Data flow scheduling method, apparatus, device, medium, and program product

Abstract

This application provides a data flow scheduling method, apparatus, device, medium, and program product, relating to the field of communication technology, for achieving resource scheduling to meet business needs and improve the accuracy of resource allocation. The method includes: receiving business event information from a digital service platform, the business event information including the business event type, the geographical area associated with the business event, and the expected time window of the business event; determining target data flow characteristics based on the geographical area, time window, and data flow characteristics of network data, used to indicate that the physical location of the source device is within the geographical area and that the data flow is related to the time window; generating network control instructions based on the business event type and target data flow characteristics, and sending the network control instructions to the network device corresponding to the business event. This application is applied to network data dynamic scheduling scenarios driven by digital twin business events.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to data stream scheduling methods, apparatus, devices, media, and program products. Background Technology

[0002] As smart cities accelerate towards deeper intelligence, digital twin technology has become the core for achieving intelligent interaction between the physical and information worlds. Digital twins synchronize the state of physical entities in real time, simulate and generate decisions in virtual space, and ultimately feed optimized strategies back to the physical world, forming an intelligent closed loop. In this process, the data processing and management system is responsible for collecting, transmitting, and processing massive amounts of spatiotemporally correlated data.

[0003] Traditional network data scheduling methods are typically static or semi-static, with scheduling strategies largely based on preset rules or passive responses to the current network state. These methods primarily focus on the network characteristics of the data flow itself or the real-time load of devices, and their decision-making logic is relatively independent.

[0004] Therefore, when a digital twin platform generates or predicts a critical event, traditional scheduling methods cannot automatically and promptly allocate the necessary network resources to ensure the quality of data transmission related to that event. Consequently, current data flow scheduling methods suffer from problems such as inaccurate resource allocation and inability to meet business needs when facing dynamic business scenarios driven by digital twins. Summary of the Invention

[0005] This application provides a data flow scheduling method, apparatus, device, medium, and program product for achieving resource scheduling that meets business needs and improves the accuracy of resource allocation.

[0006] In a first aspect, this application provides a data flow scheduling method, comprising: receiving business event information from a digital service platform, the business event information including a business event type, a geographical area associated with the business event, and a time window in which the business event is expected to occur; determining target data flow characteristics based on the geographical area, the time window, and data flow characteristics of network data, the target data flow characteristics indicating that the physical location of the source device is within the geographical area and is related to the time window; generating a network control instruction based on the business event type and the target data flow characteristics, and sending the network control instruction to the network device corresponding to the business event, the network control instruction instructing the network device to perform resource scheduling on the data flow indicated by the target data flow characteristics within the time window.

[0007] The technical solution provided in this application offers at least the following benefits: By receiving business event information from a digital service platform, the trigger source of network scheduling actions is directly associated with upper-layer business events, enabling the network to possess business awareness capabilities. Then, based on the geographical area and time window in the event information, the characteristics of the target data stream are determined, thereby accurately identifying key data related to specific spatiotemporal events from massive data streams. Network control commands are generated based on the business event type and target data stream characteristics, enabling differentiated resource guarantee strategies for different business scenarios (such as traffic congestion and emergency response) and different data characteristics (such as video streams and sensor streams), achieving on-demand scheduling. This ensures that the scheduling strategy meets business needs. Finally, network control commands are generated and sent, instructing network devices to execute scheduling within the time window, ensuring that the strategy is executed accurately and promptly within a forward-looking time window, achieving dynamic and precise resource allocation. Thus, this solution enables resource scheduling to meet business needs, thereby improving the accuracy of resource allocation.

[0008] One possible implementation involves determining the target data stream characteristics based on the geographical area, time window, and data stream characteristics of the network data. This includes: performing deep stream inspection on the network data to identify the application type and source device identifier of each data stream; obtaining the physical spatial location of the source device corresponding to each data stream based on the source device identifier; determining candidate data whose physical spatial location is within the geographical area based on the physical spatial location of the source device corresponding to each data stream; and determining the target data stream characteristics based on the application type of the candidate data streams. The target data stream characteristics are used to indicate data streams whose source device physical location is within the geographical area, whose application type is related to the business event type, and whose time window overlaps with the data stream.

[0009] Another possible implementation involves generating network control instructions based on the service event type and target data flow characteristics, including: generating a resource scheduling policy based on the service event type and target data flow characteristics, wherein the resource scheduling policy is used to allocate network resources to the data flow indicated by the target data flow characteristics; and generating network control instructions based on the resource scheduling policy.

[0010] Another possible implementation involves generating a resource scheduling policy based on the business event type and target data flow characteristics. This includes: obtaining a set of network resource scheduling actions that match the business event type and target data flow characteristics from a policy template database to obtain the resource scheduling policy. The policy template database includes an event type-policy mapping table and a policy set table. The event type-policy mapping table records the policy set identifiers corresponding to different business event types and different data flow characteristics. The policy set table records at least one network resource scheduling action and parameter contained in each policy set identifier.

[0011] Another possible implementation, the above-mentioned generation of network control instructions based on resource scheduling strategy, includes: obtaining a device configuration command template corresponding to the network resource scheduling action and the manufacturer model of the network device according to the network resource scheduling action contained in the resource scheduling strategy; writing the identification information of the data stream indicated by the target data stream characteristics and the parameter values ​​in the resource scheduling strategy into the corresponding fields in the device configuration command template to generate network control instructions.

[0012] Another possible implementation, in which the network control command is sent to the network device corresponding to the service event, includes: before or at the start of the time window, sending the network control command to the Software Defined Network (SDN) controller through the first interface, and then having the SDN controller convert the network control command into a flow table and send it to the network device; or, before or at the start of the time window, sending the network control command to the edge gateway or base station through the second interface.

[0013] Another possible implementation method includes: after the time window ends, and / or when an event end notification is received from the digital service platform, and / or when the data stream corresponding to the service event is detected to have stopped transmitting, sending a policy revocation instruction to the network device, which is used to instruct the release of network resources allocated for the data stream corresponding to the service event.

[0014] Secondly, this application provides a data flow scheduling device, comprising: a receiving module, a processing module, and a sending module; the receiving module is configured to receive business event information from a digital service platform, the business event information including a business event type, a geographical area associated with the business event, and a time window in which the business event is expected to occur; the processing module is configured to determine target data flow characteristics based on the geographical area, the time window, and the data flow characteristics of the network data, the target data flow characteristics indicating that the physical location of the source device is within the geographical area and is related to the time window; the processing module is further configured to generate network control instructions based on the business event type and the target data flow characteristics; the sending module is configured to send the network control instructions to the network device corresponding to the business event, the network control instructions instructing the network device to perform resource scheduling on the data flow indicated by the target data flow characteristics within the time window.

[0015] One possible implementation is that the aforementioned processing module is specifically used for: performing deep stream detection on network data to identify the application type and source device identifier of each data stream in the network data; obtaining the physical spatial location of the source device corresponding to each data stream based on the source device identifier; determining candidate data whose physical spatial location is within a geographical area from the network data based on the physical spatial location of the source device corresponding to each data stream; and determining target data stream characteristics based on the application type of the data streams of the candidate data, wherein the target data stream characteristics are used to indicate data streams whose source device physical location is within a geographical area, whose application type is related to the business event type, and whose time window overlaps.

[0016] Another possible implementation, the above processing module is specifically used to: generate a resource scheduling strategy based on the business event type and the characteristics of the target data flow, the resource scheduling strategy being used to allocate network resources to the data flow indicated by the characteristics of the target data flow; and generate network control instructions based on the resource scheduling strategy.

[0017] Another possible implementation is that the aforementioned processing module is specifically used to: obtain a set of network resource scheduling actions that match the business event type and target data flow characteristics from the policy template database, thereby obtaining a resource scheduling policy; wherein, the policy template database includes an event type-policy mapping table and a policy set table; the event type-policy mapping table is used to record the policy set identifiers corresponding to different business event types and different data flow characteristics; the policy set table is used to record at least one network resource scheduling action and parameter contained in each policy set identifier.

[0018] Another possible implementation is that the above processing module is specifically used to: obtain the device configuration command template corresponding to the network resource scheduling action and the manufacturer model of the network device according to the network resource scheduling action contained in the resource scheduling strategy; write the identification information of the data stream indicated by the target data stream characteristics and the parameter value in the resource scheduling strategy into the corresponding field in the device configuration command template to generate network control instructions.

