Isolation vector device lateral performance expansion system

Abstract

The application relates to an isolated vector device horizontal performance expansion system. The system comprises an isolated vector device cluster unit, at least two isolated vector device nodes with power special horizontal isolation capability; a cluster cooperative management unit in communication connection with the isolated vector device nodes; a data shunting and parallel processing unit between the cluster cooperative management unit and each isolated vector device node; a safe strategy disturbance-free synchronization unit connected with the cluster cooperative management unit; and a state monitoring and disaster recovery scheduling unit in communication connection with the isolated vector device cluster unit and the cluster cooperative management unit. The isolated vector device horizontal performance expansion system can balance high safety, high reliability and business adaptability, and meet the application requirements of a new power system source network load storage interaction scene.

Description

Technical Field

[0001] This application relates to the field of power system network security technology, and in particular to a system for horizontal performance expansion of an isolation vector device. Background Technology

[0002] With the continuous advancement of the construction of new power systems, a wide range of entities, including wind power, photovoltaics, energy storage, and virtual power plants, are widely connected to the grid. Data interaction between the production control area and the management information area is showing new characteristics of high-frequency interaction, high-volume transmission, and high requirements for both real-time performance and security. This has placed new application demands on network security isolation devices, which serve as the core of horizontal security protection for the power secondary system, in terms of throughput, forwarding latency, scalability, and security protection capabilities.

[0003] Power-specific forward and reverse network security isolation devices, through a physically isolated unidirectional ferry architecture, achieve secure data exchange across security zones. This is a core technology for horizontal boundary protection in the power industry and has been widely applied in scenarios such as power grid dispatching and new energy power plants. Traditional isolation devices, with their fixed single-device architecture, can meet the low-frequency, low-volume cross-regional data exchange needs of traditional power grids. The industry has also explored technologies related to multi-device collaboration and load balancing, resulting in solutions such as multi-device concurrent scheduling, array gateway load balancing, and repeater cluster deployment. These solutions provide technical approaches to improving the throughput of cross-regional isolation exchange systems and have achieved certain performance improvements in traditional power grid fixed business scenarios.

[0004] However, traditional clustered scheduling often adopts an external architecture, using external gateways and repeaters to achieve multi-device traffic scheduling. The cluster management capabilities and the inherent security architecture of the isolation devices are independent of each other. In high-frequency collaborative scenarios, the control signaling transmission is difficult to fully adapt to the compliance requirements of power secondary system security protection. Existing load balancing often uses general server performance indicators or basic file-level hash rules, which are not well matched with the performance bottleneck of the dedicated switching chip in the core of the isolation device, making it difficult to achieve a linear increase in system throughput with the number of nodes. At the same time, in mixed transmission scenarios, session timing disorder and core business bandwidth are prone to occur. Moreover, the session state synchronization of power long-connection services is not considered during fault switching, which can easily lead to service session interruption and is difficult to adapt to the core requirements of continuous operation of new power system services. Summary of the Invention

[0005] Therefore, it is necessary to provide a horizontal performance expansion system for isolation vector devices that can balance high security, high reliability, and business adaptability to meet the application requirements of new power system source-grid-load-storage interaction scenarios, addressing the aforementioned technical issues.

[0006] In a first aspect, this application provides a horizontal performance expansion system for an isolation vector device, comprising: an isolation vector device cluster unit, a cluster collaborative management and control unit, a data splitting and parallel processing unit, a security policy non-disruptive synchronization unit, and a status monitoring and disaster recovery scheduling unit.

[0007] An isolation vector device cluster unit includes at least two isolation vector device nodes with dedicated power lateral isolation capabilities;

[0008] The cluster collaborative management and control unit communicates with the isolation vector device nodes to realize the lifecycle management of cluster nodes, dynamic topology discovery, expansion node management, and global coordination of cluster operation status.

[0009] The data splitting and parallel processing unit is located between the cluster collaborative management and control unit and each isolation vector device node. It is used to distribute the power big data exchange traffic to be processed to the isolation vector device nodes to perform isolation forwarding processing in parallel, and to complete the orderly reassembly of the processed packets.

