Data stream processing method and device, edge node and storage medium

Abstract

The invention discloses a data stream processing method and device, an edge node and a storage medium. The method comprises the steps that multiple to-be-processed data streams are obtained, feature analysis is conducted on the to-be-processed data streams, feature requirements are obtained, and one to-be-processed data stream corresponds to one feature requirement; determining the current state information of the edge node, the current state information being used for reflecting the current data processing capability of the edge node; dividing the plurality of to-be-processed data streams into a first data stream and a second data stream based on the feature demand, the current state information and a preset data processing rule; and processing the first data stream, and sending the second data stream to other edge nodes, so that other edge nodes process the second data stream. According to the method provided by the invention, the management of the data stream to be processed is optimized, the utilization efficiency of the edge nodes is also improved, and the optimal data processing effect can be realized when each edge node is in different current state information.

Description

Technical Field

[0001] The present invention relates to the field of edge node technology, and in particular to a data stream processing method, device, edge node and storage medium. Background Art

[0002] With the continuous development of edge computing technology, edge nodes are added to the near-field environment of most IoT devices, so that multiple edge nodes can be used to process data near the source of data generation to reduce the data processing pressure of the central IoT server.

[0003] However, due to the large number of real-time data types, different data sizes, and uncertain network status in the near-field environment, the processing effect of edge nodes on real-time data is limited. Summary of the invention

[0004] The present invention provides a data stream processing method, device, edge node and storage medium, which optimizes the management of the data stream to be processed, improves the utilization efficiency of the edge nodes, and ensures that each edge node can achieve the optimal data processing effect under different current state information.

[0005] According to one aspect of the present invention, a data stream processing method is provided, the method comprising:

[0006] Acquire multiple data streams to be processed, and perform feature analysis on the data streams to be processed to obtain feature requirements, wherein one data stream to be processed corresponds to one feature requirement;

[0007] Determine the current state information of the edge node, wherein the current state information is used to reflect the current data processing capability of the edge node;

[0008] Based on feature requirements, current state information and preset data processing rules, multiple data streams to be processed are divided into a first data stream and a second data stream;

[0009] The first data stream is processed, and the second data stream is sent to other edge nodes, so that the other edge nodes process the second data stream.

[0010] According to another aspect of the present invention, there is provided a data stream processing device, the device comprising:

[0011] An acquisition module is used to acquire multiple data streams to be processed, and perform feature analysis on the data streams to be processed to obtain feature requirements, wherein one data stream to be processed corresponds to one feature requirement;

[0012] A determination module, used to determine the current state information of the edge node, wherein the current state information is used to reflect the current data processing capability of the edge node;

[0013] A division module, used for dividing a plurality of to-be-processed data streams into a first data stream and a second data stream based on feature requirements, current state information and preset data processing rules;

[0014] The processing module is used to process the first data stream and send the second data stream to other edge nodes so that the other edge nodes process the second data stream.

[0015] According to another aspect of the present invention, an edge node is provided, the edge node comprising:

[0016] at least one processor; and

[0017] a memory communicatively connected to the at least one processor; wherein,

[0018] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the data stream processing method described in any embodiment of the present invention.

[0019] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the data stream processing method described in any embodiment of the present invention when executed.

[0020] The data stream processing method provided by the embodiment of the present invention, on the one hand, obtains multiple data streams to be processed, and performs feature analysis on the data streams to be processed to obtain feature requirements, thereby realizing the analysis of the processing requirements of each data stream to be processed, so as to determine the corresponding data processing method based on the feature requirements of each data stream to be processed. On the other hand, the current state information of the edge node is determined, wherein the current state information is used to reflect the current data processing capability of the edge node, and provides a basis for determining the data processing method for each data stream to be processed based on the current state information and the feature requirements of each data stream to be processed. On the other hand, based on the feature requirements, the current state information and the preset data processing rules, the multiple data streams to be processed are divided into a first data stream and a second data stream. In this way, the processing position allocation of multiple data streams to be processed is realized, and it is determined whether the data stream to be processed is processed by the current edge node or by other edge nodes, which ensures the rapid processing of the data stream to be processed and improves the response speed and processing efficiency of the data stream to be processed. Finally, the first data stream is processed, and the second data stream is sent to other edge nodes so that other edge nodes process the second data stream. All data to be processed can be processed as much as possible, thereby greatly improving the response speed and operating efficiency of the IoT system while ensuring the quality of data processing. At the same time, this data stream processing method not only optimizes the management of the data stream to be processed, but also improves the utilization efficiency of edge nodes, ensuring that each edge node can achieve the best data processing effect under different current state information.

[0021] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0023] Figure 1 A schematic diagram of a data stream processing method provided by an embodiment of the present invention;

[0024] Figure 2 A schematic diagram of a flow chart of another data stream processing method provided by an embodiment of the present invention;

[0025] Figure 3 A schematic diagram of the structure of a data stream processing device provided by an embodiment of the present invention;

[0026] Figure 4 A schematic diagram of the structure of an edge node provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0027] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0028] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0029] Figure 1 A flow chart of a data stream processing method provided by an embodiment of the present invention, this embodiment can be applied to how to determine the processing method of each data stream to be processed when the current edge node obtains multiple data streams to be processed, the method can be executed by a data stream processing device, the data stream processing device can be implemented in the form of hardware and / or software, and the data stream processing device can be configured in the edge node. In this implementation, the edge node can be a controller, a server or an electronic device.

[0030] Specifically, edge nodes are an important part of distributed computing and network architecture, located at the edge of the network, close to users or data sources, and used to provide fast data transmission, processing, computing and storage services. The role of edge nodes is to bring data and computing resources as close to data sources and end users as possible, thereby reducing latency, improving response speed, and reducing the load on central servers.

[0031] In this embodiment, multiple edge nodes are deployed in the near-field environment of the IoT device, such as factory workshops, warehouses, smart buildings and other key areas, so that the edge nodes can achieve maximum range coverage and data flow collection.

