Multi-node data traffic regulation and control system for BSP development management platform

Abstract

The invention provides a multi-node data traffic regulation and control system for a BSP development management platform, which comprises a BSP platform server, a node classification module, a traffic monitoring module and a dynamic regulation and control module, and is characterized in that the BSP platform server interacts with the node classification module, data traffic transmission is carried out through the node classification module, and the dynamic regulation and control module is connected with the node classification module; the node classification module divides nodes into four types of real-time nodes, high-throughput nodes, low-power-consumption nodes and cooperative computing nodes according to attributes and data traffic requirements of the nodes; after receiving transmission data, the real-time node dynamically allocates priority for a data packet, adds an accurate timestamp, encapsulates meta-information to the head of the data packet, selects an optimal path to transmit the data based on a time-triggered communication protocol and a multi-path redundancy mechanism, and feeds back through confirmation after first transmission; and after the high-throughput node receives the transmission data, a dynamic bandwidth allocation mechanism is adopted, and bandwidth resources are adjusted according to real-time flow requirements.

Description

Technical Field

[0001] The present invention belongs to the field of traffic control, and in particular relates to a multi-node data traffic control system for a BSP development management platform. Background Art

[0002] At present, in the BSP (Board Support Package) development and management platform, the multi-node data flow control system is one of the key components to ensure the efficient operation of the system. The BSP development and management platform is usually used for the development of embedded systems, involving the collaborative work of multiple hardware nodes, which may include processors, memories, peripheral interfaces, etc. With the increasing complexity of embedded systems, data flow management between multiple nodes becomes particularly important. Existing multi-node data flow control technologies mainly rely on static scheduling algorithms, fixed priority scheduling, and time-triggered communication mechanisms. These technologies can meet the real-time and reliability requirements of the system to a certain extent, but with the expansion of system scale and the increase of task complexity, the limitations of existing technologies gradually emerge.

[0003] Although static scheduling algorithms can determine the execution order and communication time of tasks through offline analysis during the design phase, they lack dynamic adaptability and cannot effectively cope with changes in the runtime environment. For example, when a node in the system fails or the load suddenly increases, the static scheduling algorithm cannot adjust the distribution of data traffic in real time, resulting in system performance degradation or even task failure. Although the fixed priority scheduling mechanism is simple and easy to implement, it is easy to cause priority inversion problems in multi-node systems. When a high-priority task depends on the resources of a low-priority task, the low-priority task may be blocked for a long time, which in turn affects the execution of the high-priority task, resulting in the inability to ensure the real-time performance of the system. Although the time-triggered communication mechanism can provide deterministic communication delay, it has extremely high requirements for clock synchronization. Once the clock deviates, it may cause communication failure or data loss. In a complex multi-node system, the accuracy and stability of clock synchronization are difficult to guarantee, especially in a distributed environment, where the cumulative effect of clock deviation will further aggravate the uncertainty of communication.

[0004] Existing technologies perform poorly when dealing with bursty data traffic. In the BSP development and management platform, certain tasks may generate large amounts of data in a sudden manner, such as centralized collection of sensor data or sudden event processing. Existing traffic control mechanisms usually adopt fixed bandwidth allocation strategies and are unable to dynamically adjust bandwidth resources according to real-time traffic, resulting in bandwidth bottlenecks under bursty data traffic, affecting the overall performance of the system. In addition, existing technologies are also insufficient in multi-node collaborative optimization. Data traffic control in a multi-node system needs to consider not only the performance of a single node, but also optimize the communication efficiency between nodes from a global perspective.

[0005] Existing control mechanisms are often limited to local optimization and lack a global perspective, resulting in the overall performance of the system not being optimal. For example, when multiple nodes compete for bandwidth resources at the same time, existing scheduling algorithms may not be able to allocate resources fairly, causing some nodes to be in a state of resource scarcity for a long time while other nodes have excess resources.

