1 to S4, according to the scheduler parameters and frame structure, to obtain the target scheduling time slots for each service flow, and generates orchestration data based on each target scheduling time slot. The orchestration data is sent to the TSN gating to ensure that each communication terminal sends data packets according to the orchestration data; the orchestration data is also sent to the scheduler of the network equipment at the base station, and the scheduler prepares scheduling resources according to the orchestration data. For example, {reference time, orchestration cycle, opening time point of each data packet, data packet size} is sent to the TSN; {reference time, orchestration cycle, time offset of each data packet, data packet size} is sent to the base station; where the reference time can represent the system time corresponding to the first uplink time slot in a radio frame, or the system time corresponding to the first uplink time slot of the receiving service flow in a radio frame; the data packet opening time point is the time offset relative to the reference time, and the data packet time offset also represents the time point offset relative to the reference time.

[0134] At this point, once the orchestration data takes effect at the TSN gate, the timestamp of the business flow obtained by the business flow feature learning module is the orchestrated timestamp, which is no longer the timestamp of the original business model. In the scenario in Figure 8, during the orchestration runtime, the business model learning module can only learn the changes in the business flow, that is, it can only obtain some business features, such as the changes in {packet size, packet period}.

[0135] During the scheduling pause, perform the following step S6.

[0136] S6: Orchestration stoppage period, learning business model, orchestration data invalidation, specifically the following steps S6.1 to S6.3.

[0137] S6.1 disables gating to obtain the actual service model, such as {data packet arrival time}. Specifically, as shown in Figure 8, the cross-layer orchestration module notifies the TSN gating to disable gating to the relevant UE, and notifies the TSN gating of terminal information such as: {UE ID}.

[0138] S6.2: The cross-layer orchestration module notifies the business flow feature learning module to clear historical data and relearn. The clearing instruction is as follows: {UE ID, clearing instruction flag}.

[0139] S6.3: The business feature learning module notifies the cross-layer orchestration module of the business flow features obtained after learning. For example: {packet size, packet duration, packet arrival time}.

[0140] At this point, during the orchestration and runtime phase, you can jump to S5 for orchestration processing.

[0141] The following describes a service orchestration method of a specific embodiment of this application with reference to FIG8, and an example with reference to FIG8 is shown below.

[0142] S1: The cross-layer orchestration module subscribes to the terminal information of the communication terminals that support the gated TSN gate, such as obtaining the terminal information as {UE1,UE2,UE3,UE4,UE5,UE6,UE7}.

[0143] S2: The cross-layer orchestration module subscribes to service flow features from the service flow feature learning module to obtain service layer data information, as shown in Table 1 above. Each communication terminal has a service flow, and each service flow feature includes {packet size, period, arrival time}. In this method, the arrival time changes of each flow in Table 1 are valid before the orchestration results take effect. After the orchestration results take effect, the data packet arrival time obtained is no longer the actual service arrival time. Therefore, in each orchestration cycle, only the feature {packet size} can be reflected in real time, while {period, arrival time} can only be obtained during the orchestration stop period, i.e., after relearning the service model as shown in S6 of Figure 8. The priority of each service flow is obtained from a fixed configuration.

[0144] S3: The cross-layer orchestration module subscribes to radio scheduling data from the base station. The frame structure information in the radio scheduling data includes frame type and time slot duration. Taking a DS frame structure (i.e., one downlink 0 and one uplink 1) and a time slot duration of 0.5ms as an example, the arrival time of each service flow can be converted into a slot offset based on the synchronization information of system time and air interface time. Referring to Figure 4, the communication terminals in the deterministic scheduling cell are {UE1, UE2, UE3, UE4, UE5, UE6}. The scheduler parameters in the radio scheduling data indicate that only 2 service flows can be scheduled in each time slot, and only 10Kbytes of data can be scheduled in each time slot.

[0145] After matching with the wireless scheduling data of the base station, we get: If cross-layer orchestration is not used, as shown in Table 2 above, time slot 0 can only schedule two service flows, and the system can only schedule four service flows in the end, namely two service flows in time slot 0, one service flow in time slot 2, and one service flow in time slot 4.