[0032] Continue to refer to Figure 1 ,like Figure 1 As shown, the method includes:

[0033] S101. Acquire multiple data streams to be processed, and perform feature analysis on the data streams to be processed to obtain feature requirements, wherein one data stream to be processed corresponds to one feature requirement.

[0034] Among them, the data stream is a data sequence of a group of ordered bytes with a starting point and an end point. It includes input streams and output streams. In this embodiment, after receiving real-time data from the near-field environment of the Internet of Things, the edge node will determine the data stream based on the real-time data. For example, taking a large chemical plant as an example, data from different sensors in the factory are received as a data stream for monitoring the industrial production of the factory; traffic operation status information of transport vehicles in the factory and traffic flow and accident reports around the factory are received as a data stream for monitoring factory traffic information; real-time video surveillance data and safety behavior data from the factory are received as a data stream for monitoring factory safety accidents. The data stream to be processed is the data stream waiting to be processed by the current edge node.

[0035] Specifically, multiple data streams to be processed are obtained, and feature analysis is performed on each data stream to be processed to determine processing requirements that need to be met for processing each data stream to be processed, and the processing requirements that meet each data stream to be processed are used as feature requirements of each data stream to be processed.

[0036] For example, for the data stream that monitors the industrial production of the factory, since it characterizes the real-time status of the industrial production of the factory, such as whether the current production temperature rises, whether the production process is abnormal, etc. Therefore, this type of data stream has high real-time requirements, high delay sensitivity and high priority. For the data stream that monitors the factory traffic information, although it needs to respond quickly to the current traffic conditions, it will not cause serious consequences compared to the data stream of the industrial production of the factory. Therefore, the real-time requirements, delay sensitivity and priority of this type of data stream are medium. For the data stream that monitors the safety accidents of the factory, due to its monitoring functional requirements, its delay sensitivity requirements are high, but the real-time requirements are slightly lower than the data stream based on the monitoring of the industrial production of the factory. Therefore, this type of data stream has a high real-time requirement, high delay sensitivity and high priority. Among them, the real-time requirements, delay sensitivity and priority of the data stream, as well as other information such as network bandwidth requirements, data size and data type, can be used as feature requirements of the data stream.

[0037] The priority of each data stream can be determined based on the data type, real-time degree, and delay sensitivity of the data stream. For example, different weights are assigned to the data type, real-time degree, and delay sensitivity of each data stream, and the sum of the weights corresponding to each data stream is obtained. The sum of the weights is used as the priority of each data stream. Alternatively, the priority of each data stream is determined in a scoring manner.

[0038] In this embodiment, when the current edge node receives multiple data streams to be processed, feature analysis is performed on each data stream to be processed to obtain feature requirements of each data stream to be processed, thereby analyzing the processing requirements of each data stream to be processed so as to determine the corresponding data processing method based on the feature requirements of each data stream to be processed.

[0039] S102: Determine current status information of the edge node, where the current status information is used to reflect the current data processing capability of the edge node.

[0040] The edge node is an edge node that currently receives multiple data streams to be processed, that is, a current edge node.

[0041] Specifically, since the data processing capability of the current edge node may not be able to meet the data processing requirements for all the multiple data streams to be processed, it is necessary to determine the current state information of the current edge node.

[0042] Exemplarily, the current status information may include the current processing tasks of the current edge node, tasks waiting to be processed, available network status, whether multiple groups of data streams can be processed simultaneously, and the remaining amount of storage resources.

[0043] In this embodiment, by determining the current state information of the edge node, a basis is provided for subsequently determining a data processing method for each data stream to be processed based on the current state information and the characteristic requirements of each data stream to be processed.

[0044] S103: Based on feature requirements, current state information and preset data processing rules, multiple to-be-processed data streams are divided into a first data stream and a second data stream.

[0045] Among them, the preset data processing rules may include priority rules, real-time rules, network demand rules, etc. The first data stream is a data stream to be processed that can be processed by the current edge node. The second data stream is a data stream to be processed that needs to be processed by other edge nodes. Exemplarily, the data stream to be processed that needs to be processed by other edge nodes is not necessarily because the current edge node cannot process it, but because it needs to be processed immediately, but the current edge node cannot process the data stream to be processed immediately. The data stream to be processed can only wait for the processing of the current edge node. Therefore, the data stream to be processed needs to be used as the second data stream and distributed to other edge nodes for immediate processing.

[0046] For example, the priority rule is used to indicate that high-priority data needs to be processed first. The real-time rule is used to indicate that data flows with high real-time or low latency requirements need to be processed quickly. The network requirement rule is used to indicate whether the available bandwidth of the current edge node meets the bandwidth requirements of the data flow.

[0047] Specifically, in one implementation, the initial processing order of each data stream can be determined based on the feature requirements, and then the data stream to be processed that can be processed by the current edge node can be determined based on the initial processing order, current state information, and preset data processing rules, and the data stream to be processed is used as the first data stream. In addition, the data stream to be processed that needs to be processed by other edge nodes is used as the second data stream.

[0048] In another implementation, based on the historical feature requirements of a large number of historical data streams, the historical state information of the current edge node in the historical process, and the preset data processing rules, the machine learning algorithm model is trained so that the machine learning algorithm model learns the relationship between the processing urgency and importance of data streams with different feature requirements from the historical feature requirements, the historical state information, and the preset data processing rules, and obtains a data stream processing model that can output the processing strategy corresponding to each data stream to be processed when the feature requirements, the current state information, and the preset data processing rules are input. The processing strategy corresponding to the data stream to be processed includes the edge node and the processing method for processing the data stream to be processed.

[0049] Exemplarily, by inputting feature requirements, current state information, and preset data processing rules, the data stream processing model can output whether the data stream to be processed corresponding to the feature requirements is the first data stream or the second data stream. If it is the first data stream, the data stream processing model will also determine the processing method of the current edge node for the first data stream. If it is the second data stream, the data stream processing model can determine other edge nodes that process the second data stream.

[0050] In this embodiment, based on feature requirements, current status information and preset data processing rules, multiple data streams to be processed are divided into a first data stream and a second data stream, and processing position allocation of multiple data streams to be processed is realized. It is determined whether the data stream to be processed is processed by the current edge node or by other edge nodes, which ensures the rapid processing of the data stream to be processed and improves the response speed and processing efficiency of the data stream to be processed.