[0019] Another possible implementation is that the aforementioned sending module is specifically used to: send network control commands to the SDN controller through the first interface before or at the start of the time window, and then convert the network control commands into flow tables and send them to the network devices via the SDN controller; or, send network control commands to the edge gateway or base station through the second interface before or at the start of the time window.

[0020] In another possible implementation, the aforementioned sending module is further configured to: send a policy revocation instruction to the network device when the time window ends, and / or when an event end notification is received from the digital service platform, and / or when the data stream corresponding to the service event is detected to have stopped transmitting. This policy revocation instruction is used to instruct the release of network resources allocated for the data stream corresponding to the service event.

[0021] Thirdly, this application provides an electronic device comprising: a processor and a memory; the memory storing processor-executable instructions; when the processor is configured to execute the instructions, causing the electronic device to implement the method of the first aspect described above.

[0022] Fourthly, this application provides a computer-readable storage medium comprising: computer software instructions; which, when executed in an electronic device, cause the electronic device to implement the method described in the first aspect.

[0023] Fifthly, this application provides a computer program product comprising a computer program; when the computer program is run in an electronic device, the electronic device performs the method described in the first aspect.

[0024] The beneficial effects of the second to fifth aspects mentioned above are described in the corresponding description of the first aspect and will not be repeated here. Attached Figure Description

[0025] Figure 1 This is a schematic diagram illustrating the application environment of a data stream scheduling method provided in an embodiment of this application. Figure 2 A flowchart illustrating a data stream scheduling method provided in an embodiment of this application; Figure 3 A flowchart illustrating another data stream scheduling method provided in an embodiment of this application; Figure 4 A schematic diagram illustrating the overall implementation process of a data stream scheduling method provided in this application embodiment; Figure 5 This application provides a schematic diagram of the architecture of a data stream scheduling system. Figure 6 This is a schematic diagram of the composition of a data stream scheduling device provided in an embodiment of this application; Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0026] The data stream scheduling method, apparatus, equipment, medium, and program products provided in this application will now be described in detail with reference to the accompanying drawings.

[0027] In this article, the term "and / or" is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone.

[0028] The terms "first" and "second," etc., used in the specification and drawings of this application are used to distinguish different objects or to distinguish different treatments of the same object, rather than to describe a specific order of objects.

[0029] Furthermore, the terms "comprising" and "having," and any variations thereof, used in the description of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the steps or units listed, but may optionally include other steps or units not listed, or may optionally include other steps or units inherent to such process, method, product, or apparatus.

[0030] It should be noted that in the embodiments of this application, the words "exemplary" or "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design scheme described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of the words "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0031] To facilitate a clear description of the technical solutions of the embodiments of this application, the terms "first" and "second" are used in the embodiments of this application to distinguish the same or similar items with essentially the same function and effect. Those skilled in the art can understand that the terms "first" and "second" are not intended to limit the quantity or execution order.

[0032] In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0033] The data flow scheduling method, apparatus, device, medium, and program product provided in this application embodiment can be applied to smart city scenarios driven by digital twin technology to provide dynamic network resource protection for massive, multi-source, and spatiotemporally correlated key business data flows.

[0034] Specifically, in the operation and management of smart cities, digital twin systems continuously simulate, analyze, and predict the physical city (such as traffic networks, environmental areas, and public venues). When the system predicts or senses that a critical event (such as severe traffic congestion, large-scale public events, or sudden environmental incidents) will occur in a specific area (such as a core intersection) during a specific time period (such as the morning rush hour), it generates structured business event information. However, to respond to the real-time decision-making required for such events (such as traffic light optimization, pedestrian flow management, and emergency response), it is highly dependent on the stable, low-latency, and highly reliable transmission of real-time data streams (video streams, sensor data streams, and trajectory information) from various IoT devices (cameras, sensors, and vehicles) within the event area to the digital twin platform for processing.

[0035] The technical solution of this application receives predictive event commands issued by a digital twin platform, automatically and accurately identifies all data streams spatiotemporally related to the event, and reserves or allocates necessary resources (such as bandwidth and priority) in the network in advance for these key data streams, thereby ensuring that the transmission channels for key data remain unobstructed when the event actually occurs. When the event ends, the system can automatically reclaim resources, realizing dynamic and efficient utilization of network resources.

[0036] For example, the solution of this application is applied to, but is not limited to, the following specific processes: Intelligent transportation: When traffic congestion warnings or accident responses are issued, priority is given to ensuring the transmission of video streams from cameras and data from geomagnetic sensors at relevant intersections, providing a lossless data foundation for real-time signal control and traffic guidance.

[0037] Large-scale event security: During sporting events and concerts, provide high-bandwidth guarantees in advance for public communication traffic and security video streams in core areas to improve network experience and security response speed.

[0038] Urban emergency management: In the event of emergencies such as fires and floods, quickly establish and ensure dedicated high-priority communication channels for data transmission from rescue teams, emergency drones, and on-site sensors.

[0039] Environmental monitoring and control: When heavy pollution weather is predicted, priority should be given to ensuring the real-time reporting of air quality sensor data in key monitoring areas to support precise governance decisions.

[0040] This solution enables the transformation of network resource scheduling from passive response and static configuration to proactive prediction and dynamic adaptation. It is a key technology for building an intelligent and elastic network that can truly understand and serve upper-layer services, and strongly supports the evolution of smart cities from informatization to deep intelligence.

[0041] Currently, as smart cities accelerate their progress towards deep intelligence, digital twin technology has become the core for realizing intelligent interaction between the physical and information worlds. Digital twins synchronize the state of physical entities in real time, simulate and generate decisions in virtual space, and ultimately feed optimized strategies back to the physical world, forming an intelligent closed loop. In this process, the data processing and management system is responsible for the collection, transmission, and processing of massive amounts of spatiotemporally correlated data. However, traditional network data scheduling methods are usually static or semi-static, exhibiting significant limitations.

[0042] When a critical event occurs on a digital twin platform, the real-time performance and reliability of all relevant data streams (such as video monitoring, vehicle trajectory, and environmental sensor data) become crucial. Traditional network data scheduling methods have limitations and cannot automatically and timely identify relevant spatiotemporal critical streams and allocate the necessary network resources to them, resulting in incomplete or delayed data for decision-making.

[0043] Thus, current data flow scheduling methods suffer from problems such as inability to meet business needs and inaccurate resource allocation when facing dynamic business scenarios driven by digital twins.

[0044] To address the aforementioned technical issues, this application provides a data flow scheduling method, apparatus, device, medium, and program product. Based on digital twin-driven dynamic data flow scheduling, it utilizes spatiotemporal event information published by the digital twin platform, identifies data flow characteristics through DFI technology, matches them with a device-location mapping library to obtain the geographical location of the data source, and combines event information and data flow location information for matching and decision-making to generate a dynamic scheduling strategy. This strategy is then distributed to the SDN controller or edge scheduler to adjust the resource allocation of network devices, achieving dynamic and automated data flow scheduling. In a digital twin scenario, this solution features business event-driven data flow scheduling across physical and virtual spaces; automated data flow mobilization and dynamic strategy triggering modes driven by business semantics; and spatial association between network data and physical entities to accurately identify data flow characteristics in geographical areas related to the current business event. Thus, by responding to business event information published by the digital twin platform that represents changes in the physical world's state, this solution automatically adjusts the network scheduling strategy for relevant data flows, thereby achieving dynamic adaptation of network resources to the needs of digital twin services.

[0045] The embodiments provided in this application will now be described in detail with reference to the accompanying drawings.

[0046] The data stream scheduling method provided in this application embodiment can be applied to, for example... Figure 1 The application environment shown. For example... Figure 1As shown, the application environment includes a data flow scheduling device 101 and a front-end device 102. The data flow scheduling device 101 and the front-end device 102 are interconnected.

[0047] In some embodiments, the data flow scheduling device 101 may be a server cluster consisting of multiple servers, a single server, a computer, or a processor or processing chip in a server or computer. This application does not limit the specific device form of the data flow scheduling device 101. Figure 1 The data flow scheduling device 101 is used as an example of a single server.

[0048] In some embodiments, the front-end device 102 can be a device with wireless transceiver capabilities, such as a mobile phone, tablet computer, wearable device, in-vehicle device, augmented reality (AR) / virtual reality (VR) device, laptop computer, ultra-mobile personal computer (UMPC), netbook, personal digital assistant (PDA), etc. This application embodiment does not limit the specific device form of the front-end device 102. Figure 1 The example shown is a mobile phone, with the front-end device 102 as the illustration.

