Method and device for load balancing of video equipment access based on cluster architecture

Abstract

The application discloses a kind of method and equipment based on cluster architecture video equipment access load balancing, it is related to intelligent monitoring technical field, method includes: constructing the video access cluster including master node and slave node, provides the unified access management interface of equipment addition, modification, deletion to outside, and complete real-time data synchronization of equipment information in master-slave switching process;Receive the access request initiated by video equipment, in combination with equipment protocol type and each access node real-time running state, call load balancing scheduling algorithm to calculate optimal access node;When detecting node failure, offline or load exceeds preset threshold, automatically trigger equipment secondary distribution process, and migrate affected equipment to healthy access node;According to scheduling allocation result, load access driver matched with equipment protocol, complete equipment and access node protocol adaptation, continuously maintain equipment running state information, guarantee video stream stable access and transmission.

Description

Technical Field

[0001] This invention relates to the field of intelligent monitoring technology, and in particular to a method and device for accessing load balancing of video devices based on a cluster architecture. Background Technology

[0002] With the advancement of intelligent and visualized construction in railway transportation systems, massive amounts of high-definition video surveillance equipment have been deployed in stations, lines, tunnels, and trains. Equipment access is characterized by distributed deployment, high concurrency, strong dynamic switching, and multiple protocol coexistence, becoming a core support for railway security early warning, train operation command, and environmental monitoring. Currently, the industry generally faces problems such as high equipment access concentration, high risk of node failure, complex protocol adaptation, and slow response of onboard mobile equipment scheduling, making it difficult to guarantee the continuous and stable operation of monitoring services.

[0003] Currently, most mainstream solutions adopt single-node access or simple gateway forwarding. Some systems only achieve simple distribution through basic polling, which has obvious drawbacks: single nodes are prone to becoming access bottlenecks, and short-term concentrated requests can easily cause blocking; node failures can directly lead to the shutdown of a batch of devices, and there is no fault tolerance mechanism; load distribution only relies on fixed rules and cannot perceive the real-time status of CPU, bandwidth, and number of sessions; multi-vendor protocol adaptation is fixed, resulting in high expansion and maintenance costs.

[0004] Existing solutions still cannot solve the core challenges of balanced access of large-scale equipment, automatic migration of node failures, dynamic scheduling of mobile devices, and unified compatibility of multiple protocols in various railway scenarios. There is an urgent need for a highly available, adaptive, and scalable clustered load balancing solution to meet the real-time and high reliability requirements of railway video surveillance. Summary of the Invention

[0005] This invention provides a method and device for load balancing of video device access based on a cluster architecture, which solves the technical problems of uneven load and difficulty in multi-protocol compatibility of railway video device access.

[0006] On one hand, the present invention provides a method for load balancing of video devices based on a cluster architecture, the method comprising the following steps: Step S1: Construct a video access cluster containing master and slave nodes, provide a unified access management interface for adding, modifying, and deleting devices, realize automatic switching between master and slave nodes in the cluster, and complete real-time data synchronization of device information during the master-slave switching process to ensure high availability of access management services; Step S2: Receive the access request initiated by the video device, combine the device protocol type and the real-time operating status of each access node, call the load balancing scheduling algorithm to calculate the optimal access node, and complete the targeted allocation of access nodes for the video device; Step S3: Periodically collect the health status and load data of each access node. When a node failure, offline status or load exceeding a preset threshold is detected, the device secondary allocation process is automatically triggered to migrate the affected device to a healthy access node. Step S4: Based on the scheduling and allocation results, load the access driver that matches the device protocol, complete the protocol adaptation between the device and the access node, continuously maintain the device operating status information, and ensure stable access and transmission of video streams.

[0007] In one implementation of the present invention, in step S1, the unified access management interface is a RESTful interface. The master and slave nodes automatically switch based on the registration center to complete service registration and master selection logic. The master and slave nodes establish a stable communication link through RPC long connection. The master node synchronizes incremental device data to the slave node at a fixed period. When the slave node comes online, the master node automatically initiates full device information synchronization to ensure that the device data of the master and slave nodes are completely consistent.