[0010] The security policy non-disruptive synchronization unit is connected to the cluster collaborative management and control unit to synchronize the existing security policies under the original boundary isolation scenario to the isolation vector device node.

[0011] The status monitoring and disaster recovery scheduling unit communicates with the isolation vector device cluster unit and the cluster collaborative management and control unit to collect the operating data of each isolation vector device node and perform load balancing scheduling and traffic switching of faulty nodes within the cluster.

[0012] In one embodiment, the isolation vector device node is also equipped with a standardized data interaction interface and a dual-host dual-ferry isolation hardware architecture; the standardized data interaction interface is used to build a boundary isolation resource pool; the dual-host dual-ferry isolation hardware architecture includes an inner host and an outer host, and a physically isolated one-way ferry channel for business data and a one-way ferry channel for cluster control are set between the inner host and the outer host.

[0013] In one embodiment, the cluster collaborative management and control unit is embedded in the dual-host dual-ferry isolation hardware architecture of each isolation vector device node. The cluster collaborative management and control unit has a built-in power-specific isolation vector quantization model to realize the quantitative scheduling of global isolation resources of the cluster.

[0014] In one embodiment, the data offloading and parallel processing unit incorporates a power protocol-aware service characteristics and network 5-tuple dual-dimensional session consistency offloading algorithm to perform power service session-level consistent traffic distribution; and / or

[0015] The status monitoring and disaster recovery scheduling unit uses the dynamic isolation vector values ​​generated by each isolation vector device node through the power-specific isolation vector quantization model as the core scheduling basis to perform load balancing and disaster recovery switching.

[0016] In one embodiment, the cluster collaborative management and control unit adopts a redundant architecture with one master and multiple backups, including a master management node and backup management nodes.

[0017] The master control node is used for the global configuration and scheduling of cluster collaborative control units;

[0018] The backup management node is used to synchronize the configuration data and operating status of the primary management node in real time, and to seamlessly switch over to take over management authority in the event of a failure of the primary management node.

[0019] In one embodiment, the security policy non-disruptive synchronization unit incorporates a policy format conversion subunit, a policy consistency verification subunit, and a policy atomic synchronization subunit. The policy format conversion subunit is compatible with the security policy format of the power forward and reverse isolation devices, enabling the import of existing policies without modification. The policy consistency verification subunit is used to verify in real time whether the security policies of all isolation vector device nodes in the cluster are completely consistent, and automatically triggers synchronization correction when deviations occur. The policy atomic synchronization subunit ensures that policy changes take effect simultaneously throughout the entire cluster, and policy changes only need to be configured once at the global entry point to achieve synchronized effect across the entire cluster.

[0020] In one embodiment, the operational data collected by the status monitoring and disaster recovery scheduling unit for each isolation vector device node includes: CPU utilization, memory usage, throughput, bandwidth usage, number of session connections, link connectivity status, and the success rate of matching isolation forwarding rules.

[0021] In one embodiment, each isolated vector device node has a built-in dedicated isolated data exchange chip, trusted computing module, and national cryptographic encryption component. The trusted computing module is used to realize trusted startup, version integrity verification, and trusted supply chain management of the isolated vector device node. The dedicated isolated data exchange chip is equipped with multiple physically isolated hardware shuttle queues of different priorities, which correspond one-to-one with the power business priorities. The bandwidth between the hardware shuttle queues is hard isolated and does not crowd each other. The cluster collaborative management unit establishes a global resource pool for hardware shuttle queues of different priorities and hard reserves the preset total cluster bandwidth for the highest priority hardware shuttle queue.

[0022] In one embodiment, the security policy non-disruptive synchronization unit adopts a three-tiered architecture consisting of a global basic policy, a scenario-specific policy, and a node-specific policy.

[0023] In one embodiment, a cross-platform secure communication component is also included, deployed between the cluster collaborative management unit and each access service terminal. The cross-platform secure communication component includes:

[0024] The national cryptographic authentication module is used to achieve two-way identity authentication and communication link encryption between cluster collaborative management and control units and between cluster collaborative management and control units and access service terminals.