[0006] The existing technology also has defects in energy consumption management. With the increasing demand for low power consumption in embedded systems, multi-node data flow control systems need to reduce energy consumption as much as possible while ensuring performance. However, the existing control mechanisms usually only focus on performance optimization and ignore energy consumption management, resulting in high energy consumption while the system is running efficiently. In summary, the existing multi-node data flow control technology has many shortcomings in dynamic adaptability, priority management, clock synchronization, burst traffic processing, global optimization and energy consumption management. These shortcomings limit the further application of the BSP development and management platform in complex embedded systems. There is an urgent need for a more flexible, efficient and intelligent multi-node data flow control system to cope with the growing system complexity and performance requirements. There is an urgent need for a multi-node data flow control system for the BSP development and management platform. Summary of the invention

[0007] The present invention proposes a multi-node data flow control system for a BSP development and management platform, which solves the problems of real-time performance, bandwidth utilization, power consumption management and collaborative computing efficiency of data flow transmission of multiple types of nodes in the BSP development and management platform, and improves the overall performance and resource utilization of the system.

[0008] The technical solution of the present invention is implemented as follows: a multi-node data flow control system for a BSP development management platform, comprising a BSP platform server, a node classification module, a flow monitoring module and a dynamic control module, wherein the BSP platform server interacts with the node classification module, and data flow transmission is performed through the node classification module, wherein the node classification module divides the nodes into four types: real-time nodes, high-throughput nodes, low-power nodes and collaborative computing nodes according to the attributes of the nodes and the data flow requirements; After receiving the transmission data, the real-time node dynamically assigns priority to the data packet, adds an accurate timestamp and encapsulates the meta-information into the data packet header, selects the optimal path to transmit the data based on the time-triggered communication protocol and multi-path redundancy mechanism, and provides confirmation feedback after the first transmission; After receiving the transmission data, the high-throughput node uses a dynamic bandwidth allocation mechanism to adjust bandwidth resources according to real-time traffic demand, and establishes a redundant bandwidth trigger mechanism to activate redundant bandwidth to temporarily increase bandwidth allocation when the instantaneous traffic exceeds a set threshold; After receiving the transmission data, the low-power node uses a load-based power consumption management mechanism to monitor the power consumption of the BSP platform server, and controls the communication unit of the BSP platform server to enter a sleep mode when the load is low to reduce energy consumption; dynamically adjust the operating frequency and voltage when the load is high to balance performance and power consumption; The collaborative computing node interacts with the real-time node, the high-throughput node and the low-power node to coordinate the communication and computing tasks between the multiple nodes and reduce communication delays and resource conflicts; The traffic monitoring module monitors the data traffic status of each node in real time, including bandwidth utilization, delay and packet loss rate indicators, and sends the data to the dynamic control module at regular intervals; The dynamic control module adopts an adaptive bandwidth allocation algorithm to dynamically adjust bandwidth resources according to real-time traffic demand; the adaptive bandwidth allocation algorithm adaptively adjusts bandwidth allocation according to bandwidth utilization and delay indicators to ensure the stability of the data channel and reduce the packet loss rate.

[0009] Existing technologies usually adopt a unified traffic management strategy, which is unable to perform differentiated regulation according to the characteristics of different types of nodes. This solution divides nodes into four types: real-time nodes, high-throughput nodes, low-power nodes, and collaborative computing nodes through a node classification module, and designs a special regulation mechanism for each node, which significantly improves the accuracy and efficiency of traffic management. Traditional real-time nodes mostly use fixed priority scheduling and cannot dynamically adapt to changes in network status. This solution achieves high real-time and high reliability of data transmission by dynamically allocating priorities, adding precise timestamps, and based on time-triggered communication protocols and multi-path redundancy mechanisms, while further reducing the packet loss rate through confirmation feedback mechanisms. The bandwidth allocation of existing high-throughput nodes is mostly static or semi-dynamic, which is difficult to cope with instantaneous traffic fluctuations. This solution adopts a dynamic bandwidth allocation mechanism and a redundant bandwidth trigger mechanism to adjust bandwidth resources according to real-time traffic demand, and activate redundant bandwidth when the instantaneous traffic exceeds the set threshold, significantly improving bandwidth utilization and system stability.