[0008] In one implementation of the present invention, in step S2, the load balancing scheduling algorithm includes four types: minimum connection number strategy, weighted round-robin strategy, bandwidth priority strategy, and CPU load priority strategy. The scheduling strategy is dynamically configured and switched according to the business requirements of the railway scenario and the hardware configuration of the nodes to adapt to the load distribution requirements of different devices.

[0009] In one implementation of the present invention, the minimum number of connections strategy monitors the number of active connections of each node device in real time through the RPC interface and assigns new devices to the node with the fewest connections; the weighted round-robin strategy sets static / dynamic weights based on the node's CPU and memory hardware performance and allocates access requests in sequence according to the weight ratio; the bandwidth priority strategy collects the real-time bandwidth utilization rate of the nodes and prioritizes the nodes with the most remaining bandwidth; the CPU load priority strategy obtains the real-time CPU utilization rate of the nodes and prioritizes the nodes with the lowest CPU load.

[0010] In one implementation of the present invention, in step S2, the device protocol type covers multi-vendor proprietary video access protocols and Onvif and Rtsp general standard protocols. The device parameters and access rules of different protocols are abstracted into a unified data structure and stored in a standardized device information table to achieve unified management of multi-protocol devices.

[0011] In one implementation of the present invention, in step S3, the monitoring indicators of the node include four core data items: node online status, CPU load rate, network bandwidth utilization rate, and number of device session connections. Periodic status collection is performed at a preset period. When any indicator exceeds a preset threshold, or when the node is offline or malfunctions, device migration is triggered.

[0012] In one implementation of the present invention, in step S4, the access driver of the device adopts a component-based plug-in design. When the access node starts, it automatically scans the local directory and subdirectories, identifies and dynamically loads the driver plug-in of the corresponding protocol. After loading, a driver instance is created to complete the device protocol parsing and adaptation, and supports independent extension, replacement and upgrade of the driver plug-in.

[0013] In one implementation of the present invention, step S3, the secondary allocation and migration process of the device specifically involves: the master and slave nodes continuously synchronizing the device-node binding relationship, and the migration process does not require the device to re-initiate the access request, thereby achieving seamless switching of video stream access; accessing video and offline scenes of railway stations, tunnels, and vehicle-mounted mobile video devices.

[0014] In one implementation of the present invention, the video access cluster adopts a distributed process separation architecture, in which the device scheduling management module and the device scheduling module are integrated and run in the same process, while the device access module is deployed and run independently; the modules communicate with each other through RPC remote procedure calls to issue instructions, report status, and transmit data, thereby reducing process coupling and improving cluster stability.

[0015] On the other hand, the present invention also provides a video device access load balancing device based on a cluster architecture, the device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the aforementioned method for video device access load balancing based on a cluster architecture.

[0016] The present invention provides a method and device for load balancing of video devices based on a cluster architecture, which has the following beneficial effects: 1. By constructing a multi-node collaborative cluster and adopting a master-slave automatic switching and health detection mechanism, automatic fault tolerance for access node failures and seamless equipment migration are achieved, avoiding the downtime of batch equipment due to single point of failure, greatly improving the continuous availability and operational stability of the railway video surveillance system, and ensuring that critical monitoring services are not interrupted.

[0017] 2. By using a dynamic scheduling and allocation algorithm based on real-time load, intelligent load balancing with multi-dimensional perception of CPU, bandwidth, and number of connections is achieved. This effectively disperses the high-concurrency access pressure in hotspot areas such as stations and tunnels, eliminates local congestion and node overload, and significantly improves the access efficiency of large-scale devices and the system throughput.