[0051] S104: Process the first data stream, and send the second data stream to other edge nodes, so that the other edge nodes process the second data stream.

[0052] Specifically, after dividing a plurality of data streams to be processed into a first data stream and a second data stream, the current edge node can determine a processing method for each first data stream based on the characteristic requirements of each first data stream, the current state information of the current edge node, and the preset data processing rules, and process the first data stream. Also, based on the current state information of other edge nodes, the geographic location rules in the preset data processing rules, and the characteristic requirements of the second data stream, determine other edge nodes for processing the second data stream, and send the second data stream to the other edge nodes, so that the other edge nodes process the second data stream.

[0053] Exemplarily, following the above embodiment, in one implementation, the data stream processing model is not used, and the processing method for each first data stream is determined based on the initial processing order, current state information, and preset data processing rules. For example, the data stream for monitoring the industrialized production of the factory, the data stream for monitoring the traffic information of the factory, and the data stream for monitoring the safety accidents of the factory are sorted, and the initial processing order can be determined as: the data stream for monitoring the industrialized production of the factory, the data stream for monitoring the safety accidents of the factory, and the data stream for monitoring the traffic information of the factory. If the available bandwidth of the current edge node can simultaneously meet the processing of the data stream for monitoring the industrialized production of the factory and the data stream for monitoring the safety accidents of the factory, and the data processing capacity of the current edge node can simultaneously process two data streams with higher priority, at this time, the data stream for monitoring the industrialized production of the factory and the data stream for monitoring the safety accidents of the factory can be processed simultaneously. And, after processing these two data streams, the data stream for monitoring the traffic information of the factory is processed. If the available bandwidth of the current edge node cannot simultaneously meet the processing needs of the data stream monitoring the industrial production of the factory and the data stream monitoring the factory's safety accidents and / or the data processing capacity of the current edge node cannot simultaneously process two data streams with higher priorities, then because the total score / weight of the priority, real-time requirements and delay sensitivity corresponding to the data stream monitoring the factory's industrial production is greater than the total score / weight of the priority, real-time requirements and delay sensitivity corresponding to the data stream monitoring the factory's safety accidents, the data stream monitoring the factory's industrial production is processed first, and the data stream monitoring the factory's safety accidents is processed immediately after the processing is completed, and finally the data stream monitoring the factory's traffic information is processed.

[0054] In another implementation, a data stream processing model is used, and the feature requirements, current state information, and preset data processing rules are input. The data stream processing model can output whether the data stream to be processed corresponding to the feature requirements is the first data stream or the second data stream. If it is the first data stream, the data stream processing model will also determine the processing method of the current edge node for the first data stream. If it is the second data stream, the data stream processing model can determine other edge nodes that process the second data stream.

[0055] In this embodiment, after the data stream to be processed is divided into the first data stream and the second data stream, the first data stream is processed by the current edge node, and the second data stream is sent to other edge nodes to realize the rapid processing of the second data stream, so that all the data to be processed can be processed as much as possible, thereby greatly improving the response speed and operation efficiency of the Internet of Things system while ensuring the quality of data processing. At the same time, this data stream processing method not only optimizes the management of the data stream to be processed, but also improves the utilization efficiency of the edge nodes, ensuring that each edge node can achieve the optimal data processing effect under different current state information.

[0056] The data stream processing method provided by the embodiment of the present invention, on the one hand, obtains multiple data streams to be processed, and performs feature analysis on the data streams to be processed to obtain feature requirements, thereby realizing the analysis of the processing requirements of each data stream to be processed, so as to determine the corresponding data processing method based on the feature requirements of each data stream to be processed. On the other hand, the current state information of the edge node is determined, wherein the current state information is used to reflect the current data processing capability of the edge node, and provides a basis for determining the data processing method for each data stream to be processed based on the current state information and the feature requirements of each data stream to be processed. On the other hand, based on the feature requirements, the current state information and the preset data processing rules, the multiple data streams to be processed are divided into a first data stream and a second data stream. In this way, the processing position allocation of multiple data streams to be processed is realized, and it is determined whether the data stream to be processed is processed by the current edge node or by other edge nodes, which ensures the rapid processing of the data stream to be processed and improves the response speed and processing efficiency of the data stream to be processed. Finally, the first data stream is processed, and the second data stream is sent to other edge nodes so that other edge nodes process the second data stream. All data to be processed can be processed as much as possible, thereby greatly improving the response speed and operating efficiency of the IoT system while ensuring the quality of data processing. At the same time, this data stream processing method not only optimizes the management of the data stream to be processed, but also improves the utilization efficiency of edge nodes, ensuring that each edge node can achieve the best data processing effect under different current state information.

[0057] Figure 2 A flowchart of another data stream processing method provided by an embodiment of the present invention, based on the above embodiment and optional implementation methods, this embodiment describes in detail the steps of "performing feature analysis on the data stream to be processed to obtain feature requirements", "dividing multiple data streams to be processed into a first data stream and a second data stream based on feature requirements, current state information and preset data processing rules", "after dividing multiple data streams to be processed into a first data stream and a second data stream", "processing the first data stream" and "before sending the second data stream to other edge nodes". Figure 2 As shown, the method includes:

[0058] S201. Obtain multiple data streams to be processed.

[0059] Specifically, after receiving the data, the current edge node determines a plurality of data streams to be processed based on the data type of the real-time data and other information.

[0060] S202: Determine the data type, data volume, and delay sensitivity of the data stream to be processed.

[0061] Among them, characteristic requirements include bandwidth requirements, delay sensitivity and priority score.