[0049] In some embodiments, the data flow scheduling device 101 first receives service event information published by an upstream digital service platform, which includes the event type, geographical area range, and expected time window. Then, the data flow scheduling device 101 identifies the target data flow characteristics based on the service event information and the data flow characteristics of the network data. The data flow scheduling device 101 queries a pre-set policy template database to generate a corresponding resource scheduling policy based on the event type and the target data flow characteristics. The data flow scheduling device 101 converts the policy into a device configuration command template adapted to the specific network device (such as a switch or router, which belongs to the network infrastructure) manufacturer and model, and issues network control instructions through a northbound interface (such as to an SDN controller) or a southbound interface (such as a direct connection to an edge gateway) to instruct these network devices to allocate corresponding network resources (such as priority queues and guaranteed bandwidth) to the data flow indicated by the target data flow characteristics within a specified time window.

[0050] Front-end devices 102 (such as users' mobile phones, IoT sensors, and traffic cameras) act as data sources in this process. Located in the physical world, they continuously generate and send raw data streams (such as video streams, sensor data, and location information). When the physical location of front-end device 102 falls within the event geographic area predicted by the digital business platform, and its data stream type (such as camera video streams) is related to the event type (such as traffic congestion), the generated data stream will be identified by the data stream scheduling device 101. During the effective period of the scheduling policy issued by the data stream scheduling device 101, these data streams will receive priority during network transmission, enabling more stable and lower-latency transmission to the back-end processing system (such as a digital twin platform), providing high-quality data support for upper-level business decisions. Throughout this process, front-end device 102 does not need to be aware of complex scheduling logic; it only participates as a guaranteed data producer.

[0051] It should be noted that the system architecture described in the embodiments of this application is for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and does not constitute a limitation on the technical solutions provided in the embodiments of this application. As those skilled in the art will know, with the evolution of system architecture, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0052] See Figure 2 This is a flowchart illustrating a data stream scheduling method provided in an embodiment of this application. Figure 2 As shown, the data flow scheduling method provided in this application embodiment can be implemented by the above-mentioned data flow scheduling device, specifically including the following steps 201 to 203.

[0053] Step 201: The data flow scheduling device receives business event information from the digital business platform. The business event information includes the business event type, the geographical area associated with the business event, and the expected time window for the business event to occur.

[0054] In some embodiments, the aforementioned digital business platform may be a digital twin platform.

[0055] In some embodiments, the data flow scheduling device can accept business event information / spatiotemporal event information published by the digital twin platform, providing business-driven goals and context for the entire process. The digital twin platform performs simulation analysis and decision generation in virtual space by synchronizing the state of physical entities in real time, forming an intelligent closed-loop management of the physical world. When the platform predicts or senses that a critical event will occur in a specific area of ​​the physical world within a specific future time window based on real-time data and historical models, it generates and publishes this structured business event information.

[0056] In some embodiments, the digital twin platform performs simulation prediction by fusing multi-source data (such as sensor data and historical models) to determine the probability, type, and severity of events that will occur in a certain geographic area (such as Region_A) within a future time window, and then automatically generates structured information containing fields such as event type, geographic boundary (coordinate set), and time window.

[0057] In some embodiments, the data flow scheduling device is based on a digital twin platform and, according to real-time data and historical models, determines the timing of business events within the expected time window, i.e., the future time window / future specific time window [T]. start T end Within a specific geographic region (Region_A), a critical event will occur, generating structured event information. This event information may include: event type, geographic boundaries (such as geofence coordinates), and time window.

[0058] It is understandable that the aforementioned data flow scheduling device can receive business event information from the digital twin platform that represents changes in the physical world state, and is a key instruction connecting upper-layer business decisions and lower-layer network scheduling.

[0059] In some embodiments, the above-mentioned business event type includes the type of the event itself and the severity level of the event.

[0060] In some embodiments, the above-mentioned business event type is a category that identifies a specific business scenario, such as traffic congestion, emergency response, large-scale events, environmental monitoring, etc. This type may further include the type of the event itself and the severity level of the event (e.g., high, medium, low, severe, etc.).

[0061] In some embodiments, the geographic area associated with the above-mentioned business event defines the spatial boundary of the event's impact, typically represented in the form of geofence coordinates (polygonal geometric area).

[0062] In some embodiments, the time window in which the above-mentioned business event is expected to occur defines the period during which the event is expected to last, denoted as the time interval T_start and T_end.

[0063] In some embodiments, the data flow scheduling device can receive or subscribe to the aforementioned business event information from the digital twin platform asynchronously or synchronously through a predefined application programming interface (API) (such as a RESTful API) or a message queue.

[0064] Step 202: The data flow scheduling device determines the target data flow characteristics based on the geographical area, time window, and data flow characteristics of the network data. The target data flow characteristics are used to indicate the data flow where the physical location of the source device is within the geographical area and is related to the time window.

[0065] In some embodiments, the data flow scheduling device can determine the target data flow characteristics based on the geographical area and time window, as well as the data flow characteristics of continuously flowing network data, thereby indicating the data flow whose physical location of the source device is within the geographical area and is related to the time window.

[0066] In some embodiments, the aforementioned network data (or raw data stream) originates from various terminal devices deployed in the physical world, including but not limited to mobile devices (such as smartphones), Internet of Things (IoT) devices (such as smart meters and environmental sensors), and dedicated sensing devices (such as traffic cameras and geomagnetic coils). These devices, acting as data sources, continuously generate and transmit data packet sequences with spatiotemporal attributes.

[0067] In some embodiments, combined with Figure 2 ,like Figure 3 As shown, step 202 above can be specifically implemented as steps 202a to 202d.

[0068] Step 202a: The data flow scheduling device performs deep flow detection on the network data to identify the application type and source device identifier of each data flow in the network data.

[0069] In some embodiments, the data stream scheduling device can perform real-time analysis / deep stream inspection on the raw bitstream in network data, extracting the 5-tuple (source IP, destination IP, source port, destination port, transport protocol) and features such as stream duration, packet size distribution, packet time interval, and source device identifier, to build a classification model to identify application types (such as video streams and IoT sensor streams). For example, "video streams" are typically large and continuous data packets, and the strategy will focus on high bandwidth; "IoT sensor streams" are typically small and periodic data packets, and the strategy will focus on high reliability and low power consumption rather than high bandwidth.

[0070] It is understandable that the data flow scheduling device performs deep flow inspection (DFI) on the incoming network data. The process includes: Real-time traffic analysis: Real-time capture and analysis of raw bitstreams; Feature extraction: Extract the five-tuple information (source IP address, destination IP address, source port number, destination port number, transport layer protocol) for each data stream, as well as richer flow statistical features, such as flow duration, packet size distribution, packet arrival time interval, and source device identifiers that can uniquely identify the source device (such as device ID, MAC address). Classification Modeling: Based on extracted features, pre-trained machine learning classification models (such as pattern recognition models based on traffic statistics features) are used to intelligently identify the application type of data streams. For example, video streams (characterized by large, continuous data packet sequences with high bandwidth requirements) and IoT sensor streams (characterized by small, periodic data packets that are sensitive to reliability and low power consumption with low bandwidth requirements) can be identified.

[0071] Step 202b: The data stream scheduling device obtains the physical spatial location of the source device corresponding to each data stream based on the source device identifier.

[0072] In some embodiments, the data stream scheduling device can construct a device-location mapping library based on the actual location information of the source device and prior knowledge to display the precise geographical location information corresponding to the physical device. For the identified data stream, the device-location mapping library is queried through its source IP address or device ID to obtain the precise geographical location (Location_Device) of the physical device that generated the data stream, i.e., the physical spatial location of the source device corresponding to the data stream.

[0073] It is understandable that the data stream scheduling device can query a pre-set device-location mapping database based on the source device identifier (such as source IP address or device ID) extracted from the data stream. This mapping database is a database that stores the correspondence between physical device identifiers and their precise geographic coordinates (latitude and longitude). By querying, the device can obtain the precise geographic location (Location_Device) of the physical device that generated the data stream.

[0074] Step 202c: The data flow scheduling device determines candidate data whose physical spatial location is within a geographical area from the network data based on the physical spatial location of the source device corresponding to each data flow.

[0075] In some embodiments, the data stream scheduling device can use deep stream detection technology to identify the application type and source device identifier of the data stream; then query the device-location mapping library to obtain the physical spatial location corresponding to the source device identifier; and then use geospatial calculation to determine whether the physical spatial location falls within the geographical area.

[0076] In some embodiments, the data stream scheduling device may also employ a geolocation analysis method based on deep packet inspection (DPI) to determine whether the physical location falls within a geographic area.