[0025] The credential collection module is used to collect device fingerprints, operating system integrity, security patch status, and access behavior characteristics of accessing business terminals;

[0026] The terminal identity verification module is used to verify the integrity of access service terminals and control authorized access, allowing only access service terminals that pass the verification to access the cluster collaborative management unit.

[0027] The aforementioned horizontal performance expansion system for isolation vector devices comprises an isolation vector device cluster unit, including at least two isolation vector device nodes with dedicated power horizontal isolation capabilities; a cluster collaborative management unit, communicatively connected to the isolation vector device nodes, used for lifecycle management of cluster nodes, dynamic topology discovery, management of extended nodes, and global coordination of cluster operation status; a data distribution and parallel processing unit, located between the cluster collaborative management unit and each isolation vector device node, used to distribute the power big data exchange traffic to be processed to the isolation vector device nodes for parallel isolation forwarding processing and to complete the orderly reassembly of processed packets; a security policy non-disruptive synchronization unit, connected to the cluster collaborative management unit, used to synchronize existing security policies under the original boundary isolation scenario to the isolation vector device nodes; and a status monitoring and disaster recovery scheduling unit, connected to the isolation vector device cluster unit. The communication connection between the data collection device cluster unit and the cluster collaborative management and control unit is used to collect the operating data of each isolated vector device node and perform load balancing scheduling and traffic switching of faulty nodes within the cluster. This overcomes the problems of traditional technologies, such as the difficulty of clustered scheduling fully adapting to the compliance requirements of power secondary system security protection, the easy occurrence of session timing disorder in mixed transmission scenarios, and the squeezing of core business bandwidth. It takes into account high security, high reliability, and business adaptability, meets the application requirements of the new power system source-grid-load-storage interaction scenario, and is compatible with the deployment architecture, security policies, and business processes of existing power system horizontal isolation devices. It can be deployed and launched without modifying existing business systems or interrupting existing business, and does not require adaptation and modification of existing business terminals, thus reducing the upgrade and transformation cost and deployment cycle of the new power system horizontal security protection system. Attached Figure Description

[0028] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1 This is a block diagram of a lateral performance extension system for an isolation vector device provided in one embodiment.

[0030] Figure 2 This is a block diagram of a security policy non-interference synchronization unit in a lateral performance extension system of an isolation vector device provided in one embodiment. Detailed Implementation

[0031] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0032] It is understood that the terms “first,” “second,” etc., used in this application may be used herein to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from another element.

[0033] With the continuous advancement of the construction of new power systems, a wide range of entities, including wind power, photovoltaics, energy storage, and virtual power plants, are widely connected to the grid. Data interaction between the production control area and the management information area is showing new characteristics of high-frequency interaction, high-volume transmission, and high requirements for both real-time performance and security. This has placed new application demands on network security isolation devices, which serve as the core of horizontal security protection for the power secondary system, in terms of throughput, forwarding latency, scalability, and security protection capabilities.

[0034] Power-specific forward and reverse network security isolation devices, through a physically isolated unidirectional ferry architecture, achieve secure data exchange across security zones. This is a core technology for horizontal boundary protection in the power industry and has been widely applied in scenarios such as power grid dispatching and new energy power plants. Traditional isolation devices, with their fixed single-device architecture, can meet the low-frequency, low-volume cross-regional data exchange needs of traditional power grids. The industry has also explored technologies related to multi-device collaboration and load balancing, resulting in solutions such as multi-device concurrent scheduling, array gateway load balancing, and repeater cluster deployment. These solutions provide technical approaches to improving the throughput of cross-regional isolation exchange systems and have achieved certain performance improvements in traditional power grid fixed business scenarios.