[0062] Specifically, based on the data stream to be processed, the data type of the data stream to be processed can be determined. For example, the data type is video data, audio data or sensor data. Further, the data type can be determined as production monitoring data, traffic operation data, safety accident data and other types of data based on the source of the data stream to be processed. And, the amount of data can also be determined based on the data stream to be processed. For example, determine the data size of the data stream to be processed, the bandwidth required for the data stream to be processed, etc. And, determine the delay sensitivity based on the data type of the data stream to be processed. For example, if the data type of the data stream to be processed is production monitoring data of production activities whose danger level exceeds a preset threshold, the delay sensitivity of the data stream to be processed is high. If the data type of the data stream to be processed is normal traffic operation data, the delay sensitivity of the data stream to be processed is low. If the data type of the data stream to be processed is safety accident data with a higher confidentiality level or greater security requirements, the delay sensitivity of the data stream to be processed is high.

[0063] S203: Determine the priority score of the data stream to be processed according to the data type and delay sensitivity; and determine the bandwidth requirement of the data stream to be processed according to the data type and data volume.

[0064] Specifically, the risk level and potential impact of each data stream to be processed can be determined based on the data type, and the score of the data type can be determined based on the risk level and potential impact of the data stream to be processed. The priority score of the data stream to be processed can be determined based on the score of the data type and the delay sensitivity score.

[0065] Exemplarily, in production activities, the monitoring data of production activities is used to ensure whether the equipment and personnel on the production line are in normal working condition. If it is in an abnormal state, the specific location of the abnormal state can be directly determined based on monitoring equipment such as sensors, and an alarm can be promptly issued to the personnel at that location or relevant processing personnel to remind the staff on the production line that there may be danger and ensure the safety of the staff. Therefore, if the data type is production monitoring data, it can be determined that its real-time requirements, risk level and potential impact are large. Therefore, the data type score A can be determined when the data stream to be processed is production monitoring data based on real-time requirements, risk level and potential impact. At the same time, the delay sensitivity score B can also be determined when the data stream to be processed is production monitoring data. The priority score when the data stream to be processed is production monitoring data is determined based on A+B.

[0066] It is worth noting that the priority score can also be determined by using the score and weight of the data type and the score and weight of the delay sensitivity. For example, the score of the data type multiplied by the corresponding weight plus the score of the delay sensitivity multiplied by the corresponding weight can determine the priority score.

[0067] Specifically, it can be determined whether the data is video, audio or sensor data based on the data type. Also, the real-time requirements, risk level and potential impact of the data can be determined based on the data type. Furthermore, based on the data volume, the data size of the data stream to be processed can be determined. Therefore, based on the above characteristics, the bandwidth requirements corresponding to the data stream to be processed can be determined.

[0068] For example, data with high real-time requirements, high risk levels, and high potential impacts, as well as large data volumes, has high bandwidth requirements, while data with low real-time requirements, low risk levels, and low potential impacts, as well as small data volumes, has low bandwidth requirements.

[0069] It is worth noting that bandwidth refers to the amount of data that can be transmitted in a fixed time, that is, the ability to transmit data in a transmission pipeline. Therefore, the bandwidth requirement of each data stream to be processed can be directly determined based on the data type and data volume. Optionally, similar to the priority score, the bandwidth requirement can also determine the bandwidth requirement sequence of each data stream to be processed based on the data type score and the data volume score. After determining the processing position of each data stream to be processed, the bandwidth amount allocated to each data to be processed is determined based on the bandwidth requirement sequence and other characteristic requirements.

[0070] In this embodiment, the priority score of the data stream to be processed is determined according to the data type and delay sensitivity, so that the importance of multiple data streams to be processed can be determined based on the priority of the data stream to be processed, providing support for subsequently determining the processing order and processing position of each data stream to be processed.

[0071] S204: Determine current status information of the edge node.

[0072] Specifically, since the data processing capability of the current edge node may not be able to meet the data processing requirements for all the multiple data streams to be processed, it is necessary to determine the current state information of the current edge node.

[0073] S205 , sorting multiple data streams to be processed according to preset data processing rules and feature requirements to obtain an initial processing order.

[0074] The preset data processing rules may also include data size rules, source and destination rules, and quality of service rules.

[0075] Exemplarily, the data size rule identifies the packet size of the data stream to be processed, and divides the data into different categories according to its size, such as small packets (log records and small sensor readings), medium packets (medium resolution video clips and database query results) and large packets (high resolution video streams and large-scale analysis results). For data streams with large data volumes, such as high-definition video streams, more storage capacity can be allocated to them. For data streams with small data volumes, less storage capacity can be allocated to them. The source and purpose rules determine the importance of the data stream according to the source of the data and the purpose to be achieved by the data stream. For example, if the source is a monitoring data stream on a production line, and its purpose is to monitor the production status of production equipment, then the real-time requirements and delay sensitivity of the data stream are high. Therefore, a faster data processing flow can be determined for it. For example, the data stream from the environmental monitoring station has a low delay sensitivity and a high real-time requirement because its data is collected regularly and analyzed for long-term trends. Therefore, a data processing flow that can wait or is not processed immediately can be determined for it. Quality service rules are rules for data generated by more important social tasks. For example, in the financial services or medical fields, quality of service rules need to be used to optimize the processing of these data streams, such as using QoS (Quality of Service) security mechanisms.

[0076] Specifically, according to the characteristic requirements of each data stream to be processed and the corresponding preset data processing rules, multiple data streams to be processed can be sorted to obtain an initial processing order.

[0077] Exemplarily, since each feature requirement of the data stream to be processed has multiple requirements, the weight of each requirement can be determined based on the data processing rule, and the total weight score of the data stream to be processed can be determined based on the weight of each requirement. And the multiple data streams to be processed are sorted based on the total weight score value of each data stream to be processed.

[0078] In one implementation, the preset data processing rule that best matches the data stream to be processed is determined based on the data type and data source of the data stream to be processed. For example, if the data stream to be processed is in the financial service field or the medical field, the data processing rule that best matches it is the quality service rule. At this time, the total weight score of the data stream to be processed can be determined directly based on the quality service rule and each characteristic requirement of the data stream to be processed. In another implementation, since the characteristic requirements include bandwidth requirements, delay sensitivity and priority scores, as well as other requirements. Therefore, the weight can be determined according to the data processing rules that match some or each requirement, and the weight of each characteristic requirement of the data stream to be processed is used to calculate the total weight score. For example, the weight corresponding to the bandwidth requirement is determined according to the bandwidth requirement and the network status rule, the weight corresponding to the delay sensitivity is determined according to the delay sensitivity and the network status rule, and the weight corresponding to the priority score is determined according to the priority score and the priority rule. The weight corresponding to the bandwidth requirement, the weight corresponding to the delay sensitivity and the weight corresponding to the priority score are used to determine the total weight score of the data stream to be processed.