[0010] Traditional low-power nodes mostly use fixed power consumption modes and cannot dynamically adjust power consumption according to the load. This solution uses a load-based power consumption management mechanism to put the communication unit into sleep mode when the load is low, and dynamically adjust the operating frequency and voltage when the load is high, thereby achieving a balance between performance and power consumption and significantly reducing system energy consumption. In the prior art, collaborative computing nodes usually lack efficient communication and computing task coordination mechanisms, which easily lead to resource conflicts and communication delays. This solution optimizes task allocation and resource scheduling through the interaction of collaborative computing nodes with real-time nodes, high-throughput nodes and low-power nodes, and reduces communication delays and resource conflicts. This solution uses the traffic monitoring module to monitor indicators such as bandwidth utilization, delay and packet loss rate in real time, and combines the adaptive bandwidth allocation algorithm of the dynamic control module to achieve real-time perception of traffic status and dynamic adjustment of bandwidth resources, significantly improving the response speed and stability of the system.

[0011] As a preferred implementation, the dynamic priority allocation of the real-time node is allocated based on a weighted calculation of the task urgency and the real-time traffic demand. By setting the basic priority weight, the weight coefficient is dynamically adjusted according to the real-time traffic load, and when the real-time node is transmitted, a hardware timestamp generator is used to add a nanosecond-precision timestamp to the data packet, and the timestamp is synchronized and calibrated with the global clock.

[0012] As a preferred implementation, the dynamic bandwidth allocation mechanism of the high-throughput node is processed by a redundant bandwidth trigger module. By presetting the bandwidth occupancy threshold, when the instantaneous traffic exceeds the threshold, the backup communication channel is automatically activated and redundant bandwidth is allocated. When transmitting through the high-throughput node, the large data stream is divided into data blocks of adaptive sizes through a dynamic block transmission algorithm, and the block size is adjusted according to the real-time bandwidth.

[0013] As a preferred implementation, the load-based power consumption management mechanism adopted in the low-power node is to customize the adjustment strategy by real-time collection of CPU load, cache occupancy rate and active duration of the communication unit, and generate frequency-voltage combination strategy in real time according to load changes. When the data cache reaches a preset capacity threshold, a hierarchical wake-up mechanism is triggered, and different power consumption management strategies are sent to the BSP platform server.

[0014] As a preferred embodiment, the collaborative computing node is configured with a distributed collaborative optimization engine, which is a communication path optimizer based on a genetic algorithm, calculates the optimal transmission path between nodes in real time, establishes a resource conflict prediction model, analyzes the probability of resource competition through historical communication data, and migrates excess computing tasks to idle nodes for execution.

[0015] As a preferred embodiment, the traffic monitoring module is integrated with an intelligent prediction unit that uses a time series prediction model to predict traffic trends and dynamically adjusts the monitoring sampling frequency, using millisecond sampling under high load and second sampling under low load.

[0016] As a preferred implementation, the adaptive bandwidth allocation algorithm of the dynamic control module adopts a bandwidth utilization-delay joint optimization function, and the calculation formula is:

[0017] in Q It is a comprehensive optimization index, which represents the comprehensive optimization target value of bandwidth allocation and is used to evaluate the pros and cons of the current bandwidth allocation strategy. U 带宽 is bandwidth utilization; D 延迟 is the data transmission delay; α and β are dynamic adjustment coefficients, which are configured differently according to the node type.

[0018] After adopting the above technical scheme, the beneficial effect of the present invention is: this multi-node data flow control system significantly improves the performance and resource utilization of the BSP development management platform through innovative designs such as node classification, real-time optimization, bandwidth management, power consumption control, collaborative computing and adaptive control.