[0035] However, in scenarios involving frequent interactions between power generation, grid, load, and storage in new power systems, traditional clustered dispatching often employs an external architecture. The cluster management capabilities and the intrinsic security architecture of the isolation devices are independent of each other. Control signaling transmission in high-frequency collaborative scenarios is difficult to fully adapt to the compliance requirements of power secondary system security protection. The performance bottleneck of the dedicated switching chip at the core of the traditional load balancing and isolation devices is insufficient, making it difficult to achieve a linear increase in system throughput with the number of nodes. At the same time, traditional load balancing algorithms are not deeply adapted to the session characteristics of power-specific protocols and cannot distinguish the priority of grid control services and non-real-time big data services. In mixed transmission scenarios, session timing disorder and core service bandwidth squeezing are prone to occur. Furthermore, the session state synchronization of long-connection power services is not considered during fault switching, which can easily lead to service session interruption and is difficult to adapt to the core requirements of continuous operation of new power system services.

[0036] In view of this, embodiments of this application provide a system for laterally scaling the performance of an isolated vector device. In one exemplary embodiment, such as Figure 1 As shown, this application provides a horizontal performance expansion system for an isolation vector device. The horizontal performance expansion system for an isolation vector device may include: an isolation vector device cluster unit 101, a cluster collaborative management and control unit 102, a data offloading and parallel processing unit 103, a security policy non-disruptive synchronization unit 104, and a status monitoring and disaster recovery scheduling unit 105. Wherein:

[0037] The isolation vector device cluster unit 101 includes at least two isolation vector device nodes with dedicated power horizontal isolation capabilities. The cluster collaborative management unit 102 is communicatively connected to the isolation vector device nodes and is used to achieve lifecycle management of cluster nodes, dynamic topology discovery, expansion node management, and global coordination of cluster operation status. The data distribution and parallel processing unit 103 is located between the cluster collaborative management unit 102 and each isolation vector device node, and is used to distribute the power big data exchange traffic to be processed to the isolation vector device nodes for parallel isolation forwarding processing, and to complete the orderly reassembly of processed packets. The security policy non-disruptive synchronization unit 104 is connected to the cluster collaborative management unit 102 and is used to synchronize existing security policies under the original boundary isolation scenario to the isolation vector device nodes. The status monitoring and disaster recovery scheduling unit 105 is communicatively connected to the isolation vector device cluster unit 101 and the cluster collaborative management unit 102, and is used to collect the operating data of each isolation vector device node and perform load balancing scheduling and traffic switching of faulty nodes within the cluster.

[0038] As an example, each isolation vector device node is equipped with a standardized data interaction interface and an intrinsically secure dual-host dual-ferry isolation hardware architecture; the standardized data interaction interface is used to build a boundary isolation resource pool; the dual-host dual-ferry isolation hardware architecture is an intrinsically secure dual-host dual-ferry isolation hardware architecture, including an inner host and an outer host, and a physically isolated one-way business data ferry channel and a cluster control one-way ferry channel are set between the inner host and the outer host.

[0039] As an example, the operational data collected by the status monitoring and disaster recovery scheduling unit 105 for each isolated vector device node includes: CPU utilization, memory usage, throughput, bandwidth usage, number of session connections, link connectivity status, and the success rate of matching isolation forwarding rules.

[0040] Optionally, the isolation vector device cluster unit 101 can consist of four isomorphic isolation vector device nodes with gigabit-level power-specific isolation capabilities, named Node 1, Node 2, Node 3, and Node 4, respectively. Each isolation vector device node has a rated unidirectional forwarding throughput of 1Gbps, and the overall design throughput of the isolation vector device cluster unit 101 is at least 3.8Gbps, preferably reaching 3.92Gbps, which is 3.92 times the throughput of a single isolation vector device node. This meets the high-frequency big data exchange needs of source-grid-load-storage entities such as new energy power plants, virtual power plants, and energy storage power stations, with a linear improvement ratio exceeding 98%, satisfying the design requirements for linear improvement. The end-to-end average forwarding latency is 120μs, which is not significantly different from the forwarding latency of a single node, meeting the low-latency transmission requirements of new power systems.

[0041] As an example, each isolation vector device node has a built-in dedicated isolation data exchange chip, trusted computing module, and national cryptographic encryption components, among which:

[0042] The Trusted Computing module is used to implement trusted startup, version integrity verification, and trusted supply chain management for isolated vector device nodes.