[0077] In some embodiments, the data stream scheduling device can perform geospatial calculations using the physical location of the source device (Location_Device) of each data stream and the geographic region range (Region_A, typically defined by a set of polygon vertex coordinates) obtained from business event information.

[0078] A typical algorithm is the Point-in-Polygon (PIP) algorithm, such as the ray casting method. This algorithm emits a horizontal ray from the target point and calculates the number of intersections between the ray and the edge of the polygon. If the number of intersections is odd, the point is inside the polygon; if it is even, it is outside. This calculation determines whether the device's location falls within the event's geographic area.

[0079] In some embodiments, the data stream scheduling device can perform geospatial calculations using Location_Device and Region_A received from the digital twin platform to determine whether the location of the source device falls within the event area / geographic area. For example, the Location_Device of camera ID-001 is (116.3734, 39.3653), and the geofence coordinates of Region_A are typically a polygonal geometric region, [(116.3121, 39.9112), (116.5645, 39.9112), (116.3121, 39.1101), (116.5645, 39.1101)]. The point-in-polygon (PIP) algorithm (ray casting: a ray is cast horizontally from the target point to the right (or left), and the number of times the ray intersects the polygon boundary is calculated. If the number of intersections is odd, the point is inside the polygon; if it is even, it is outside) is used to determine whether the device location (Location_Device) is within the event region (Region_A).

[0080] It should be noted that, in addition to the DFI combined with a location database, geolocation determination can also employ Deep Packet Inspection (DPI) technology, which analyzes location-related information within the data packet payload. The geospatial calculation algorithm is not limited to the PIP algorithm; other computational geometry methods can be used, and this application's embodiments do not impose any limitations.

[0081] Step 202d: The data flow scheduling device determines the target data flow characteristics based on the application type of the candidate data flow. The target data flow characteristics are used to indicate the data flow whose physical location of the source device is within a geographical area, whose application type is related to the business event type, and whose time window overlaps with the data flow.

[0082] In some embodiments, the data stream scheduling device accurately determines all data stream characteristics that are spatiotemporally related to the predicted event by matching the aforementioned spatial location information with features such as the relevant device type and the application type parsed by DFI, and uses this information to subsequently determine the data streams that need to be scheduled. For example, within the same critical area, there may be traffic cameras, environmental sensors, Wi-Fi users, etc. If the predicted event type is traffic congestion, only the video streams related to the cameras are filtered out, while the data streams related to environmental sensors and ordinary internet users are not considered.

[0083] It is understandable that the data flow scheduling device can make judgments based on the type of business event and the time window: Business relevance judgment: Determine whether the application type of the candidate data is logically related to the type of the received business event. For example, if the business event is traffic congestion, only candidate data with the application type of video stream (from traffic cameras) will be retained, while irrelevant data streams such as environmental sensor streams or ordinary Internet traffic in the same area will be filtered out. Temporal relevance assessment: Determine whether the expected transmission time or active transmission period of the candidate data overlaps with the event's time window. This is based on predictions of data stream transmission patterns or real-time monitoring (e.g., predicting that the camera will continue to operate and generate video streams during the event window).

[0084] Thus, the data stream that simultaneously meets the two conditions of being related to both the application type and the business event type, and whose expected transmission time overlaps with the time window, is ultimately determined as the target data stream that needs to be scheduled for resources.

[0085] Thus, by performing deep stream inspection on network data to identify application types and source device identifiers, it surpasses traditional five-tuple identification, gaining a deeper understanding of the business semantics of data streams (such as distinguishing between video streams and sensor streams), and providing a key index for subsequent association with physical locations. By obtaining physical spatial locations based on source device identifiers and combining this with a location database for spatial filtering, it ensures that only data streams physically located within the event's impact area enter the candidate set. Based on the application type of the candidate data streams, target data stream characteristics are determined. These characteristics indicate data streams whose source device's physical location is within a geographical area, whose application type is related to the business event type, and whose time window overlaps with the data stream's location. This achieves dual filtering: firstly, precise matching based on business logic is achieved through application type relevance (e.g., excluding irrelevant environmental sensor streams in traffic congestion events), ensuring that the selected data streams are necessary for business operations; secondly, time overlap judgment ensures that these data streams are time-critical, meaning their active period coincides with the event's occurrence, requiring resource guarantees at that moment. Therefore, this approach improves the accuracy, reliability, and automation level of data stream identification through three dimensions: deep semantic understanding, precise spatial positioning, and spatiotemporal business relevance judgment.

[0086] Step 203: The data flow scheduling device generates network control commands based on the service event type and the characteristics of the target data flow, and sends the network control commands to the network device corresponding to the service event.

[0087] In some embodiments, the network control instructions described above are used to instruct network devices to perform resource scheduling on data streams indicated by target data stream characteristics within a time window.

[0088] It should be noted that the data stream indicated by the above target data stream characteristics can also be understood as the data stream corresponding to the business event (i.e., the data stream corresponding to the business event mentioned below). Both refer to the data stream whose physical location of the source device is within the above geographical area and is related to the above time window. Specifically, it can refer to the data stream whose physical location of the source device is within the above geographical area, whose application type is related to the above business event type, and whose time window overlaps with the above time window.

[0089] In some embodiments, step 203 can be specifically implemented as steps 203a and 203b.

[0090] Step 203a: The data flow scheduling device generates a resource scheduling strategy based on the business event type and the characteristics of the target data flow.

[0091] In some embodiments, the resource scheduling strategy described above is used to allocate network resources to data streams indicated by target data stream characteristics.

[0092] In some embodiments, the data flow scheduling device can generate corresponding resource scheduling policies based on the type of service event (including severity level) and the characteristics of the target data flow (such as the type of data flow). This policy essentially defines a set of network resource allocation actions, the purpose of which is to ensure network transmission quality for data flows corresponding to anticipated future service events.

[0093] In some embodiments, step 203a above can be specifically implemented as step 203a1.

[0094] Step 203a1: The data flow scheduling device obtains a set of network resource scheduling actions that match the characteristics of the business event type and the target data flow from the policy template database based on the business event type and the target data flow type, and obtains the resource scheduling policy.

[0095] In some embodiments, the policy template database includes an event type-policy mapping table and a policy set table.

[0096] The aforementioned event type and policy mapping table records the policy set identifiers corresponding to different service event types and different data flow characteristics. The aforementioned policy set table records at least one network resource scheduling action and parameters contained in each policy set identifier.

[0097] In some embodiments, the data flow scheduling device can use the event type, severity level of the service event information and the characteristics of the target data flow (such as the type / data type of the data flow) as joint inputs to query the policy template database; obtain a policy set consisting of multiple network control instructions from the policy template database; and convert the policy set into instructions adapted to specific network devices.

[0098] In some embodiments, the strategy template database described above is a predefined, structured database (SQL tables) built based on expert experience, and its main fields include the fields shown in Table 1 below: Table 1

[0099] The core fields of Table 1 include: event_type (event type): Used to store the classification of different business scenarios. Example values ​​include traffic_congestion, emergency_response, large_event, and environmental_monitoring. event_severity (event level): Used to identify the severity or urgency of an event. Example values ​​include high, medium, low, and critical. data_type (data type): Used to distinguish the application type of the data stream transmitted in the network. Example values ​​include video_stream (video stream), iot_sensor (IoT sensor stream), and v2x_message (vehicle-to-everything message). network_action: Defines the network resource configuration operations to be performed for a specific data flow. Example values ​​include set_dscp_ef (sets DSCP to EF for accelerated forwarding), allocate_min_bandwidth (allocates minimum guaranteed bandwidth), and assign_priority_queue (assigns a priority queue).

[0100] A basic policy template database (as shown in Table 1) can contain the following fields: event_type, event_severity, data_type, and network_action. By configuring these fields, empirical policies such as "video streams during traffic congestion events should guarantee low latency (corresponding to setting DSCP to EF and assigning a priority queue)" and "video streams during large events should guarantee high bandwidth" can be predefined. Then, the policy template database can be queried based on the event type and severity level. For example, for traffic congestion events, the matching policy is to guarantee low latency; for large event events, the matching policy is to guarantee high bandwidth.

[0101] In some embodiments, Tables 2, 3, and 4 below are SQL tables constructed based on expert experience. The SQL input consists of event information (event_type, event_severity) and data flow information (data_type), where Table 2 is the event type-policy mapping table and Table 3 is the policy set table.