[0019] The node classification module divides the nodes into four types and designs a special control mechanism for each node, realizing the precision and efficiency of traffic management. Real-time nodes ensure the high real-time and high reliability of data transmission through dynamic priority allocation, precise timestamp, time-triggered communication protocol and multi-path redundancy mechanism, and further reduce the packet loss rate through confirmation feedback mechanism. High-throughput nodes adopt dynamic bandwidth allocation mechanism and redundant bandwidth trigger mechanism, which significantly improves bandwidth utilization and system stability, and can effectively cope with instantaneous traffic fluctuations. Low-power nodes enter sleep mode at low load and dynamically adjust the working frequency and voltage at high load through load-based power management mechanism, achieving a balance between performance and power consumption and significantly reducing system energy consumption. Collaborative computing nodes reduce communication delays and resource conflicts by optimizing task allocation and resource scheduling, and improve the overall efficiency of the system. The combination of traffic monitoring module and dynamic control module realizes real-time perception of traffic status and dynamic adjustment of bandwidth resources, significantly improving the response speed and stability of the system.

[0020] The adaptive bandwidth allocation algorithm ensures the stability of the data channel and reduces the packet loss rate by adjusting bandwidth resources in real time. In summary, this system has significant advantages in real-time, bandwidth utilization, power consumption management, and collaborative computing efficiency, and is suitable for the complex and changeable BSP development and management platform environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. 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 paying creative labor.

[0022] Figure 1 It is a flow chart of the system of the present invention. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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 are within the scope of protection of the present invention.

[0024] Example: like Figure 1 As shown, in the specific implementation scenario of this application document, the working principle and workflow of the multi-node data traffic control system for the BSP development management platform are as follows: First, the BSP platform server interacts with the node classification module, which divides the nodes into four types: real-time nodes, high-throughput nodes, low-power nodes, and collaborative computing nodes according to the node attributes and data traffic requirements. After receiving the transmission data, the real-time node dynamically assigns priority to the data packet, adds an accurate timestamp and encapsulates the metadata to the data packet header, selects the optimal path to transmit data based on the time-triggered communication protocol and multi-path redundancy mechanism, and ensures the real-time and reliability of data transmission through confirmation feedback after the first transmission.

[0025] After receiving the transmission data, the high-throughput node adopts a dynamic bandwidth allocation mechanism to adjust bandwidth resources according to real-time traffic demand, and establishes a redundant bandwidth trigger mechanism. When the instantaneous traffic exceeds the set threshold, the redundant bandwidth is activated to temporarily increase the bandwidth allocation to ensure the stability and efficiency of high-throughput data transmission.

[0026] After receiving the transmission data, the low-power node uses a load-based power management mechanism to monitor the power consumption of the BSP platform server. When the load is low, the communication unit is controlled to enter the sleep mode to reduce energy consumption; when the load is high, the operating frequency and voltage are dynamically adjusted to balance performance and power consumption to ensure the energy saving effect of the low-power node. The collaborative computing node interacts with the real-time node, the high-throughput node and the low-power node to coordinate the communication and computing tasks between multiple nodes, reduce communication delays and resource conflicts, and ensure the coordination and efficiency of the overall operation of the system. The traffic monitoring module monitors the data traffic status of each node in real time, including bandwidth utilization, delay and packet loss rate indicators, and sends the data to the dynamic control module regularly to provide real-time data support for the dynamic control module. The dynamic control module adopts an adaptive bandwidth allocation algorithm to dynamically adjust bandwidth resources according to real-time traffic demand to ensure the stability of the data channel and reduce the packet loss rate.

[0027] In the entire workflow, the BSP platform server, as the core of the system, is responsible for interacting with the node classification module to ensure smooth data transmission. The node classification module divides nodes into four types according to the node attributes and data traffic requirements to ensure that the data transmission requirements of different types of nodes are met. Real-time nodes ensure the real-time and reliability of data transmission by dynamically allocating priorities and selecting the optimal path. High-throughput nodes ensure the stability and efficiency of high-throughput data transmission through dynamic bandwidth allocation mechanism and redundant bandwidth trigger mechanism. Low-power nodes ensure energy saving and performance balance through load-based power consumption management mechanism. Collaborative computing nodes ensure the coordination and efficiency of the overall operation of the system by coordinating the communication and computing tasks between multiple nodes. The traffic monitoring module provides real-time data support for the dynamic control module by monitoring the data traffic status of each node in real time. The dynamic control module dynamically adjusts bandwidth resources according to real-time traffic requirements through an adaptive bandwidth allocation algorithm to ensure the stability of the data channel and reduce the packet loss rate. Through the collaborative work of various components, the system realizes efficient control of multi-node data traffic, ensuring the real-time, stability, efficiency and energy-saving effect of data transmission.