[0043] The dedicated isolated data exchange chip is equipped with multiple physically isolated hardware shuttle queues of different priorities, which correspond one-to-one with the priorities of power services. The bandwidth between the hardware shuttle queues is hard isolated and does not crowd each other out.

[0044] The cluster collaborative management and control unit 102 establishes a global resource pool for hardware transfer queues of different priorities, and reserves the preset total cluster bandwidth for the highest priority hardware transfer queue.

[0045] As an example, two sets of physically isolated dedicated ASIC isolation transfer chip channels can be set up, namely a one-way transfer channel for business data and a one-way transfer channel for cluster control.

[0046] As an example, version integrity verification performs a step-by-step check on the integrity of the BIOS, embedded operating system, and isolated applications. If the check fails, the node is prevented from starting. Optionally, the trusted computing module has a built-in trusted version management submodule that works in conjunction with the trusted supply chain management platform to verify the firmware and software upgrade packages of the nodes. Only legitimate versions that pass the verification are allowed to be upgraded, and all version change operations are recorded to achieve full lifecycle audit traceability.

[0047] Optionally, the hardware transfer queue can be set up in 4 groups: Level I power grid control business queue, Level II power grid status business queue, Level III real-time acquisition business queue, and Level IV non-real-time big data business queue.

[0048] Specifically, the cluster collaborative management and control unit 102 hard-resers 20% of the total cluster bandwidth for the Level I power grid control service queue to ensure priority transmission of core control services in extreme scenarios. Preferably, under full cluster bandwidth conditions, the end-to-end forwarding latency fluctuation of Level I power grid control services does not exceed 10μs, with no packet loss, fully meeting the high reliability requirements of power grid control services.

[0049] For details, please continue reading. Figure 1 The dual-host, dual-ferry isolated hardware architecture allows the inner host (specifically via a gigabit Ethernet port) to connect to the power production control zone 106, integrating modules for cluster management and control proxy, traffic distribution execution, policy synchronization, and session state management. The outer host (specifically via a gigabit Ethernet port) connects to the power management information zone 107, integrating modules for data reassembly, session keep-alive, and status reporting. There is no bidirectional network layer connection between the inner and outer hosts. The one-way ferry channel for business data between the inner and outer hosts only supports one-way ferry of application layer data, transmitting power business messages or files unidirectionally, disabling all network layer protocols. The one-way ferry channel for cluster control only transmits cluster coordination signaling unidirectionally. The two channels are hardware isolated and have no data overlap. The signaling transmission of the one-way ferry channel for cluster control and the transmission of the one-way ferry channel for business data are strictly time-sequential and only executed during the idle window of the business channel.

[0050] As an example, the cluster collaborative management and control unit 102 is deeply embedded in the dual-host dual-ferry isolation hardware architecture of each isolation vector device node, without external forwarding devices or gateway devices. The cluster collaborative management and control unit 102 has a built-in power-specific isolation vector quantization model to realize the quantitative scheduling of global isolation resources of the cluster.

[0051] Specifically, the cluster collaborative management and control unit 102 is completely embedded in the internal host of each isolated vector device node, without any external array gateway or repeater device. All cluster control signals are transmitted unidirectionally through the built-in cluster control one-way transfer channel of the isolated vector device node, without bidirectional connection across the network layer of the security zone, which fully complies with the security protection requirements of the power secondary system.

[0052] In one embodiment, the cluster collaborative management and control unit 102 adopts a redundant architecture with one master and multiple backups, including a master management and control node and backup management and control nodes. The master management and control node is used for the global configuration and scheduling of the cluster collaborative management and control unit 102; the backup management and control node is used to synchronize the configuration data and running status of the master management and control node in real time, and to complete a seamless switchover within a preset time (for example, it can be but is not limited to 500ms) to take over the management and control authority when the master management and control node fails.