[0018] 3. Through component-based driving and unified multi-protocol adaptation design, it achieves compatible access to various private protocols and general standard protocols, supports dynamic driver loading and flexible expansion, reduces the cost of device access modification and maintenance, and can quickly adapt to diverse scenario requirements such as railway lines, vehicle-mounted, and stations. Attached Figure Description

[0019] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this invention, illustrate exemplary embodiments of the invention and are used to explain the invention, but do not constitute an undue limitation of the invention. In the drawings: Figure 1 A flowchart illustrating a method for accessing load balancing for video devices based on a cluster architecture, provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of a video device accessing a load balancing device based on a cluster architecture, provided in an embodiment of the present invention. Figure 3 This is a schematic diagram of the system architecture provided in an embodiment of the present invention; Figure 4 A flowchart for adding / modifying devices provided in the embodiments of the present invention; Figure 5 A flowchart for device deletion provided in an embodiment of the present invention; Figure 6 A master-slave switching flowchart provided for embodiments of the present invention; Figure 7 A data synchronization timing diagram provided for embodiments of the present invention; Figure 8 This is a flowchart of the dynamic loading driver process for the access module provided in an embodiment of the present invention; Figure 9 The timing diagram for secondary allocation and dynamic scheduling of equipment provided in the embodiments of the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0021] This invention provides a method and device for load balancing of video devices based on a cluster architecture. The technical solution proposed in this invention will be described in detail below with reference to the accompanying drawings.

[0022] Figure 1This is a flowchart illustrating a method for load balancing of video devices based on a cluster architecture, provided as an embodiment of the present invention. Figure 1 As shown, the method mainly includes the following steps: Step S1: Construct a video access cluster that includes master and slave nodes, providing a unified access management interface for adding, modifying, and deleting devices, enabling automatic switching between master and slave nodes in the cluster, and completing real-time data synchronization of device information during the master-slave switching process to ensure high availability of access management services.

[0023] It should be noted that the unified access management interface is a RESTful interface. The automatic switching of master and slave nodes relies on the registration center to complete the service registration and master election logic. The master and slave nodes establish a stable communication link through RPC long connection. The master node synchronizes incremental device data to the slave node at a fixed period. When the slave node comes online, the master node automatically initiates full device information synchronization to ensure that the device data of the master and slave nodes are completely consistent.

[0024] Step S2: Receive the access request initiated by the video device, combine the device protocol type and the real-time operating status of each access node, call the load balancing scheduling algorithm to calculate the optimal access node, and complete the targeted allocation of access nodes for the video device.

[0025] It should be noted that the load balancing scheduling algorithm includes four types: minimum connection count strategy, weighted round-robin strategy, bandwidth priority strategy, and CPU load priority strategy. The scheduling strategy is dynamically configured and switched according to the business needs of the railway scenario and the hardware configuration of the nodes to adapt to the load distribution requirements of different devices.

[0026] The minimum number of connections strategy monitors the number of active connections of each node device in real time through the RPC interface and assigns new devices to the node with the fewest connections; the weighted round-robin strategy sets static / dynamic weights based on the node's CPU and memory hardware performance and allocates access requests in sequence according to the weight ratio; the bandwidth priority strategy collects the real-time bandwidth utilization of the nodes and prioritizes the nodes with the most remaining bandwidth; the CPU load priority strategy obtains the real-time CPU utilization of the nodes and prioritizes the nodes with the lowest CPU load.

[0027] The device protocol types cover multi-vendor proprietary video access protocols and Onvif and RTSP general standard protocols. The device parameters and access rules of different protocols are abstracted into a unified data structure and stored in a standardized device information table to achieve unified management of multi-protocol devices.

[0028] Step S3: Periodically collect the health status and load data of each access node. When a node failure, offline status, or load exceeding a preset threshold is detected, the device secondary allocation process is automatically triggered to migrate the affected device to a healthy access node.

[0029] It should be noted that the node monitoring metrics include four core data points: node online status, CPU load rate, network bandwidth utilization rate, and number of device session connections. Periodic status collection is performed at a preset interval. When any metric exceeds the preset threshold, or when the node goes offline or fails, device migration is triggered.

[0030] The secondary allocation and migration process of the device is as follows: the master and slave nodes continuously synchronize the device-node binding relationship, and the migration process does not require the device to re-initiate the access request, realizing seamless switching of video stream access; access to video and offline scenarios of railway stations, tunnels, and vehicle-mounted mobile video equipment.