[0028] The dynamic allocation priority allocation of the real-time node is allocated based on the weighted calculation of the task urgency and the real-time traffic demand. By setting the basic priority weight, the weight coefficient is dynamically adjusted according to the real-time traffic load, and the hardware timestamp generator is used to add nanosecond precision timestamps to the data packets when the real-time node is transmitted, and the timestamps are synchronized and calibrated with the global clock. The dynamic allocation priority allocation of the real-time node is allocated based on the weighted calculation of the task urgency and the real-time traffic demand. By setting the basic priority weight, the system can dynamically adjust the weight coefficient according to the real-time traffic load to ensure that high-priority tasks can be transmitted first. For example, in an industrial automation control system, when the real-time node receives emergency alarm data from the sensor, the system will dynamically adjust the priority weight according to the urgency of the task and the current network traffic load to ensure that the alarm data can be transmitted first. At the same time, the real-time node adds nanosecond precision timestamps to the data packets through the hardware timestamp generator, and synchronizes and calibrates with the global clock to ensure the time accuracy and synchronization of data transmission. This design is particularly important in scenarios where high-precision time synchronization is required, such as financial trading systems or power dispatching systems, which can ensure the transmission order of data packets and the accuracy of timestamps, and avoid system errors caused by time asynchrony.

[0029] The dynamic bandwidth allocation mechanism of the high-throughput node is processed by a redundant bandwidth trigger module. By presetting the bandwidth occupancy threshold, when the instantaneous flow exceeds the threshold, the backup communication channel is automatically activated and redundant bandwidth is allocated. When transmitting through the high-throughput node, the large data stream is divided into data blocks of adaptive size by a dynamic block transmission algorithm, and the block size is adjusted according to the real-time bandwidth. When the instantaneous flow exceeds the preset bandwidth occupancy threshold, the system automatically activates the backup communication channel and allocates redundant bandwidth to ensure the stability of high-throughput data transmission. For example, in a video streaming server, when the number of users surges and causes the instantaneous flow to exceed the threshold, the system automatically activates the backup communication channel and allocates redundant bandwidth to ensure the smooth transmission of the video stream. At the same time, the high-throughput node divides the large data stream into data blocks of adaptive size through a dynamic block transmission algorithm, and adjusts the block size according to the real-time bandwidth. This design is particularly important in scenarios where large-capacity data needs to be transmitted, such as cloud computing data centers or large-scale data transmission systems, which can dynamically adjust the data block size according to the real-time bandwidth situation to ensure the efficiency and stability of data transmission.

[0030] The load-based power management mechanism adopted in the low-power node is to customize the adjustment strategy by real-time collection of CPU load, cache occupancy rate and communication unit active duration, and generate frequency-voltage combination strategy in real time according to load changes, trigger the hierarchical wake-up mechanism when the data cache reaches the preset capacity threshold, and send different power management strategies to the BSP platform server. The load-based power management mechanism adopted by the low-power node customizes the adjustment strategy by real-time collection of CPU load, cache occupancy rate and communication unit active duration. For example, in an IoT device, when the device is in a low-load state, the system controls the communication unit to enter sleep mode to reduce energy consumption; when the device is in a high-load state, the system dynamically adjusts the operating frequency and voltage to balance performance and power consumption. At the same time, the system generates a frequency-voltage combination strategy in real time according to load changes, triggers a hierarchical wake-up mechanism when the data cache reaches the preset capacity threshold, and sends different power management strategies to the BSP platform server. This design is particularly important in IoT devices or mobile devices that need to run for a long time, which can effectively extend the service life of the device and reduce energy consumption.