[0053] Optionally, the cluster collaborative management and control unit 102 can adopt a redundant architecture of one primary and two backups. The primary management and control node is deployed on the internal host of node 1, the first backup management and control node is deployed on the internal host of node 2, and the second backup management and control node is deployed on the internal host of node 3. The primary management and control node is responsible for the global configuration and scheduling of the cluster, and the backup management and control node synchronizes the configuration data of the primary node in real time with a synchronization period of 100ms. When the primary management and control node fails, the first backup management and control node can complete seamless switching within 500ms and take over the cluster management and control authority.

[0054] Specifically, during the initial deployment of the cluster collaborative management and control unit 102, it is necessary to import the existing security policies of the original horizontal isolation device, configure the global parameters, three-level policy architecture rules, isolation vector value calculation weights, and service priority bandwidth reservation ratios of the cluster collaborative management and control unit 102, and perform initial configuration of the main management and control node. After the other standby management and control nodes are powered on, they send a registration request to the main management and control node through a dedicated out-of-band security management port. The main management and control node performs dual trusted verification based on national cryptographic standards on the standby management and control nodes. The dual trusted verification includes: first, completing two-way identity authentication through the national cryptographic SM2 algorithm to verify the legality of the power dedicated equipment certificate of the management and control node; then, completing the full integrity verification of the BIOS, operating system, and isolation application of the standby management and control node through the trusted computing module. After the verification is passed, the isolation vector value benchmark test of the standby management and control node is automatically completed. After the test result reaches the minimum threshold of the cluster collaborative management and control unit 102, the global security policy and the configuration parameters of the cluster collaborative management and control unit 102 are synchronized to the newly added standby management and control node, completing the automatic management of the standby management and control node. Once the management is completed, the newly added backup management node can enter the cluster isolation resource pool to take over traffic forwarding tasks, and the entire process does not require interruption of the original business operation.

[0055] In the example above, the inherently secure dual-host, dual-ferry isolation hardware architecture eliminates the need for external gateways or repeater devices. It resolves the issue of non-compliance in control signaling across security zones by using physically isolated one-way ferry channels for business data and cluster control. The primary management node performs dual trusted verification based on national cryptographic standards on the backup management node, improving verification reliability. Upon successful verification, the backup management node is automatically added to the management system without interrupting existing business operations.

[0056] Optionally, the power-specific isolation vector quantization model defines the dynamic isolation vector value for each isolation vector device node. It is generated by weighted calculation of six power-specific indicators: one-way ferry hardware throughput, security rule matching concurrency, national cryptographic encryption and decryption performance, node trust security level, power long connection session processing capability, and historical fault weight. The cluster collaborative management and control unit 102 constructs a dynamic computing power view of the cluster isolation resource pool based on the dynamic isolation vector values ​​of global nodes. All traffic splitting, load balancing, disaster recovery switching, and node management operations are based solely on the dynamic isolation vector values.

[0057] In the above example, the power-specific isolation vector quantization model uses the dynamic isolation vector value as the sole core basis for the cluster collaborative management and control unit 102, rather than general CPU or memory indicators, to match the hardware performance bottleneck of the dedicated shuttle chip of the isolation device, thereby improving scheduling accuracy and adaptability.

[0058] As an example, the data offloading and parallel processing unit 103 incorporates a power protocol-aware service characteristics and network 5-tuple dual-dimensional session consistency offloading algorithm to perform power service session-level consistent traffic distribution.

[0059] As an example, the data splitting and parallel processing unit 103 includes the following three types: power protocol-aware service characteristics, network five-tuple two-dimensional session consistency splitting algorithm, and power protocol-aware service characteristics and network five-tuple two-dimensional session consistency splitting algorithm. Specifically, the execution logic of the power protocol-aware service characteristics and network five-tuple two-dimensional session consistency splitting algorithm is as follows: First, by parsing the power application layer protocol of the message, accurately identify IEC104, IEC61850MMS / GOOSE, and Modbus power-specific protocols, extract the core power service characteristics including service type (remote control / remote adjustment / remote signaling / remote measurement), intelligent electronic device (IED) device address, application service data unit (ASDU) identifier, application session ID, and source or destination service address, and generate power protocol-aware service characteristics. Then, combine the extracted power protocol-aware service characteristics with the network layer five-tuple to generate a two-dimensional hash factor, and calculate the target isolation vector device node based on the consistent hashing algorithm to ensure that all messages of the same power service session are 100% distributed to the same isolation vector device node for processing.