[0079] Afterwards, after calculating the total weight score of each data stream to be processed, the total weight scores of multiple data streams to be processed can be arranged from large to small to obtain an initial processing order.

[0080] In this embodiment, the preset data processing rules and feature requirements are matched, and multiple data streams to be processed are sorted according to the matching results to obtain an initial processing order, thereby realizing a preliminary determination of the processing order of multiple data streams to be processed. The process of preliminary determination of the processing order of multiple data streams to be processed realizes the determination of the processing importance of each data stream to be processed, and determines its initial processing order according to the processing importance of each data stream to be processed, which provides a basis for determining the processing method of the data stream to be processed later.

[0081] S206: Divide the multiple data streams to be processed into a first data stream and a second data stream according to the initial processing order and the current state information.

[0082] Specifically, after determining the initial processing order of multiple data streams to be processed, according to the initial processing order, each data stream to be processed and the current state information are sequentially determined to determine whether each data stream to be processed can be processed by the current edge node. If it can be processed by the current edge node, it is divided into the first data stream; if it needs to be processed by other edge nodes, it is divided into the second data stream.

[0083] Exemplarily, there are now four data streams to be processed, A, B, C and D. Among them, A: high bandwidth demand, high priority, high delay sensitivity, mission-critical data. B: medium bandwidth demand, medium priority, medium delay sensitivity, regular business data. C: low bandwidth demand, high priority, high delay sensitivity, mission-critical data. D: high bandwidth demand, low priority, low delay sensitivity, non-mission-critical data. According to the above method, the four data streams to be processed, A, B, C and D, can be sorted, and the initial processing order is A, C, B, D. And, the current status information is: available bandwidth: 100Mbps (megabits per second), processing capacity: able to process multiple high-priority data streams at the same time. Storage resources: sufficient. At this time, it can be determined in order whether each data stream to be processed can be processed by the current edge node. For A, the current edge node needs to process it immediately. For C, after determining that the current edge node processes A immediately, the remaining available bandwidth can only process data with lower bandwidth requirements, and because the current edge node can process multiple high-priority data streams at the same time. Therefore, C can be processed by the current edge node at the same time as A. For B, since it is of medium priority and medium delay sensitivity, and is regular business data, B can be processed after the current edge node processes A and C. For D, since it is low priority, low delay sensitivity, and non-critical mission data, D can be processed after the current edge node processes B. Alternatively, due to its high bandwidth requirements and the long waiting time for processing after B is processed, it can be transferred to other idle edge nodes for processing. In summary, A, B, and C can be divided into the first data stream, and according to the current processing data and the number of processing tasks, it can be determined whether D is divided into the first data stream or the second data stream. For example, when the current edge node has less data to be processed or tasks to be processed, D can be determined as the first data stream. When the current edge node has more data to be processed or tasks to be processed, D can be determined as the second data stream. The specific situation depends on the current task volume and the number of data streams to be processed.

[0084] In this embodiment, according to the initial processing order and the current state information, multiple data streams to be processed are divided into a first data stream and a second data stream, thereby realizing the allocation of processing positions for multiple data streams to be processed, and determining whether the data stream to be processed is processed by the current edge node or by other edge nodes, thereby ensuring the rapid processing of the data stream to be processed and improving the response speed and processing efficiency of the data stream to be processed.

[0085] S207 . According to the position of each first data stream in the initial processing sequence, the characteristic requirements of the first data stream, and the current state information, sequentially predict the remaining state information of the edge node when processing each first data stream.

[0086] Specifically, for each first data stream to be processed, since it has different feature requirements, and the current state information of the current edge node will change with the data stream to be processed, it is necessary to predict the remaining state information of the edge node when processing each first data stream in turn according to the position of each first data stream in the initial processing order, the feature requirements of the first data stream and the current state information.

[0087] Exemplarily, following the above example, assuming that D is the second data stream, the initial processing order of the first data stream is A, C, and B. For A, its initial processing order is the first, and its feature requirements are higher than C and B. Therefore, if the current edge node is not processing any data stream at this time, it can be determined that when processing A, the remaining state information of the edge node is the current state information, that is, the available bandwidth is sufficient, it can process multiple high-priority data streams at the same time, and the storage resources are sufficient. For C, its initial processing order is the second, and its feature requirements are higher than B. And, since it has been determined to process A, the remaining state information of the current edge node should be: "low bandwidth, sufficient storage resources, and can also process a high priority and a data stream below the high priority" when predicting the processing of C. Similarly, when processing B, it can be assumed that the current edge node has processed A and C at the same time. Therefore, it can be predicted that the remaining state information of the current edge node is insufficient remaining bandwidth, sufficient storage resources, and no longer able to process data streams simultaneously.

[0088] S208. For a first data stream, determine whether the remaining state information of the edge node when processing the first data stream meets the characteristic requirements of the first data stream; if so, execute S210; if not, execute S209.

[0089] Specifically, for each first data stream, the remaining state information of the edge node when processing the first data stream can be determined in the above manner. If the remaining state information of the edge node cannot meet the characteristic requirements of the first data stream, it means that the current edge node cannot immediately process the first data stream. If the remaining state information of the edge node can meet the characteristic requirements of the first data stream, it means that the current edge node can immediately process the first data stream.