[0102] Table 2

[0103] Table 2 defines the specific policy sets corresponding to different combinations of business events and data flow types. Specifically, the event type-policy mapping table records the unique policy set identifier (qos_policy_set_id, e.g., policy_set_1) corresponding to different combinations of (event type, severity level, data flow type). It establishes a mapping relationship from business semantics to the policy framework. Its core fields and examples are as follows: event_type (event type): Example values ​​are traffic_congestion, emergency_response, large_event, and environmental_monitoring. event_severity (event level): Example values: high, critical, medium, low; data_type (data type): Example values ​​are video_stream (video stream), all (all), and iot_sensor (IoT sensor stream); qos_policy_set_id (Quality of Service Policy Set Identifier): Example values ​​are policy_set_1 (policy set 1), policy_set_2 (policy set 2), policy_set_3 (policy set 3), and policy_set_4 (policy set 4). This field acts as a foreign key, linking to specific policy action details.

[0104] Table 3

[0105] Table 3 defines the specific network control action sequence and its parameters for each policy set identifier. In other words, the policy set table records the specific network resource scheduling action sequence (action) and its parameters (param) for each policy set identifier. The action defines the operation to be performed (e.g., set_dscp_ef, allocate_min_bandwidth), and the parameter quantifies the operation (e.g., bandwidth value of 10Mbps). Its core fields and examples are as follows: policy_set_id (policy set identifier): Example values: policy_set_1 (policy set 1), policy_set_2 (policy set 2), policy_set_3 (policy set 3); sequence (execution order): Used to define the execution order of multiple actions in the same policy set. Example value is the numerical sequence 1, 2; action: The specific network control command, example values ​​are set_dscp_ef (set DSCP to EF), assign_priority_queue (assign a priority queue), and allocate_min_bandwidth (allocate the minimum guaranteed bandwidth). param (parameter): The configuration parameter value required for the corresponding action. Example values ​​include null (empty, indicating no parameter is required or the default value is used), 10Mbps (10 megabits per second), 20Mbps (20 megabits per second), and 15Mbps (15 megabits per second).

[0106] For example, the query input includes the business event type (e.g., traffic_congestion), event severity level (e.g., high), and data stream type (e.g., video_stream) as joint query conditions. Policy retrieval: A database query is executed. For example, given the input (traffic_congestion, high, video_stream), the corresponding policy_set_1 is found through the mapping table, and then the specific action sequence contained in that policy set [{action:set_dscp_ef,param:null},{action:assign_priority_queue,param:10Mbps}] is retrieved from the policy set table. This action set is the generated resource scheduling policy.

[0107] It should be noted that the query method based on the predefined template database can be replaced by other intelligent methods, such as a strategy recommendation system based on a lightweight artificial intelligence (AI) model, which can dynamically generate scheduling suggestions based on historical data and real-time context. This application embodiment does not impose any restrictions.

[0108] In some embodiments, when the data stream indicated by the target data stream characteristics is a "long and few" continuous data stream (such as a video surveillance stream), the above resource scheduling strategy is a strategy for allocating resources to the continuous data stream after determining the type of the continuous data stream.

[0109] In some embodiments, when the data stream indicated by the target data stream characteristics is a "short and numerous" data stream (such as IoT sensor pulses), after determining the type of data stream, the above resource scheduling strategy can be a strategy for the data source (the data stream may no longer exist when the expected time arrives), for example, allocating a low-priority queue to all traffic from a certain port of the sensor gateway.

[0110] It should be noted that when the IP address changes after the data flow passes through a route, it can be updated through methods such as SDN controller awareness and DFI-based data flow feature analysis (e.g., specific encoding features). When the data flow does not pass through fixed devices, the SDN controller can achieve global vision capture (matching packet features) or distribute global policies based on flow features (distributed to all potential entry devices).

[0111] Thus, based on the business event type and target data flow characteristics, a matching set of network resource scheduling actions is retrieved from the policy template database. This transforms the policy generation process from relying on complex real-time calculations or manual configuration into efficient database query operations. The event type-policy mapping table establishes a direct mapping from business scenarios to the policy framework, while the policy set table pre-configures mature policy schemes optimized by expert experience or historical data, containing specific actions and parameters (such as guaranteed bandwidth values ​​and priorities). This approach ensures the speed and real-time nature of policy generation, meeting the low-latency response requirements of digital twin services, improving policy consistency and reliability, avoiding policy differences caused by different instances or operators, and ensuring that network behavior conforms to business expectations. Furthermore, when new business event types emerge or policies need adjustment, only the pre-configured template database needs to be updated, without modifying the core scheduling logic, reducing operational complexity and enabling the system to flexibly adapt to constantly changing business needs.

[0112] Step 203b: The data flow scheduling device generates network control commands based on the resource scheduling strategy.

[0113] In some embodiments, the network control instructions described above are used to instruct network devices to perform resource scheduling on data streams indicated by target data stream characteristics within a time window.

[0114] The data stream scheduling method provided in this application receives business event information from a digital service platform and directly associates the trigger source of network scheduling actions with upper-layer business events, enabling the network to have business awareness capabilities. Then, based on the geographical area and time window in the event information, the characteristics of the target data stream are determined, thereby accurately identifying key data related to specific spatiotemporal events from massive data streams. Network control commands are generated based on the business event type and target data stream characteristics, enabling differentiated resource guarantee strategies for different business scenarios (such as traffic congestion and emergency response) and different data characteristics (such as video streams and sensor streams), achieving on-demand scheduling. This ensures that the scheduling strategy meets business needs. Finally, network control commands are generated and sent, instructing network devices to execute scheduling within the time window, ensuring that the strategy is executed accurately and promptly within a forward-looking time window, achieving dynamic and precise resource allocation. Thus, this solution enables resource scheduling to meet business needs, thereby improving the accuracy of resource allocation.

[0115] In some embodiments, step 203b can be specifically implemented as steps 204a and 204b.

[0116] Step 204a: The data flow scheduling device obtains the device configuration command template corresponding to the network resource scheduling action and the manufacturer and model of the network device, based on the network resource scheduling action contained in the resource scheduling strategy.

[0117] In some embodiments, the data flow scheduling device can query another preset instruction dictionary table based on each abstract network resource scheduling action (such as set_dscp_ef) in the resource scheduling policy, combined with the specific vendor (device_vendor) and model (device_model) of the network device on the data flow path. This instruction dictionary table stores the mapping from (action, vendor, model) to specific device configuration command templates (cli_template), thereby obtaining the device configuration command template corresponding to the network resource scheduling action and the vendor and model of the network device by querying the instruction dictionary table.

[0118] Step 204b: The data flow scheduling device writes the identification information of the data flow indicated by the target data flow characteristics and the parameter values ​​in the resource scheduling strategy into the corresponding fields in the device configuration command template, and generates network control instructions.

[0119] In some embodiments, the data flow scheduling device can write (or fill in, replace) the identification information of the data flow indicated by the target data flow characteristics (such as mapping its five-tuple information to {map_name} and {class_name}) and the parameter values ​​in the resource scheduling policy (such as filling 10Mbps into {bandwidth_kbps}) into the corresponding placeholder fields in the device configuration command template. After the filling is completed, a specific network control instruction (such as a CLI command string) that can be directly recognized and executed by a specific network device is obtained.

[0120] In some embodiments, the data flow scheduling device can translate service requirements into precise instructions that the network infrastructure can understand and execute. For example, the final instruction might be: "Assign the highest priority queue to the data flow indicated by the target data flow characteristics (all packets with source IP being IP_001) and guarantee a minimum bandwidth of 10 Mbps."

[0121] The specific strategy generation and instruction conversion process: The strategy conversion process involves executing a logic similar to an SQL query. Table 4 shows the instruction dictionary table.

[0122] Table 4

[0123] Table 4 stores templates that translate abstract network actions into specific executable commands for network devices from different manufacturers and models. Its core fields and examples are as follows: action: Corresponds to the action field in Table 3. Example values ​​are set_dscp_ef (set DSCP to EF), allocate_min_bandwidth (allocate minimum guaranteed bandwidth), and assign_priority_queue (assign priority queue). device_vendor (device vendor): The manufacturer of the network equipment; example values ​​are C, H, and J. device_model (device model): The specific model of the network device. Example values ​​are IX (C IX operating system), NE (H NE series routers), and M series (JM series routers). `cli_template` (command line template): A configuration command template string specific to a particular manufacturer and model of device. Example: For the set_dscp_ef action on C IX devices, the template is: "policy-map {map_name} class{class_name} set dscp ef".

[0124] For the set_dscp_ef action of H NE device, the template is: "traffic behavior {behavior_name} remark ip-dscp 46".