[0146] S4: The cross-layer orchestration module subscribes to the latency measurement results of the communication terminals {UE1,UE2,UE3,UE4,UE5,UE6} participating in the orchestration from the latency measurement module. Based on the latency target value and the latency measurement results, the latency margin of each stream is obtained, as shown in Table 1.

[0147] When the orchestration cycle is complete, the orchestration data obtained can be shown in Table 3, completing the orchestration of service layer data into the radio frame structure. After orchestration, the system can finally schedule 6 users.

[0148] S5: The orchestration data is sent simultaneously to the TSN gating and the base station scheduler. The TSN gating needs to convert the air interface time back to the system time. Then, each UE opens the gating at the time specified in Table 3, and the scheduler performs scheduling at the specified time. For example, UE3 and UE4 open the gating at the system time corresponding to time slot 0, respectively, to allow the service flow to be sent out. The scheduler prepares to schedule 200 bytes and 2 service flows in time slot 0.

[0149] When the configured orchestration stop period arrives, such as 00:00 to 05:00 daily, the cross-layer orchestration module will stop periodic orchestration and enter a learning period for the original business model. It will then notify the TSN gating to stop gating the managed UEs, such as {UE1, UE2, UE3, UE4, UE5, UE6, UE7}. Details are as follows.

[0150] S6.1: Close the gating.

[0151] S6.2: Notify the business flow feature learning module to clear the historical data of {UE1,UE2,UE3,UE4,UE5,UE6,UE7} and re-collect the business flow features of {UE1,UE2,UE3,UE4,UE5,UE6,UE7}; UE9 has no gating, so its business model has always been the original business model and therefore does not need to be processed.

[0152] S6.3: The service flow feature learning module periodically feeds back the service flow features of {UE1, UE2, UE3, UE4, UE5, UE6, UE7} to the cross-layer orchestration module.

[0153] Once the orchestration pause period ends and the orchestration run resumes, the cross-layer orchestration module will re-orchestrate the business based on the received information.

[0154] The following describes a service orchestration method according to a specific embodiment of this application with reference to FIG7.

[0155] S1: The cross-layer orchestration module subscribes to the terminal information of the communication terminals that support the gated TSN gate, such as obtaining the terminal information as {UE1,UE2,UE3,UE4,UE5,UE6,UE7}.

[0156] S2: The cross-layer orchestration module subscribes to the UE's service flow characteristics from the service flow feature learning module, resulting in Table 1. Each UE has one service flow, and each service flow has the following characteristics: packet size, period, and arrival time. In this approach, the arrival time of each flow in Table 1 is the arrival time of the original service model, which is not affected by whether the orchestration result is effective. Therefore, when the service model changes, it can be reflected in real time in each orchestration cycle.

[0157] S3: The cross-layer orchestration module subscribes to radio scheduling data from the base station. Radio scheduling data includes frame type and slot duration. The frame type indicates whether the frame structure is a DS frame structure, and each slot duration is 0.5ms. The arrival time of each service flow is converted into a slot offset based on the synchronization information of system time and air interface time. Referring to Figure 4, the UEs in the deterministic scheduling cell are {UE1, UE2, UE3, UE4, UE5, UE6}. Scheduler parameters include the ability to schedule only 2 service flows per slot and a maximum data volume of 10Kbytes per slot.

[0158] After matching with the base station information, it is found that if cross-layer orchestration is not used, as shown in Table 4, slot 0 can only schedule two service flows, and ultimately the system can only schedule 2 service flows. Because the radio layer does not know the delay margin of the service layer, it will not schedule UE1, UE2, and UE6 with delay.

[0159] Table 4

[0160] S4: The cross-layer orchestration module subscribes to the latency measurement results of the communication terminals {UE1,UE2,UE3,UE4,UE5,UE6} participating in the orchestration from the latency measurement module. Based on the latency target value and the latency measurement results, the latency margin of each service flow is obtained, as shown in Table 4.

[0161] When the orchestration cycle is over, the orchestration data can be as shown in Table 5. As shown in Table 5, after the scheduling orchestration is adopted, 5 users are finally scheduled, and the latency of each user meets the standard, and the air interface resources are fully utilized.