[0090] Exemplarily, following the above example, it is predicted that when processing C, the remaining state information of the edge node is low bandwidth, sufficient storage resources, and can process a high priority and data streams below the high priority. The characteristic requirements of C are low bandwidth requirements, high priority, high delay sensitivity, and mission-critical data. Therefore, it can be determined that the remaining state information of the current edge node meets the characteristic requirements of C. And when processing B, it is predicted that the remaining state information of the current edge node is insufficient remaining bandwidth, sufficient storage resources, and no simultaneous data streams. The characteristic requirements of B are medium bandwidth requirements, medium priority, medium delay sensitivity, and regular business data. Therefore, it can be determined that the remaining state information of the current edge node does not meet the characteristic requirements of B.

[0091] S209, putting the first data stream into the second processing queue; and after S211 is completed, executing S212.

[0092] The second processing queue is a queue of pending data flows waiting to be processed in the current edge node.

[0093] Specifically, if the remaining state information of the edge node cannot meet the characteristic requirements of the first data flow, it means that the current edge node cannot process the first data flow immediately. Therefore, the first data flow can be put into the second processing queue.

[0094] Exemplarily, following the above example, since it is predicted that when processing B, the remaining state information of the current edge node cannot meet the feature requirements of processing B, B can be put into the second processing queue. After waiting for the first processing queue to process the data stream to be processed, S212 is executed to process the second processing queue.

[0095] S210: Put the first data stream into a first processing queue.

[0096] The first processing queue is a queue of data flows to be processed immediately in the current edge node.

[0097] Specifically, if the remaining state information of the edge node can meet the characteristic requirements of the first data flow, it means that the current edge node can immediately process the first data flow. Therefore, the first data flow can be put into the first processing queue.

[0098] Exemplarily, following the above example, since it is predicted that when processing A and C, the remaining state information of the current edge node can meet the feature requirements of processing A and the feature requirements of processing C, A and C can be put into the first processing queue.

[0099] S211, processing the first processing queue.

[0100] Specifically, after the first processing queue is determined, the current edge node may immediately process the first data flow in the first processing queue.

[0101] S212: Process the second processing queue.

[0102] Specifically, after waiting for the current edge node to finish processing the first data flow in the first processing queue, the current edge node may continue to process the first data flow in the second processing queue.

[0103] In this embodiment, the first data stream is processed according to its position, characteristic requirements and current status information in the initial processing order to determine whether the remaining status information of the edge node meets the characteristic requirements of the first data stream when processing each first data stream, and the first data stream that meets the characteristic requirements of the first data stream is determined as the first processing queue, and the first data stream that does not meet the characteristic requirements of the first data stream is determined as the second processing queue, so as to predict the data streams that can be processed immediately and the data streams that can wait for processing based on the processing capacity of the current edge node and the predicted processing status of each data stream to be processed, thereby improving the efficient processing capability and stable operation capability of the edge node.

[0104] S213: Determine the real-time status information of the original edge node.

[0105] The original edge node is an edge node other than the current edge node. The real-time status information includes the current data processing capability of the original edge node, whether the original edge node currently has a data stream to be processed, and the amount of the data stream to be processed.

[0106] Specifically, after the multiple to-be-processed data streams are divided into the second data stream, the real-time status information of each original edge node needs to be determined.

[0107] S214: Determine a candidate edge node from the original edge nodes based on the feature requirements of the second data stream, the real-time status information, and a preset data processing rule.

[0108] The data processing rules also include geolocation rules. In this embodiment, the geolocation rules are used to send data to the edge node that is geographically closest and can provide the required service, thereby reducing transmission delay and improving response speed.

[0109] Specifically, it is necessary to first determine which original edge nodes among all edge nodes have data processing capabilities that can meet the characteristic requirements of the second data stream, and whose distance from the current edge node is less than or equal to a preset distance threshold, based on the characteristic requirements of the second data stream, the real-time status information of each original edge node, and the geographical location rules. Through the above method, multiple candidate edge nodes can be determined from the original edge nodes.

[0110] Exemplarily, when determining the candidate edge nodes, information such as network delay and transmission cost may also be used as selection conditions to determine the candidate edge nodes.

[0111] S215: Determine other edge nodes according to the location information of the candidate edge nodes.

[0112] Among them, the other edge nodes are the ones closest to the edge nodes among the candidate edge nodes.

[0113] Specifically, according to the preset geographical location rule, an edge node among the candidate edge nodes that is closest to the current edge node is directly determined as the other edge node.

[0114] Exemplarily, when determining other edge nodes, information such as network delay and transmission cost may also be used as selection conditions to determine other edge nodes.

[0115] S216: Send the second data stream to other edge nodes, so that the other edge nodes process the second data stream.

[0116] Specifically, after determining the other edge nodes corresponding to each second data stream, the second data stream is sent to the other edge nodes corresponding to the second data stream, so that the other edge nodes process the second data stream.

[0117] In this embodiment, based on the characteristic requirements, real-time status information and preset data processing rules of the second data stream, candidate edge nodes are determined from the original edge nodes, and then other edge nodes are determined based on the candidate edge nodes, so as to determine the processing position of each second data stream, so that all data to be processed can be processed as much as possible, thereby greatly improving the response speed and operation efficiency of the Internet of Things system while ensuring the quality of data processing. At the same time, this data stream processing method not only optimizes the management of the data stream to be processed, but also improves the utilization efficiency of the edge nodes, ensuring that each edge node can achieve the optimal data processing effect under different current status information.