[0125] For the allocate_min_bandwidth action of JM series devices, the template is: "firewall policer{policer_name} if-exceeding bandwidth-limit 10m".

[0126] For the assign_priority_queue action on C IX devices, the template is: "policy-map {map_name} class {class_name} priority {bandwidth_kbps}".

[0127] Note: {map_name}, {class_name}, {behavior_name}, {policer_name}, {bandwidth_kbps}, etc. in the template are all placeholders. They will be replaced (instantiated) according to the data flow information and policy parameter values ​​when the actual instruction is generated.

[0128] For example, combining Tables 2, 3, and 4 above, the input is (traffic_congestion, high, video_stream), and the internal query logic is "SELECT action, param FROM policy_table WHERE event_type='traffic_congestion' AND severity='high' AND data_type='video_stream';" (policy_table is a merged table of Tables 2 and 3); the output is {"action": set_dscp_ef, "param": null}, {"action": assign_priority_queue, "param": 10Mbps}. The abstract action field is then transformed into executable instructions for specific network devices using the instruction dictionary in Table 4. For example, “set_dscp_ef” is translated into the C IOS XE command: “policy-map MY_POLICY;class VIDEO;set dscp ef”, and “assign_priority_queue” is translated into the C IOS XE command: “policy-map MY_POLICY;class VIDEO;priority 10000”.

[0129] In this way, based on network resource scheduling actions, device configuration command templates corresponding to network equipment manufacturers and models are obtained, establishing a cross-manufacturer, cross-model instruction translation layer. This allows the same set of business policies to automatically adapt to devices from different manufacturers, shielding the heterogeneity of underlying devices. The identification information of the target data stream (such as the five-tuple) and policy parameter values ​​(such as bandwidth values) are written into the corresponding fields of the template, completing the instantiation from a general template to specific instructions. This generates unambiguous device-level configuration commands (such as CLI commands) that can be directly issued, achieving a high degree of automation in network configuration. This avoids tedious and error-prone manual command-line configuration while ensuring the accuracy of policy execution.

[0130] In some embodiments, the step 203 above, "sending network control instructions to the network device corresponding to the service event", can be specifically implemented as step 204c or step 204d.

[0131] Step 204c: Before or at the start of the time window, the data flow scheduling device sends network control commands to the SDN controller through the first interface, and the SDN controller converts the network control commands into flow tables and sends them to the network devices.

[0132] In some embodiments, the first interface described above may be a northbound API (such as a RESTful API).

[0133] Step 204d: Before or at the start of the time window, the data flow scheduling device sends network control commands to the edge gateway or base station through the second interface.

[0134] In some embodiments, the second interface described above may be a southbound interface (such as NetConf).

[0135] In some embodiments, before network congestion occurs, the data flow scheduling device pre-establishes high-speed channels for the data flows corresponding to service events to ensure their transmission quality. The generated policy instructions are sent to the SDN controller via a northbound API (such as a RESTful API) or to the edge gateway / base station via a southbound interface (such as NetConf). Upon receiving the instructions, the network device immediately adjusts its scheduling algorithm (such as a weighted fair queue or a priority queue) to provide resource guarantees for the data flows corresponding to the service events.

[0136] It is understandable that the data flow scheduling device may choose one of the following methods to issue instructions before or at the start of the time window to ensure that the resource channel has been established before the event occurs: Through a centralized control path: network control commands are sent to the Software Defined Network (SDN) controller via the first interface (northbound interface, such as RESTful API). The SDN controller is responsible for converting the received service layer commands into flow table entries that the underlying network devices can understand, and then uniformly distributing them to relevant network devices such as switches and routers through its southbound interface (such as OpenFlow). Through distributed direct connection path: network control commands are sent directly to specific network devices at the edge of the link where the data flow corresponding to the service event is located, such as edge gateways or base stations, through a second interface (southbound interface, such as NETCONF, CLI). After receiving the command, these devices immediately adjust their scheduling algorithm locally (such as modifying the weighted fair queue weight and setting priority queue) and implement resource allocation.

[0137] It should be noted that the above-mentioned northbound or southbound interface distribution methods can be replaced by other methods, such as generating only high-level policy descriptions instead of directly generating device command lines, and having other models or local agents fine-tune and execute them according to global policies and local conditions.

[0138] In this way, within the future time window, a dynamic network resource scheduling strategy is implemented to ensure the transmission quality of data streams. This means that a high-quality network transmission channel is pre-established before peak traffic periods or critical events occur, providing reliable assurance from the very first moment of an event and completely avoiding data delays or loss due to network congestion. For example, sending data to the SDN controller via the first interface is suitable for centralized, software-defined modern network architectures, achieving unified orchestration and flexible control from a global perspective. Sending data directly to the edge gateway or base station via the second interface is suitable for edge computing scenarios that are more sensitive to latency or require rapid local responses, achieving distributed and efficient execution. These two methods enable this technical solution to be flexibly deployed in various network environments from the core to the edge, ensuring the timeliness and proactivity of scheduling actions.

[0139] In some embodiments, the data flow scheduling method provided in this application further includes the following step 301.

[0140] Step 301: When the time window ends, and / or when the data flow scheduling device receives an event end notification from the digital service platform, and / or when it detects that the data flow corresponding to the service event has stopped transmitting, the data flow scheduling device sends a policy cancellation command to the network device.

[0141] In some embodiments, the above policy revocation instruction is used to instruct the release of network resources allocated to the data stream corresponding to the service event.

[0142] In some embodiments, the data flow scheduling device can monitor event status: whether the digital twin platform notifies that the event has ended or the situation has changed; the data flow scheduling device can monitor data flow status: whether the data flow corresponding to the business event has been terminated; when the event ends or the data flow terminates, the data flow scheduling device can automatically trigger a policy revocation process to release network resources back to the default pool, avoiding resource waste. This ensures the system's adaptability and efficiency.

[0143] In some embodiments, the data flow scheduling device generates and issues a corresponding revocation command (such as deleting previously issued flow table entries or restoring the default queue configuration). After the network device executes the command, previously reserved resources such as bandwidth and priority are released back to the public resource pool.

[0144] Thus, triggering a cancellation command when the time window ends, when an event end notification is received, and / or when the data stream corresponding to a business event is detected to have stopped transmitting can release network resources (such as priority queues and reserved bandwidth) previously guaranteed for specific events in a timely and accurate manner. This enables on-demand allocation and timely reclamation of network resources, avoids long-term ineffective occupation of resources, improves the overall utilization and turnover rate of resources, and allows limited network resources to serve more dynamically occurring business events. It also enhances the system's self-management capabilities. Even if the business platform notifies the system of anomalies or omissions, it can trigger resource reclamation by autonomously monitoring the data stream status, thus achieving refined and intelligent control of network resources.

[0145] For example, the embodiments of this application can be applied to the dynamic scheduling of network data in a smart city scenario based on digital twins. Take intelligent traffic congestion early warning as an example.

[0146] Digital twin platform predicts event information: The digital twin platform uses simulation to predict that severe traffic congestion will occur at intersections in the city center during the morning rush hour from 8:15 to 8:45.

[0147] (Event: Traffic congestion, Geographic boundary: Crossroad area, Time window: [08:15, 08:45], Severity level: High).

[0148] Data stream dynamic scheduling process: 1) Receive information related to events predicted by the digital twin platform. (Event: Traffic congestion, Geographic boundary: Crossroad area, Time window: [08:15, 08:45], Severity level: High).

[0149] 2) Based on the deep flow detection model and the device-location mapping library, spatiotemporal information matching is performed to analyze and identify all video streams and data streams from traffic cameras and geomagnetic sensors in the Crossroad area, and the target data stream characteristics are obtained.

[0150] 3) Based on the characteristics of the target data stream and the event information, generate strategies and specific instructions: mark the DSCP values ​​of all relevant video streams as EF (accelerated forwarding), and guarantee a minimum bandwidth of 10Mbps for all relevant sensor data streams.

[0151] 4) The instruction was sent to the SDN controller of the metropolitan area network at 8:10 in advance. When congestion began to form, all key traffic perception data had been stably transmitted to the traffic management center's digital twin platform through high-priority paths, providing a lossless data foundation for real-time traffic light optimization.

[0152] The data stream scheduling method of this application embodiment is described below with reference to a specific example, such as... Figure 4 The diagram shown illustrates the overall implementation process of a data flow scheduling method provided in this application embodiment. The specific process is as follows: Receive event information from the digital twin platform.

[0153] The data flow scheduling device receives business event information from the digital twin platform, representing changes in the state of the physical world. This event information provides the business-driven goals and context for the entire scheduling process.