[0162] Table 5

[0163] The orchestration data is sent simultaneously to the TSN gating and the base station scheduler. The TSN gating needs to convert the air interface time back to the system time. Then, each communication terminal opens the gating at the time specified in Table 5, and the scheduler performs scheduling at the specified time. For example, UE3 and UE4 open the gating at the system time corresponding to time slot 0 to allow the service flow to be sent out. The scheduler prepares to schedule 200 bytes and 2 users' scheduling resources in time slot 0.

[0164] It is understood that a network device provided according to an embodiment of this application includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method described above.

[0165] It is understood that, referring to FIG9, one embodiment of this application also provides a network device, including:

[0166] At least one processor 901;

[0167] At least one memory 902 is used to store at least one program that implements the above method when the at least one program is executed by at least one processor 901.

[0168] Memory 902, as a non-transitory network system, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory 902 may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory 902 includes remotely located memories 902 relative to processor 901, which can be connected to processor 901 via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0169] The memory 902 can be implemented as a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 902 can store the operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 902 and is called and executed by the processor 901.

[0170] The processor 901 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application.

[0171] In some embodiments, the network device further includes:

[0172] Input / output interfaces are used to implement information input and output;

[0173] The communication interface is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0174] The bus transmits information between various components of the device (such as processor 901, memory 902, input / output interface, and communication interface);

[0175] The processor 901, memory 902, input / output interface, and communication interface can communicate with each other within the device via a bus.

[0176] It is understood that, in one embodiment of this application, a computer-readable storage medium is also provided, storing computer-executable instructions for performing the above-described method.

[0177] It is understood that one embodiment of this application also provides a computer program product, including a computer program or computer instructions, which are stored in a computer-readable storage medium. The processor of the network device reads the computer program or computer instructions from the computer-readable storage medium and executes the computer program or computer instructions, causing the network device to perform the above-described method.

[0178] According to the above embodiments of this application, by using the service flow characteristics and latency margin of each service flow of each communication terminal and the wireless scheduling data of the network device, the target scheduling time slot for scheduling each service flow of the network device can be determined. Based on the target scheduling time slot, service orchestration is achieved by controlling the network device to schedule the corresponding service flow and controlling the gating operation time of the service flow. At this time, the transmission of the service flow at the service layer is associated with the reception and processing of the wireless layer of the network device, so that the network device can control the transmission timing of the service flow based on the wireless resources. Furthermore, the introduction of latency margin can further ensure that the latency of the service flow meets the requirements during the service orchestration process. Therefore, compared with related technologies, the embodiments of this application can perform service orchestration for each service flow of multiple communication terminals, thereby improving the wireless resource utilization of the network device while reducing the service latency of the communication terminal.

[0179] The system architecture and application scenarios described in this application are intended to more clearly illustrate the technical solutions of this application and do not constitute a limitation on the technical solutions provided in this application. Those skilled in the art will understand that as system architectures evolve and new application scenarios emerge, the technical solutions provided in this application are also applicable to similar technical problems.

[0180] It will be understood by those skilled in the art that all or some of the steps and systems in the methods disclosed above can be implemented as software, firmware, hardware, and suitable combinations thereof. Some or all of the physical components can be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media 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 includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer. Furthermore, as is known to those skilled in the art, communication media typically contain computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0181] The above description, with reference to the accompanying drawings, illustrates some embodiments of this application, but does not limit the scope of this application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and spirit of this application shall be within the scope of this application.

Claims

1. A method for service orchestration, comprising: obtaining service flow characteristics and time delay margins of service flows sent by a communication terminal, and radio scheduling data of network devices connected in communication with the communication terminal; determining target scheduling time slots of the network devices for scheduling the service flows according to the radio scheduling data, the service flow characteristics and the time delay margins; controlling the network devices to schedule the corresponding service flows in the target scheduling time slots, and orchestrating a gating operation time of the service flows corresponding to the target scheduling time slots.