[0118] The data stream processing method provided by the embodiment of the present invention, on the one hand, determines the priority score of the data stream to be processed according to the data type and delay sensitivity, can realize the importance of multiple data streams to be processed based on the priority of the data stream to be processed, and provides support for the subsequent determination of the processing order and processing position of each data stream to be processed. On the other hand, the preset data processing rules and feature requirements are matched, and the multiple data streams to be processed are sorted according to the matching results to obtain the initial processing order, and the processing order of the multiple data streams to be processed is preliminarily determined. The process of preliminarily determining the processing order of multiple data streams to be processed realizes the determination of the processing importance of each data stream to be processed, and determines its initial processing order according to the processing importance of each data stream to be processed, which provides a basis for determining the processing method of the data stream to be processed. On the other hand, according to the initial processing order and the current state information, the multiple data streams to be processed are divided into the first data stream and the second data stream, and the allocation of the processing positions of the multiple data streams to be processed is realized, and it is determined whether the data stream to be processed is processed by the current edge node or by other edge nodes, which ensures the rapid processing of the data stream to be processed and improves the response speed and processing efficiency of the data stream to be processed. And, the first data stream is processed according to the position, characteristic requirements and current state information in the initial processing order, and it is determined whether the remaining state information of the edge node meets the characteristic requirements of the first data stream when processing each first data stream, and the first data stream that meets the characteristic requirements of the first data stream is determined as the first processing queue, and the first data stream that does not meet the characteristic requirements of the first data stream is determined as the second processing queue, so as to predict the data stream that can be processed immediately and the data stream that can be waited for processing according to the processing capacity of the current edge node and the predicted processing state of each data stream to be processed, thereby improving the efficient processing capacity and stable operation capacity of the edge node. Finally, based on the characteristic requirements, real-time status information and preset data processing rules of the second data stream, the candidate edge node is determined from the original edge node, and then other edge nodes are determined based on the candidate edge node, so as to realize the determination of the processing position of each second data stream, so that all data to be processed can be processed as much as possible, thereby greatly improving the response speed and operation efficiency of the Internet of Things system while ensuring the quality of data processing. At the same time, this data stream processing method not only optimizes the management of the data stream to be processed, but also improves the utilization efficiency of the edge node, ensuring that each edge node can achieve the optimal data processing effect under different current state information.

[0119] For example, take a specific application scenario as an example: first, in the near-field environment of the IoT device, such as factory workshops, warehouses, smart buildings and other key areas, select the deployment location of each edge node. After all edge nodes are deployed, each edge node can collect, transmit and temporarily store data, and perform data processing and analysis tasks locally. Secondly, after receiving the data stream, the edge node can perform preliminary data cleaning and formatting on the data. For example, for video data, the edge node uses efficient video encoding and decoding technology to reduce the amount of transmitted data while maintaining high-definition image quality. Alternatively, the edge node analyzes the video stream in real time, quickly identifies abnormal activities such as intrusion detection through motion detection algorithms, and immediately triggers the alarm system when an abnormality is detected. And, the edge node dynamically adjusts the video quality according to the current network conditions, optimizes bandwidth usage by adjusting the resolution and frame rate, and ensures a smooth video experience. For audio data, the edge node uses noise reduction technology, such as spectral subtraction or deep learning algorithms, to eliminate background noise and improve voice clarity. And, voice commands are converted into operations or text information to achieve automated control and recording. Again, when the edge node processes the data stream before, during, and after processing the data stream according to the above data stream processing method, it will ensure that the data is transmitted from the source device to the target device, other edge nodes, or the final server based on end-to-end data encryption technology to ensure the security of the data transmission process and prevent the data from being tampered with or intercepted during the transmission process. For example, when the data leaves the source device, it will be encrypted by an encryption algorithm, such as using the AES-256 (256-bit Advanced Encryption Standard) standard to generate an encryption key that is shared only between the sender and the receiver. After arriving at the target node, the receiver uses the same key to decrypt the data and restore the original data. In addition, identity authentication and access control mechanisms can also be implemented. In addition, in the process of executing the data stream processing method, the resource usage of the edge node can also be monitored in real time based on the resource scheduling scheme of the edge node, and it is used as the current state information of the edge node and the real-time state information of the original edge node to provide data support for the data stream processing method, thereby realizing efficient resource utilization of each edge node, optimizing the load balancing of computing tasks, and improving energy efficiency, providing a stable, efficient, and energy-saving operating environment for the Internet of Things. Finally, the edge node also supports user customization. Based on user customization, it can be used to customize data processing logic and automated tasks for specific business needs through an intuitive graphical interface or through programming. In addition, users can continue to formulate preset data processing rules and other rules. For example, when the temperature detected by the sensor exceeds the preset threshold, the edge node automatically triggers the start of the cooling system; or when the operating status indicator of the equipment drops to a certain level, it automatically sends a maintenance reminder. The rules that users can customize include complex logical judgments, data filtering, conversion and aggregation operations, as well as interactions with other systems or services.Users can also set up automated tasks, such as regular data backup, system performance monitoring report generation, or device restart under specific conditions. The user-defined functions supported by edge nodes can not only achieve personalized business process management, but also ensure that these automated processes are closely aligned with user business goals and workflows, thereby improving operational efficiency and response speed, reducing human errors, and ultimately achieving smarter and more efficient management of IoT systems.

[0120] Exemplarily, the real-time requirement of each data stream to be processed can be determined according to the latency, throughput, error rate, etc. of the data stream to be processed.

[0121] Exemplarily, in the above-mentioned data stream processing method, if there are data streams to be processed with the same total score, that is, if the processing order of two or more data streams to be processed conflicts, the processing method can be determined based on the preset data processing rules. For example, if the data stream of the industrial control system has the highest priority score, it may be decided to send its data packet to the fastest processing node. If the data stream score of the security monitoring system is slightly lower, but the real-time demand is also high, it may be decided to send its data packet to a suboptimal but still efficient processing node. If the data stream of the traffic management system has the lowest score, it may be decided to put its data packet into a buffer and wait until other high-priority data streams are processed before processing.

[0122] Figure 3 A schematic diagram of the structure of a data stream processing device provided by an embodiment of the present invention. Figure 3 As shown, the device comprises:

[0123] The acquisition module 301 is used to acquire multiple data streams to be processed, and perform feature analysis on the data streams to be processed to obtain feature requirements, wherein one data stream to be processed corresponds to one feature requirement;

[0124] A determination module 302 is used to determine current state information of the edge node, wherein the current state information is used to reflect the current data processing capability of the edge node;

[0125] A division module 303, used to divide a plurality of to-be-processed data streams into a first data stream and a second data stream based on feature requirements, current state information and preset data processing rules;

[0126] The processing module 304 is configured to process the first data stream and send the second data stream to other edge nodes, so that the other edge nodes process the second data stream.