[0060] Specifically, the data splitting and parallel processing unit 103 can be divided into a front-end splitting module and a back-end reassembly module. The front-end splitting module is deployed at the production control zone 106 entrance switch of the cluster collaborative management and control unit 102. When the pending source-grid-load-storage business traffic enters the production control zone 106 entrance switch, the front-end splitting module performs deep parsing of the packets. It first identifies the power application layer protocol type and extracts the core business characteristics of protocols such as IEC104 / IEC61850 / Modbus, including business type (remote control / remote adjustment / remote signaling / remote measurement), IED device address, and ASDU identifier. The system identifies the application session ID and source / destination service address, then extracts the network layer five-tuple information, combines the two to generate a two-dimensional hash factor, and uses a consistent hashing algorithm to calculate the target isolation vector device node corresponding to the session to which the packet belongs. All packets of the same session are uniformly distributed to the isolation vector device node for processing, ensuring the integrity and timing correctness of power service sessions. The backend reassembly module is deployed on the management information area 107 egress switch of the cluster collaborative management and control unit 102. Each isolation vector device node has a built-in traffic distribution proxy module, which works in conjunction with the front-end traffic distribution module to perform traffic distribution and parallel processing.

[0061] Optionally, the load balancing scheduling adopts the isolation vector value redundancy priority algorithm, which prioritizes the distribution of new traffic to the node with the highest current isolation vector value redundancy. All isolation vector device nodes within the cluster collaborative management and control unit 102 synchronize the full status data of all active power sessions at a period of 10ms through the out-of-band security management channel, including the Transmission Control Protocol (TCP) sequence number, power protocol session status, security rule matching progress, and packet cache data. When a hardware failure, link interruption, or core program abnormality is detected in an isolation vector device node, the main management and control node immediately marks the node as a faulty node, stops distributing new traffic to it, and redistributes the session being processed by the node to other normal nodes through a two-dimensional session consistency traffic splitting algorithm. Normal nodes directly take over the subsequent packet processing of the session through the pre-synchronized session status data. Normal nodes directly take over the subsequent processing through the pre-synchronized session status data, without the need for the terminal to rebuild the TCP connection or re-authenticate. The service interruption time is less than 200ms, and there is no loss of service packets during the switching process. Preferably, for Level I power grid control sessions, a dual-node hot standby mode is adopted, in which the same control session is processed synchronously on the primary and backup nodes. When the primary node fails, the backup node can seamlessly take over, achieving packet loss and zero interruption.

[0062] In the example above, the monitoring data of the status monitoring and disaster recovery scheduling unit 105 is aggregated to the main control node. When a fault occurs, the main control node immediately marks the node as a faulty node and distributes the sessions being processed by the node to other normal nodes. There is no loss of service packets during the switchover process, achieving zero packet loss and zero interruption.

[0063] Optionally, during the traffic offloading process, packets can be distributed to the corresponding priority hardware transfer queues within the isolation vector device nodes according to the service type. Level I power grid control services directly enter their dedicated queues, with hard-reserved bandwidth to ensure they are not occupied by other services. After receiving the traffic, each isolation vector device node performs packet filtering, content inspection, and one-way transfer processing according to the synchronized security policy of the uninterrupted synchronization unit 104. The processed packets are then sent to the egress switch of the cluster collaborative management and control unit 102. The backend reassembly module reassembles the processed packets from each isolation vector device node in an orderly manner and sends them to the target business system of the management information zone 107.

[0064] During the traffic offloading process, the status monitoring and disaster recovery scheduling unit 105 collects the dynamic isolation vector values ​​of each isolation vector device node in real time. When the redundancy of the isolation vector value of a certain isolation vector device node is less than 20% and the bandwidth utilization rate exceeds 90%, the traffic offloading scheduling module will automatically adjust the traffic offloading weight and prioritize the distribution of new sessions to the node with the highest isolation vector value redundancy to avoid single node overload.