[0031] Step S4: Based on the scheduling and allocation results, load the access driver that matches the device protocol, complete the protocol adaptation between the device and the access node, continuously maintain the device operating status information, and ensure stable access and transmission of video streams.

[0032] It should be noted that the device's access driver adopts a component-based plug-in design. When the access node starts, it automatically scans the local directory and subdirectories, identifies and dynamically loads the corresponding protocol driver plug-in. After loading, a driver instance is created to complete the device protocol parsing and adaptation, and supports independent extension, replacement and upgrade of driver plug-ins.

[0033] It should also be noted that the video access cluster adopts a distributed process separation architecture. The device scheduling management module and the device scheduling module are integrated and run in the same process, while the device access module is deployed and run independently. The modules communicate with each other through RPC remote procedure calls to issue instructions, report status, and transmit data, thereby reducing process coupling and improving cluster stability.

[0034] The above describes a video device access load balancing system based on a cluster architecture, as provided by an embodiment of the present invention. Based on the same inventive concept, this invention also provides a video device access load balancing device based on a cluster architecture. Figure 2 This invention provides a schematic diagram of a video device accessing a load balancing device based on a cluster architecture, as shown in the embodiment of the invention. Figure 2 As shown, the device mainly includes: at least one processor 201; and a memory 202 communicatively connected to the at least one processor; wherein the memory 202 stores instructions that can be executed by the at least one processor 201, and the instructions are executed by the at least one processor 201 to enable the at least one processor 201 to complete the aforementioned method for load balancing of video device access based on a cluster architecture.

[0035] The specific implementation steps of this invention are as follows: First, let's describe the system architecture and module functions. The overall architecture of this system is as follows: Figure 3As shown. This system is suitable for distributed cluster scenarios; the system consists of a device scheduling management module, a device scheduling module, and a device access module; in a distributed cluster deployment, the device scheduling management module and the device scheduling module run integrated in the same process, while the device access module runs independently in another process, and the two communicate with each other via Remote Procedure Call (RPC); the main functions of the device scheduling management module are: unified device management interface, protocol allocation and device pool management, support for master-slave switching and high availability, and support for data synchronization between master and slave nodes.

[0036] The main functions of the device scheduling module include: supporting multiple scheduling algorithms, specifically minimum connection count, weighted round-robin, bandwidth priority, and CPU load priority strategies; minimum connection count is achieved by establishing a long connection with the device access module through Remote Procedure Call (RPC) technology, and using the RPC interface to monitor and obtain the active device connection count of each node in real time, assigning new devices to the node with the fewest current device connections; weighted round-robin is achieved by assigning a static or dynamic weight to each node based on its actual hardware performance, and allocating new devices proportionally, with higher weights being allocated more frequently. This method is suitable for inter-node CPU and CPU load balancing. In scenarios with uneven distribution of resources such as memory, the following methods are employed: Bandwidth priority is given: the bandwidth utilization of each node is obtained through system interface calls, and the node with the lowest bandwidth utilization is selected first. CPU load priority is also given: the CPU utilization of each node is obtained through system interface calls, and the node with the lowest CPU utilization is selected. This method is suitable for scenarios where nodes are performing multiple tasks simultaneously, such as transcoding, video analysis, or intelligent monitoring. Multiple configurable scheduling algorithms are also supported. Specifically, it supports access node status monitoring, periodically monitoring the status and health of access nodes, and automatically triggering device migration when a node goes offline, fails, or exceeds its load limit.

[0037] The main functions of the device access module are: supporting driver plug-in and supporting device information maintenance.

[0038] This system specifically executes the following steps through these modules: 1. Device access request processing. The business platform submits requests to add, modify, or delete devices through a unified interface; the system abstracts device information from different manufacturers and using different protocols into a unified structure and stores it in the device information table.