[0031] The collaborative computing node is configured with a distributed collaborative optimization engine, which is based on a communication path optimizer of a genetic algorithm, calculates the optimal transmission path between nodes in real time, establishes a resource conflict prediction model, analyzes the probability of resource competition through historical communication data, and migrates excess computing tasks to idle nodes for execution. The distributed collaborative optimization engine configured by the collaborative computing node is based on a communication path optimizer of a genetic algorithm, and calculates the optimal transmission path between nodes in real time. For example, in a distributed computing system, when multiple nodes need to collaborate to complete a computing task, the system will calculate the optimal transmission path in real time through a genetic algorithm to reduce communication delays and resource conflicts. At the same time, the system will establish a resource conflict prediction model, analyze the probability of resource competition through historical communication data, and migrate excess computing tasks to idle nodes for execution. This design is particularly important in scenarios that require efficient collaborative computing, such as distributed databases or large-scale parallel computing systems, which can ensure efficient execution of computing tasks and reasonable allocation of resources.

[0032] The traffic monitoring module is integrated with an intelligent prediction unit that uses a time series prediction model to predict traffic trends and dynamically adjusts the monitoring sampling frequency, using millisecond sampling when the load is high and second sampling when the load is low. The intelligent prediction unit integrated in the traffic monitoring module uses a time series prediction model to predict traffic trends and dynamically adjusts the monitoring sampling frequency. For example, in a network traffic monitoring system, when the system detects a high load, it will use millisecond sampling to monitor the network traffic status in real time; when the system detects a low load, it will use second sampling to reduce system resource consumption. This design is particularly important in scenarios where real-time monitoring of network traffic is required, such as network security monitoring systems or network performance optimization systems, which can dynamically adjust monitoring strategies according to traffic trends to ensure real-time and efficient monitoring.

[0033] As a preferred implementation, the adaptive bandwidth allocation algorithm of the dynamic control module adopts a bandwidth utilization-delay joint optimization function, and the calculation formula is:

[0034] in Q It is a comprehensive optimization index, which represents the comprehensive optimization target value of bandwidth allocation and is used to evaluate the pros and cons of the current bandwidth allocation strategy. U 带宽 is bandwidth utilization; D 延迟 is the data transmission delay; α and β are dynamic adjustment coefficients, which are configured differently according to the node type.

[0035] Q: Comprehensive optimization index, which represents the comprehensive optimization target value of bandwidth allocation and is used to evaluate the pros and cons of the current bandwidth allocation strategy. By maximizing the Q value, the system can find the best balance between bandwidth utilization and latency. U 带宽 Bandwidth utilization indicates the actual usage of current bandwidth resources, usually expressed as a percentage (for example, 80% means that 80% of the current bandwidth is used).

[0036] When bandwidth utilization U 带宽 When it is higher, the formula tends to increase bandwidth allocation to avoid network congestion; when the delay D 延迟 When it is higher, the formula tends to optimize the transmission path or increase bandwidth to reduce latency; by introducing dynamic adjustment coefficients α and β, the system can adaptively adjust the optimization strategy according to real-time network status and application requirements, significantly improving network performance.

[0037] Assume that the bandwidth utilization of a real-time node is 90%, the delay is 10ms, and the current α=0.6; β=0.4, then Q=0.6·90+0.4·(1 / 10)=54+0.04=54.04; if the delay is reduced to 5ms, then Q =0.6·90+0.4·(1 / 5)=54+0.08=54.08; at this time, the Q value increases, indicating that the system performance is optimized; This formula realizes adaptive optimization of bandwidth allocation strategy by jointly optimizing bandwidth utilization and delay, combined with dynamic adjustment coefficients α and β, significantly improving the real-time performance, resource utilization and overall performance of the system.