[0154] Identify and monitor the characteristics of target data streams.

[0155] The data flow scheduling device proactively analyzes data flows in the network based on received service event information (especially the geographical area and future time window). It identifies data flow characteristics through technologies such as deep flow inspection (DFI) and performs spatiotemporal matching using a device-location mapping library. From massive network flows, it filters out all data flows that are spatiotemporally related to the predicted events, obtains the characteristics of target data flows that need to be specially scheduled, and continuously monitors them.

[0156] Generate dynamic network resource scheduling strategies.

[0157] The data flow scheduling device, based on the type and severity level of the service event and the characteristics of the target data flow, queries a pre-set policy template library to automatically generate a structured dynamic network resource scheduling policy for the data flow indicated by the aforementioned target data flow characteristics. This policy defines how to allocate network resources (such as priority and bandwidth) to the data flow indicated by the target data flow characteristics.

[0158] Issue and implement strategies in advance.

[0159] In this system, before network congestion or service events actually occur, the data flow scheduling device transforms the generated resource scheduling policy into specific network control commands and issues them in advance through a northbound interface (e.g., sent to the SDN controller) or a southbound interface (e.g., sent directly to edge devices). Upon receiving the commands, the network devices immediately adjust their scheduling algorithms, execute the policies, and establish a guaranteed channel for the data flow corresponding to the service event.

[0160] Monitor events and data stream status.

[0161] Throughout the entire process of the business event, the data flow scheduling device continuously monitors two key states in parallel: Event Status: Monitor whether the digital twin platform has notified that the event has ended or that the situation has changed; Data stream status: Monitor whether the data stream corresponding to the business event has been terminated.

[0162] Undo or restore the strategy.

[0163] Specifically, when an event ends or the data flow corresponding to a business event terminates, the data flow scheduling device automatically triggers a policy revocation process. It issues a command to the network device to revoke the previously implemented scheduling policy, releasing network resources back to the default resource pool, thus completing a full, closed-loop dynamic scheduling process. The entire process ends here, preparing for the next business event.

[0164] It should be noted that for a detailed description of the solution in this embodiment, please refer to the description in the above embodiments, which will not be repeated here.

[0165] Figure 5 This is a schematic diagram of the architecture of a data stream scheduling system provided in an embodiment of this application. The data stream scheduling system 800 may include: an event receiving and parsing module 801, a data stream identification module 802, a policy generation module 803, and an instruction issuing and execution module 804.

[0166] The event receiving and parsing module 801 is used to receive business event information published by a digital business platform (such as a digital twin platform), parse the information, and extract the business event type, the geographical area associated with the event, and the expected time window for the event. Applied to step 201 and its related schemes, it is responsible for acquiring and structuring the original business-driven instructions.

[0167] The aforementioned data stream identification module 802 is used to determine target data stream characteristics based on the parsed geographical area, time window, and network data stream characteristics. This indicates data streams located within the event's influence area and related to the event window. It is applied to steps 202, 202a to 202d, and related schemes. Specifically: Used for deep flow inspection (DFI) of network data to identify the application type of each data stream and extract the source device identifier, applied to step 202a and its related schemes.

[0168] This is used to query a pre-set device-location mapping library based on the source device identifier to obtain the physical spatial location of the source device corresponding to the data stream, and is applied to step 202b and its related schemes.

[0169] This is used to filter data streams located within the geographical area as candidate data based on physical spatial location, and is applied to step 202c and its related schemes.

[0170] The application type of the data stream based on the candidate data is used to determine the target data stream characteristics. The target data stream characteristics are used to indicate the data stream whose physical location of the source device is within a geographical area, whose application type is related to the business event type, and whose time window overlaps. These characteristics are applied to step 202d and its related schemes.

[0171] The aforementioned strategy generation module 803 is used to generate a dynamic scheduling strategy for allocating network resources to the data flow indicated by the target data flow characteristics, based on the service event type and the target data flow characteristics. This is applied to steps 203a and 203a1 and their related schemes. Specifically: This tool is used to retrieve a set of matching network resource scheduling actions from a pre-built policy template database (including event type-policy mapping table and policy set table) by using business event type (including severity level) and target data flow characteristics as joint query conditions, and to form a resource scheduling policy.

[0172] The database is used to maintain and update the policy template database, which stores preset policy actions (such as set_dscp_ef) and parameters (such as guaranteed bandwidth values) for different data types such as video streams and sensor streams under different event types such as traffic congestion and large events. Its contents correspond to the schemes described in Tables 1, 2 and 3 above.

[0173] The aforementioned instruction issuance and execution module 804 is used to transform abstract resource scheduling policies into specific instructions that can be executed by network devices, and to issue these instructions at the correct time to control network devices to perform scheduling, and to cancel scheduling at appropriate times. It is applied to step 203 and its related schemes. Specifically: Based on the network resource scheduling actions in the resource scheduling policy, it queries the instruction dictionary table (as shown in Table 4 above) to obtain the device configuration command template corresponding to the manufacturer and model of the target network device, and fills the identification information of the data flow indicated by the target data flow characteristics and the policy parameter value into the template to generate specific network control instructions.

[0174] This is used to send instructions to the SDN controller via a first interface (such as the Northbound API) before or at the start of the time window, or to send instructions directly to the edge gateway or base station via a second interface (such as the Southbound interface).

[0175] It should be noted that for a detailed explanation of the steps performed by each module and their beneficial effects, please refer to the description in the above embodiments, which will not be repeated here.

[0176] As can be seen, the above mainly describes the solutions provided by the embodiments of this application from a methodological perspective. To achieve the above functions, the embodiments of this application provide corresponding hardware structures and / or software modules for executing each function. Those skilled in the art should readily recognize that, in conjunction with the modules and algorithm steps of the various examples described in the embodiments disclosed herein, the embodiments of this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0177] This application embodiment can divide the data flow scheduling device into functional modules according to the above method example. For example, each function can be divided into its own functional module, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware or as a software functional module. Optionally, the module division in this application embodiment is illustrative and only represents one logical functional division; other division methods may be used in actual implementation.

[0178] In some embodiments, this application also provides a data stream scheduling apparatus. The data stream scheduling apparatus may include one or more functional modules for implementing the data stream scheduling method of the above method embodiments.

[0179] For example, Figure 6 This is a schematic diagram illustrating the composition of a data stream scheduling device provided in an embodiment of this application. Figure 6 As shown, the data stream scheduling device 900 includes: a receiving module 901, a processing module 902, and a sending module 903.

[0180] The receiving module 901 is used to receive business event information from the digital service platform. This business event information includes the business event type, the geographical area associated with the business event, and the expected time window for the business event to occur. The processing module 902 is used to determine the target data flow characteristics based on the geographical area, the time window, and the data flow characteristics of the network data. The target data flow characteristics indicate that the physical location of the source device is within the geographical area and that the data flow is related to the time window. The processing module 902 is also used to generate network control instructions based on the business event type and the target data flow characteristics. The sending module 903 is used to send the network control instructions to the network device corresponding to the business event. These network control instructions instruct the network device to perform resource scheduling on the data flow indicated by the target data flow characteristics within the time window.

[0181] In some embodiments, the processing module 902 is specifically used for: performing deep stream detection on network data to identify the application type and source device identifier of each data stream in the network data; obtaining the physical spatial location of the source device corresponding to each data stream based on the source device identifier; determining candidate data whose physical spatial location is within a geographical area from the network data based on the physical spatial location of the source device corresponding to each data stream; and determining target data stream characteristics based on the application type of the data stream of the candidate data, wherein the target data stream characteristics are used to indicate data streams whose source device physical location is within a geographical area, whose application type is related to the business event type, and whose time window overlaps.

[0182] In other embodiments, the processing module 902 is specifically used to: generate a resource scheduling strategy based on the service event type and the characteristics of the target data stream, wherein the resource scheduling strategy is used to allocate network resources to the data stream indicated by the characteristics of the target data stream; and generate network control instructions based on the resource scheduling strategy.

[0183] In some other embodiments, the processing module 902 is specifically used to: obtain a set of network resource scheduling actions that match the business event type and the target data flow characteristics from the policy template database, and obtain a resource scheduling policy; wherein, the policy template database includes an event type and policy mapping table and a policy set table; the event type and policy mapping table is used to record the policy set identifiers corresponding to different business event types and different data flow characteristics; the policy set table is used to record at least one network resource scheduling action and parameter contained in each policy set identifier.