[0127] Optionally, the characteristic requirements include bandwidth requirements, delay sensitivity and priority scores; the characteristic requirements are obtained by performing characteristic analysis on the data stream to be processed, and the acquisition module 301 is specifically used for:

[0128] Determine the data type, data volume, and delay sensitivity of the data flow to be processed; determine the priority score of the data flow to be processed based on the data type and delay sensitivity; and determine the bandwidth requirement of the data flow to be processed based on the data type and data volume.

[0129] Optionally, the division module 303 is specifically used for:

[0130] According to preset data processing rules and feature requirements, multiple data streams to be processed are sorted to obtain an initial processing order; according to the initial processing order and current state information, the multiple data streams to be processed are divided into a first data stream and a second data stream.

[0131] Optionally, the edge node includes a first processing queue and a second processing queue; after dividing the multiple to-be-processed data streams into the first data stream and the second data stream, the dividing module 303 is further used to:

[0132] According to the position of each first data stream in the initial processing order, the characteristic requirements of the first data stream and the current state information, the remaining state information of the edge node when processing each first data stream is predicted in turn; for a first data stream, if the remaining state information of the edge node when processing the first data stream does not meet the characteristic requirements of the first data stream, the first data stream is placed in the second processing queue; if the remaining state information of the edge node when processing the first data stream meets the characteristic requirements of the first data stream, the first data stream is placed in the first processing queue.

[0133] Optionally, to process the first data stream, the processing module 304 is specifically configured to:

[0134] Process the first processing queue; after the first processing queue is processed, process the second processing queue.

[0135] Optionally, before sending the second data stream to other edge nodes, the processing module 304 is further configured to:

[0136] Determine the real-time status information of the original edge node; determine the candidate edge node from the original edge node based on the feature requirements of the second data stream, the real-time status information and the preset data processing rules; determine other edge nodes according to the location information of the candidate edge node.

[0137] The data stream processing device provided in the embodiment of the present invention can execute the data stream processing method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0138] Figure 4A schematic diagram of the structure of an edge node 10 that can be used to implement an embodiment of the present invention is shown. The edge node is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The edge node can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0139] like Figure 4 As shown, the edge node 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the edge node 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0140] Multiple components in the edge node 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the edge node 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0141] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 performs the various methods and processes described above, such as a data stream processing method.

[0142] In some embodiments, the data stream processing method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the edge node 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the data stream processing method described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to perform the data stream processing method in any other appropriate manner (e.g., by means of firmware).

[0143] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0144] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0145] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0146] To provide interaction with a user, the systems and techniques described herein may be implemented on an edge node having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the edge node. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0147] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0148] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the respective computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0149] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0150] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A data stream processing method, characterized in that: Applied to an edge node, the method comprises: Acquire multiple data streams to be processed, and perform feature analysis on the data streams to be processed to obtain feature requirements, wherein one data stream to be processed corresponds to one feature requirement; Determining current state information of the edge node, wherein the current state information is used to reflect the current data processing capability of the edge node; Based on the feature requirement, the current state information and a preset data processing rule, the plurality of data streams to be processed are divided into a first data stream and a second data stream; The first data stream is processed, and the second data stream is sent to other edge nodes, so that the other edge nodes process the second data stream.

2. The data stream processing method according to claim 1, characterized in that: The characteristic requirements include bandwidth requirements, delay sensitivity and priority scores; The performing feature analysis on the data stream to be processed to obtain feature requirements includes: Determining the data type, data volume and delay sensitivity of the data stream to be processed; Determine a priority score of the data stream to be processed according to the data type and the delay sensitivity; and determine a bandwidth requirement of the data stream to be processed according to the data type and the data volume.

3. The data stream processing method according to claim 2, characterized in that: The dividing the plurality of data streams to be processed into a first data stream and a second data stream based on the feature requirement, the current state information and a preset data processing rule comprises: Sorting the plurality of data streams to be processed according to preset data processing rules and the characteristic requirements to obtain an initial processing order; According to the initial processing order and the current state information, the plurality of data streams to be processed are divided into a first data stream and a second data stream.

4. The data stream processing method according to claim 3, characterized in that: The edge node includes a first processing queue and a second processing queue; After dividing the plurality of data streams to be processed into a first data stream and a second data stream, the method further includes: According to the position of each of the first data streams in the initial processing sequence, the characteristic requirements of the first data streams and the current state information, sequentially predict the remaining state information of the edge node when processing each of the first data streams; For one of the first data flows, if the remaining state information of the edge node does not meet the characteristic requirements of the first data flow when processing the first data flow, the first data flow is placed in the second processing queue; If the remaining state information of the edge node meets the characteristic requirements of the first data flow when processing the first data flow, the first data flow is placed in the first processing queue.

5. The data stream processing method according to claim 4, characterized in that: The processing of the first data stream comprises: processing the first processing queue; After the first processing queue is processed, the second processing queue is processed.

6. The data stream processing method according to claim 1, characterized in that: Before sending the second data stream to other edge nodes, the method further includes: Determine the real-time status information of the original edge node; Determine a candidate edge node from the original edge nodes based on the characteristic requirements of the second data flow, the real-time status information and a preset data processing rule; The other edge nodes are determined according to the location information of the candidate edge nodes.

7. The data stream processing method according to claim 6, characterized in that: The other edge node is one of the candidate edge nodes that is closest to the edge node.

8. A data stream processing device, characterized in that: The device comprises: An acquisition module, used for acquiring a plurality of data streams to be processed, and performing feature analysis on the data streams to be processed to obtain feature requirements, wherein one data stream to be processed corresponds to one feature requirement; A determination module, used to determine the current state information of the edge node, wherein the current state information is used to reflect the current data processing capability of the edge node; A division module, used for dividing the plurality of data streams to be processed into a first data stream and a second data stream based on the feature requirements, the current state information and a preset data processing rule; The processing module is configured to process the first data stream and send the second data stream to other edge nodes, so that the other edge nodes process the second data stream.

9. An edge node, characterized in that: include: one or more processors; a memory for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the data stream processing method according to any one of claims 1 to 7.

10. A readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the data stream processing method according to any one of claims 1 to 7 is implemented.

Images