[0039] 2. Equipment Scheduling and Allocation. The flowchart for adding / modifying equipment is as follows: Figure 4 As shown, the device removal flowchart is as follows: Figure 5As shown, when a new device is detected that needs to access the network, the scheduling module selects the optimal node based on the real-time status of the nodes. The scheduling strategies include: minimum number of connections priority: prioritizing the node with the fewest active connections; weighted round-robin: allocating resources according to the weight of the node's computational resource configuration; bandwidth priority: prioritizing the node with the lowest bandwidth usage; CPU load priority: prioritizing the node with the lowest CPU utilization. The device access information is synchronously sent to the corresponding access node to complete the device binding and unbinding.

[0040] 3. Node status monitoring and master-slave switching, such as Figure 6 As shown in the master-slave switching flowchart, the system periodically monitors the status of each access node; when a master node failure or offline is detected, a new master is re-elected through the registration center (such as Nacos) to achieve automatic master-slave switching; the new master node after switching is responsible for the scheduling and allocation of devices to avoid access interruption due to single point of failure.

[0041] 4. Device data synchronization, such as Figure 7 As shown in the data synchronization sequence diagram, the master node periodically synchronizes device information with the slave node through a long-lived Remote Procedure Call (RPC) connection; when a slave node comes online, the master node automatically establishes a connection and synchronizes data to ensure the consistency of device information between the master and slave nodes.

[0042] 5. Access driver dynamically loaded, such as Figure 8 As shown in the flowchart of the dynamic driver loading process of the access module, the access node automatically traverses the directory and dynamically loads driver plugins for different protocols when it starts up; it provides unified adaptation capabilities for devices from different manufacturers and with different protocols, ensuring scalability.

[0043] 6. Secondary allocation and dynamic migration, such as Figure 9 As shown in the device secondary allocation and dynamic scheduling sequence diagram, when an access node is detected to be offline or overloaded, the scheduling module immediately triggers secondary allocation; the affected devices are migrated to healthy and low-load nodes according to the scheduling strategy to ensure uninterrupted access.

[0044] Specific applications are as follows: Tunnel entrance environment monitoring scenario: At the tunnel entrances along railway lines, multiple high-definition video surveillance devices are usually deployed to monitor the operation status of trains entering and exiting the tunnel and the environmental conditions at the tunnel entrance in real time.

[0045] Application issues: 1. The environment at tunnel entrances is complex, and monitoring points are generally concentrated. If devices start streaming simultaneously, it can easily lead to instantaneous overload of a single node; 2. Tunnel entrances are often located in remote areas with large fluctuations in network bandwidth. Once an access node fails or bandwidth is insufficient, it will directly lead to the loss of video footage, affecting traffic safety monitoring.

[0046] Implementation process: 1. The surveillance cameras at the tunnel entrance are uniformly registered to the system through the access interface; 2. The scheduling module adopts the minimum number of connections + bandwidth priority strategy when allocating: (1) Prioritize ensuring the balanced distribution of devices at each access node; (2) At the same time, in combination with the bandwidth utilization of the nodes, prioritize the nodes with stable links and the most remaining bandwidth.

[0047] 3. When a sudden drop in network bandwidth or abnormal offline status is detected at an access node, the scheduling module immediately triggers a secondary allocation mechanism to migrate the affected tunnel entrance equipment to other normal nodes.

[0048] 4. During the migration process, device information remains synchronized between master and slave nodes to ensure rapid scheduling and switching without interruption of video feed.

[0049] Technical effects: This implementation ensures real-time access and continuous operation of the tunnel entrance monitoring equipment. Even in the event of network fluctuations or node failures, the video stream can be quickly migrated to other nodes, avoiding the loss of monitoring footage and improving the safety assurance capabilities of key points at railway tunnel entrances.

[0050] The various embodiments in this invention are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the device embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0051] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0052] The above description is merely an embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of the present invention should be included within the scope of the claims of the present invention.