[0184] In some other embodiments, the processing module 902 is specifically used to: obtain a device configuration command template corresponding to the network resource scheduling action and the manufacturer model of the network device according to the network resource scheduling action contained in the resource scheduling strategy; write the identification information of the data stream indicated by the target data stream characteristics and the parameter value in the resource scheduling strategy into the corresponding field in the device configuration command template to generate network control instructions.

[0185] In some other embodiments, the sending module 903 is specifically used to: send network control commands to the SDN controller through the first interface before or at the start of the time window, and then convert the network control commands into flow tables and send them to the network devices via the SDN controller; or, send network control commands to the edge gateway or base station through the second interface before or at the start of the time window.

[0186] In some other embodiments, the sending module 903 is further configured to: send a policy revocation instruction to the network device when the time window ends, and / or when an event end notification is received from the digital service platform, and / or when the data stream corresponding to the service event is detected to have stopped transmitting, the policy revocation instruction being used to instruct the release of network resources allocated for the data stream corresponding to the service event.

[0187] It should be noted that the data flow scheduling device can implement all the processes implemented in the above method embodiments and achieve the same beneficial effects. To avoid repetition, it will not be described again here.

[0188] In the case where the functions of the integrated modules described above are implemented in hardware, this application provides a possible structural schematic diagram of the electronic device involved in the above embodiments. For example... Figure 7 As shown, the electronic device 90 includes: a processor 92, a communication interface 93, and a bus 94. Optionally, the electronic device 90 may also include a memory 91.

[0189] Processor 92 may implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 92 may be a central processing unit, a general-purpose processor, a digital signal processor, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It may implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 92 may also be a combination that implements computational functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0190] Communication interface 93 is used to connect with other devices via a communication network. This communication network can be Ethernet, wireless access network, wireless local area network (WLAN), etc.

[0191] The memory 91 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but is not limited thereto.

[0192] As one possible implementation, the memory 91 can exist independently of the processor 92. The memory 91 can be connected to the processor 92 via a bus 94 and is used to store instructions or program code. When the processor 92 calls and executes the instructions or program code stored in the memory 91, it can implement the data flow scheduling method provided in the embodiments of this application.

[0193] In another possible implementation, memory 91 can also be integrated with processor 92.

[0194] Bus 94 can be an Extended Industry Standard Architecture (EISA) bus, etc. Bus 94 can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 7 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0195] Through the above description of the implementation methods, those skilled in the art can clearly understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the service calling device can be divided into different functional modules to complete all or part of the functions described above.

[0196] This application also provides a computer-readable storage medium. All or part of the processes in the above method embodiments can be executed by computer instructions instructing related hardware. The program can be stored in the aforementioned computer-readable storage medium, and when executed, it can include the processes of the above method embodiments. The computer-readable storage medium can be any of the foregoing embodiments or memory. The aforementioned computer-readable storage medium can also be an external storage device of the aforementioned service invocation device, such as a plug-in hard drive, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the aforementioned service invocation device. Further, the aforementioned computer-readable storage medium can include both internal storage units of the aforementioned service invocation device and external storage devices. The aforementioned computer-readable storage medium is used to store the aforementioned computer program and other programs and data required by the aforementioned service invocation device. The aforementioned computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

[0197] This application also provides a computer program product comprising a computer program that, when run on a computer, causes the computer to execute any of the data flow scheduling methods provided in the above embodiments.

[0198] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A data stream scheduling method, characterized in that, include: Receive business event information from the digital business platform, the business event information including the business event type, the geographical area associated with the business event, and the expected time window for the business event to occur; Based on the geographical area, the time window, and the data flow characteristics of the network data, target data flow characteristics are determined. These target data flow characteristics are used to indicate the data flow where the physical location of the source device is within the geographical area and is related to the time window. Based on the service event type and the target data stream characteristics, a network control command is generated and sent to the network device corresponding to the service event. The network control command is used to instruct the network device to perform resource scheduling on the data stream indicated by the target data stream characteristics within the time window.

2. The data stream scheduling method according to claim 1, characterized in that, The step of determining the target data stream characteristics based on the geographical area, the time window, and the data stream characteristics of the network data includes: Deep stream inspection is performed on the network data to identify the application type and source device identifier of each data stream in the network data; Based on the source device identifier, obtain the physical spatial location of the source device corresponding to each data stream; Based on the physical spatial location of the source device corresponding to each data stream, candidate data whose physical spatial location is within the geographical area are determined from the network data; Based on the application type of the data stream from the candidate data, the target data stream characteristics are determined. The target data stream characteristics are used to indicate data streams whose physical location of the source device is within the geographical area, whose application type is related to the business event type, and whose time window overlaps with the data stream.

3. The data stream scheduling method according to claim 1, characterized in that, The step of generating network control instructions based on the service event type and the target data stream characteristics includes: Based on the business event type and the target data flow characteristics, a resource scheduling strategy is generated, which is used to allocate network resources to the data flow indicated by the target data flow characteristics. The network control command is generated based on the resource scheduling strategy.

4. The data stream scheduling method according to claim 3, characterized in that, The step of generating a resource scheduling strategy based on the business event type and the target data stream characteristics includes: Based on the business event type and the target data flow characteristics, a set of network resource scheduling actions matching the business event type and the target data flow characteristics is obtained from the policy template database to obtain the resource scheduling policy; The policy template database includes an event type and policy mapping table and a policy set table. The event type and policy mapping table is used to record policy set identifiers corresponding to different business event types and different data flow characteristics. The policy set table is used to record at least one network resource scheduling action and parameter contained in each policy set identifier.

5. The data stream scheduling method according to claim 3, characterized in that, The generation of the network control command based on the resource scheduling policy includes: Based on the network resource scheduling actions contained in the resource scheduling strategy, obtain the device configuration command template corresponding to the network resource scheduling actions and the manufacturer and model of the network device; The identification information of the data stream indicated by the target data stream characteristics and the parameter values ​​in the resource scheduling strategy are written into the corresponding fields in the device configuration command template to generate the network control command.

6. The data stream scheduling method according to claim 1, characterized in that, Sending the network control command to the network device corresponding to the service event includes: Before or at the start of the time window, the network control command is sent to the software-defined network (SDN) controller via the first interface, and the SDN controller converts the network control command into a flow table and sends it to the network device. or, Before or at the start of the time window, the network control command is sent to the edge gateway or base station via the second interface.

7. The data stream scheduling method according to claim 1, characterized in that, The method further includes: When the time window ends, and / or when an event end notification is received from the digital service platform, and / or when it is detected that the data stream corresponding to the service event has stopped transmitting, a policy revocation instruction is sent to the network device. The policy revocation instruction is used to instruct the release of network resources allocated for the data stream corresponding to the service event.

8. A data stream scheduling device, characterized in that, include: The module consists of a receiving module, a processing module, and a sending module. The receiving module is used to receive business event information from the digital business platform. The business event information includes the business event type, the geographical area associated with the business event, and the expected time window for the business event to occur. The processing module is used to determine target data stream characteristics based on the geographical area, the time window, and the data stream characteristics of the network data. The target data stream characteristics are used to indicate that the physical location of the source device is within the geographical area and is related to the data stream of the time window. The processing module is further configured to generate network control instructions based on the business event type and the target data stream characteristics; The sending module is used to send the network control command to the network device corresponding to the service event. The network control command is used to instruct the network device to perform resource scheduling on the data stream indicated by the target data stream characteristics within the time window.

9. The data stream scheduling device according to claim 8, characterized in that, The processing module is specifically used for: Deep stream inspection is performed on the network data to identify the application type and source device identifier of each data stream in the network data; Based on the source device identifier, obtain the physical spatial location of the source device corresponding to each data stream; Based on the physical spatial location of the source device corresponding to each data stream, candidate data whose physical spatial location is within the geographical area are determined from the network data; Based on the application type of the data stream from the candidate data, the target data stream characteristics are determined. The target data stream characteristics are used to indicate data streams whose physical location of the source device is within the geographical area, whose application type is related to the business event type, and whose time window overlaps with the data stream.

10. An electronic device, characterized in that, The device includes a processor and a memory, the processor being coupled to the memory; the memory is used to store computer instructions, which are loaded and executed by the processor to enable the computer device to implement the data flow scheduling method as described in any one of claims 1 to 7.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes computer-executable instructions that, when executed on a computer, cause the computer to perform the data flow scheduling method as described in any one of claims 1 to 7.

12. A computer program product, characterized in that, The computer program product includes a computer program that, when run on an electronic device, causes the electronic device to perform the data stream scheduling method as described in any one of claims 1 to 7.

Images