2. The method of business orchestration of claim 1, wherein, The determining of the target scheduling time slots of the network devices for scheduling the service flows according to the radio scheduling data, the service flow characteristics and the time delay margins comprises: determining initial scheduling time slots of the service flows; determining a first service flow to be adjusted and a remaining part of service flows from the service flows according to a scheduler parameter in the radio scheduling data and the service flow characteristics; determining a target scheduling time slot of the first service flow based on a time delay margin of the first service flow; taking the initial scheduling time slots of the remaining part of service flows as the target scheduling time slots of the remaining part of service flows.

3. The method of business orchestration of claim 2, wherein, The determining of the first service flow to be adjusted and the remaining part of service flows from the service flows according to the scheduler parameter in the radio scheduling data and the service flow characteristics comprises: sorting the service flows according to service priorities and time delay margins of the service flows to obtain a traversal order of the service flows; traversing the service flows based on the traversal order, and determining the first service flow according to the scheduler parameter and service flow characteristics of service flows sharing the same initial scheduling time slot with the traversed service flow in the traversing process; taking each service flow except the first service flow as the remaining part of service flows.

4. The method of business orchestration of claim 1, wherein, The controlling of the network devices to schedule the corresponding service flows in the target scheduling time slots comprises: generating orchestration data of the network devices according to the target scheduling time slots; controlling the network devices to schedule the service flows corresponding to the target scheduling time slots in a preset orchestration period according to the orchestration data of the network devices.

5. The method of business orchestration of claim 1, wherein, The orchestrating of the gating operation time of the service flows corresponding to the target scheduling time slots comprises: generating orchestration data of the communication terminal according to the target scheduling time slots; forwarding the orchestration data to a preset gating module to determine the gating operation time of the service flows in a preset orchestration period.

6. The method of business orchestration of claim 1, wherein, The determining of the target scheduling time slots of the network devices for scheduling the service flows according to the radio scheduling data, the service flow characteristics and the time delay margins comprises: determining the target scheduling time slots of the network devices for scheduling the service flows according to the radio scheduling data, the service flow characteristics and the time delay margins in a case that the communication terminal supports service orchestration processing, there is guaranteed service transmission, and deterministic scheduling cells are supported.

7. The method of business orchestration of claim 1, wherein, In a case that a timestamp carried by the service flow represents a time at which the service flow is generated, the method comprises: receiving the service flows sent by the communication terminal; The service flow is feature extracted to obtain the service flow feature, and the service flow feature includes packet size, packet period, and packet arrival time.

8. The method of business orchestration of claim 1, wherein, In a case where the timestamp carried by the service flow contains a gating processing duration, the method further includes: stopping scheduling the gating operation time of the service flow corresponding to the target scheduling time slot for a preset duration; acquiring the service flow sent by the communication terminal within the preset duration; feature extracting the received service flow to obtain the service flow feature; in a case where the preset duration is reached, re-determining the target scheduling time slot of the network device scheduling each service flow according to the wireless scheduling data, the service flow feature, and the time delay margin, and controlling the network device to schedule the corresponding service flow in each target scheduling time slot and schedule the gating operation time of the service flow corresponding to the target scheduling time slot.

9. The method of business orchestration of claim 1, wherein, The service scheduling method is applied to a computing power device or the network device.

10. A network device, comprising: a time delay measurement module configured to acquire a time delay margin of a service flow sent by a communication terminal; a service flow feature learning module configured to acquire a service flow feature of the service flow; a cross-layer scheduling module configured to determine a target scheduling time slot of the network device scheduling each service flow according to wireless scheduling data, the service flow feature, and the time delay margin, control the network device to schedule the corresponding service flow in each target scheduling time slot, and schedule the gating operation time of the service flow corresponding to the target scheduling time slot.

11. A network device, comprising: at least one processor; at least one memory for storing at least one program; wherein when at least one of the programs is executed by at least one of the processors, the method for service scheduling according to any one of claims 1 to 9 is implemented.

12. A computer-readable storage medium storing computer-executable instructions, wherein, The computer executable instructions are used to execute the method for service scheduling according to any one of claims 1 to 9.

13. A computer program product comprising computer programs or computer instructions, wherein, The computer program or the computer instructions are stored in a computer readable storage medium, and the processor of the network 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 network device executes the method for service scheduling according to any one of claims 1 to 9.

Images