Claims

1. A method for load balancing of video devices based on a cluster architecture, characterized in that, The method includes the following steps: Step S1: Construct a video access cluster containing master and slave nodes, provide a unified access management interface for adding, modifying, and deleting devices, realize automatic switching between master and slave nodes in the cluster, and complete real-time data synchronization of device information during the master-slave switching process to ensure high availability of access management services; Step S2: Receive the access request initiated by the video device, combine the device protocol type and the real-time operating status of each access node, call the load balancing scheduling algorithm to calculate the optimal access node, and complete the targeted allocation of access nodes for the video device; Step S3: Periodically collect the health status and load data of each access node. When a node failure, offline status or load exceeding a preset threshold is detected, the device secondary allocation process is automatically triggered to migrate the affected device to a healthy access node. Step S4: Based on the scheduling and allocation results, load the access driver that matches the device protocol, complete the protocol adaptation between the device and the access node, continuously maintain the device operating status information, and ensure stable access and transmission of video streams.

2. The method for load balancing of video devices based on a cluster architecture according to claim 1, characterized in that, In step S1, the unified access management interface is a RESTful interface. The automatic switching of master and slave nodes relies on the registration center to complete the service registration and master selection logic. The master and slave nodes establish a stable communication link through RPC long connection. The master node synchronizes incremental device data to the slave node at a fixed period. When the slave node comes online, the master node automatically initiates full device information synchronization to ensure that the device data of the master and slave nodes are completely consistent.

3. The method for load balancing of video devices based on a cluster architecture according to claim 1, characterized in that, In step S2, the load balancing scheduling algorithm includes four types: minimum connection number strategy, weighted round-robin strategy, bandwidth priority strategy, and CPU load priority strategy. The scheduling strategy is dynamically configured and switched according to the business needs of the railway scenario and the hardware configuration of the nodes to adapt to the load distribution requirements of different devices.

4. The method for load balancing of video devices based on a cluster architecture according to claim 3, characterized in that, The minimum number of connections strategy monitors the number of active connections of each node device in real time through the RPC interface and assigns new devices to the node with the fewest connections; the weighted round-robin strategy sets static / dynamic weights based on the node's CPU and memory hardware performance and allocates access requests in sequence according to the weight ratio; the bandwidth priority strategy collects the real-time bandwidth utilization of the nodes and prioritizes the nodes with the most remaining bandwidth; the CPU load priority strategy obtains the real-time CPU utilization of the nodes and prioritizes the nodes with the lowest CPU load.

5. The method for load balancing of video devices based on a cluster architecture according to claim 1, characterized in that, In step S2, the device protocol types cover multi-vendor proprietary video access protocols and Onvif and Rtsp general standard protocols. The device parameters and access rules of different protocols are abstracted into a unified data structure and stored in a standardized device information table to achieve unified management of multi-protocol devices.

6. The method for load balancing of video devices based on a cluster architecture according to claim 1, characterized in that, In step S3, the monitoring indicators of the node include four core data items: node online status, CPU load rate, network bandwidth utilization rate, and number of device session connections. Periodic status collection is performed at a preset period. When any indicator exceeds the preset threshold, or when the node is offline or malfunctions, device migration is triggered.

7. The method for load balancing of video devices based on a cluster architecture according to claim 1, characterized in that, In step S4, the device's access driver adopts a component-based plug-in design. When the access node starts up, it automatically scans the local directory and subdirectories, identifies and dynamically loads the corresponding protocol driver plug-in. Once loaded, a driver instance is created, and device protocol parsing and adaptation are completed. Driver plugins can be independently extended, replaced, and upgraded.

8. The method for load balancing of video devices based on a cluster architecture according to claim 1, characterized in that, In step S3, the secondary allocation and migration process of the device is as follows: the master and slave nodes continuously synchronize the device-node binding relationship, and the migration process does not require the device to re-initiate the access request, realizing seamless switching of video stream access; accessing video and offline scenes of railway stations, tunnels, and vehicle-mounted mobile video devices.

9. A method for load balancing of video devices based on a cluster architecture according to claim 8, characterized in that, The video access cluster adopts a distributed process separation architecture. The device scheduling and management module and the device scheduling module run in the same process, while the device access module is deployed and runs independently. The modules communicate with each other through RPC remote procedure calls to issue commands, report status, and transmit data, thereby reducing process coupling and improving cluster stability.

10. A video device access load balancing device based on a cluster architecture, characterized in that, The device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform a method for load balancing of video device access based on a cluster architecture as described in any one of claims 1-9.

Images