[0038] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A multi-node data flow control system for a BSP development management platform, characterized in that: It includes a BSP platform server, a node classification module, a traffic monitoring module and a dynamic control module. The BSP platform server interacts with the node classification module and transmits data traffic through the node classification module. The node classification module divides the nodes into four types: real-time nodes, high-throughput nodes, low-power nodes and collaborative computing nodes according to the attributes of the nodes and the data traffic requirements. After receiving the transmission data, the real-time node dynamically assigns priority to the data packet, adds an accurate timestamp and encapsulates the meta-information into the data packet header, selects the optimal path to transmit the data based on the time-triggered communication protocol and multi-path redundancy mechanism, and provides confirmation feedback after the first transmission; After receiving the transmission data, the high-throughput node uses a dynamic bandwidth allocation mechanism to adjust bandwidth resources according to real-time traffic demand, and establishes a redundant bandwidth trigger mechanism to activate redundant bandwidth to temporarily increase bandwidth allocation when the instantaneous traffic exceeds a set threshold; After receiving the transmission data, the low-power node uses a load-based power consumption management mechanism to monitor the power consumption of the BSP platform server, and controls the communication unit of the BSP platform server to enter a sleep mode when the load is low to reduce energy consumption; dynamically adjust the operating frequency and voltage when the load is high to balance performance and power consumption; The collaborative computing node interacts with the real-time node, the high-throughput node and the low-power node to coordinate the communication and computing tasks between the multiple nodes and reduce communication delays and resource conflicts; The traffic monitoring module monitors the data traffic status of each node in real time, including bandwidth utilization, delay and packet loss rate indicators, and sends the data to the dynamic control module at regular intervals; The dynamic control module adopts an adaptive bandwidth allocation algorithm to dynamically adjust bandwidth resources according to real-time traffic demand; the adaptive bandwidth allocation algorithm adaptively adjusts bandwidth allocation according to bandwidth utilization and delay indicators to ensure the stability of the data channel and reduce the packet loss rate.

2. A multi-node data flow control system for a BSP development management platform as claimed in claim 1, characterized in that: The dynamic allocation priority of the real-time node is allocated based on the weighted calculation of the task urgency and the real-time traffic demand. By setting the basic priority weight, the weight coefficient is dynamically adjusted according to the real-time traffic load, and when the real-time node is transmitted, a nanosecond precision timestamp is added to the data packet through a hardware timestamp generator, and it is synchronized and calibrated with the global clock.

3. A multi-node data flow control system for a BSP development management platform as claimed in claim 1, characterized in that: The dynamic bandwidth allocation mechanism of the high-throughput node is processed by a redundant bandwidth trigger module. By presetting the bandwidth occupancy threshold, when the instantaneous traffic exceeds the threshold, the backup communication channel is automatically activated and redundant bandwidth is allocated. When transmitting through the high-throughput node, the large data stream is divided into data blocks of adaptive size through a dynamic block transmission algorithm, and the block size is adjusted according to the real-time bandwidth.

4. A multi-node data flow control system for a BSP development management platform as claimed in claim 1, characterized in that: The load-based power consumption management mechanism adopted in the low-power node is to customize the adjustment strategy by real-time collection of CPU load, cache occupancy rate and active duration of the communication unit, and generate frequency-voltage combination strategy in real time according to load changes. When the data cache reaches the preset capacity threshold, the hierarchical wake-up mechanism is triggered, and different power consumption management strategies are sent to the BSP platform server.

5. A multi-node data flow control system for a BSP development management platform as claimed in claim 1, characterized in that: The collaborative computing node is configured with a distributed collaborative optimization engine, which is a communication path optimizer based on a genetic algorithm. It calculates the optimal transmission path between nodes in real time, establishes a resource conflict prediction model, analyzes the probability of resource competition through historical communication data, and migrates excess computing tasks to idle nodes for execution.

6. A multi-node data flow control system for a BSP development management platform as claimed in claim 1, characterized in that: The traffic monitoring module is integrated with an intelligent prediction unit that uses a time series prediction model to predict traffic trends and dynamically adjusts the monitoring sampling frequency, using millisecond sampling when the load is high and second sampling when the load is low.

7. A multi-node data flow control system for a BSP development management platform as claimed in claim 1, characterized in that: The adaptive bandwidth allocation algorithm of the dynamic control module adopts the bandwidth utilization-delay joint optimization function, and the calculation formula is: in Q It is a comprehensive optimization index, which represents the comprehensive optimization target value of bandwidth allocation and is used to evaluate the pros and cons of the current bandwidth allocation strategy. U 带宽 is bandwidth utilization; D 延迟 is the data transmission delay; α and β are dynamic adjustment coefficients, which are configured differently according to the node type.

Images