[0065] In the above example, by employing a network five-tuple dual-dimensional session consistency traffic splitting algorithm and parallel processing mechanism that incorporates power protocol-aware service characteristics, the overall throughput of the cluster collaborative management unit 102 increases linearly with the number of nodes. The throughput of a cluster of multiple isolation vector device nodes can reach over 95% of that of a single node, resolving the bandwidth bottleneck in source-grid-load-storage interaction scenarios. All packets of the same session are uniformly distributed to this node for processing, ensuring the integrity and timing correctness of power service sessions. The traffic splitting weight is automatically adjusted based on the redundancy of the isolation vector value and bandwidth occupancy rate to avoid single-node overload.

[0066] As an example, the security policy non-disruptive synchronization unit 104 works in conjunction with the cluster collaborative management and control unit 102 to synchronize the existing security policies under the original boundary isolation scenario to all isolation vector device nodes, so as to ensure that the security protection rules of each isolation vector device node are consistent.

[0067] As an example, such as Figure 2As shown, the security policy non-disruptive synchronization unit 104 incorporates a policy format conversion subunit 1041, a policy consistency verification subunit 1042, and a policy atomic synchronization subunit 1043. The policy format conversion subunit 1041 is compatible with the security policy format of the power forward and reverse isolation devices, enabling the import of existing policies without modification. The policy consistency verification subunit 1042 is used to verify in real time whether the security policies of all isolation vector device nodes in the cluster are completely consistent, and automatically triggers synchronization correction when deviations occur. The policy atomic synchronization subunit 1043 ensures that policy changes take effect simultaneously throughout the entire cluster, and policy changes only need to be configured once at the global entry point to achieve synchronized effect across the entire cluster.

[0068] As an example, the security policy seamless synchronization unit 104 adopts a three-tiered architecture consisting of global basic policies, scenario-specific policies, and node-specific policies. The policy priorities between different levels are clearly defined, and a built-in automatic policy conflict resolution unit automatically detects policy conflicts between different levels and nodes before policy synchronization, automatically resolving them according to the principle of prioritizing higher-priority policies, while simultaneously generating conflict alarms and audit logs. Furthermore, policy versions are linked and bound to trusted node versions; each policy change generates a unique version with a national cryptographic signature, and one-click rollback and full lifecycle audit traceability are supported.

[0069] Specifically, a global basic policy configuration entry is deployed on the main control node. Users can import the existing security policies of the original traditional isolation devices through the global policy configuration entry on the main control node. Each isolation vector device node has a built-in policy execution unit that works in conjunction with the global policy configuration entry to achieve policy synchronization across the entire cluster.

[0070] Users can import existing security policies from traditional isolation devices through the global policy configuration entry of the main control node. The policy format conversion subunit 1041 of the security policy non-disruptive synchronization unit 104 automatically identifies and converts the policy format without requiring users to modify the original policy rules. After conversion, the policy is layered according to a three-level architecture of global basic policy, scenario-specific policy, and node personalized policy. The policy conflict automatic resolution module automatically detects policy conflicts between different levels and different nodes, and automatically resolves them according to the principle of prioritizing high-priority policies, generating conflict alarms and audit logs. After verification, the policy atomic synchronization subunit 1043 fully synchronizes the policy to all isolated vector device nodes in the cluster. At the same time, the policy consistency verification subunit 1042 verifies the policy of each isolated vector device node to ensure that the policies of all isolated vector device nodes are completely consistent. When a user needs to modify the policy, only one modification is required in the global configuration entry. The policy atomic synchronization subunit 1043 can then synchronize the changes to all isolated vector device nodes and ensure that the policies of all isolated vector device nodes take effect simultaneously, avoiding security vulnerabilities caused by policy asynchrony. Each policy change generates a unique version with a national cryptographic signature, which is bound to the trusted version of the isolated vector device node, supporting one-click rollback and full lifecycle auditing.

[0071] In the above example, the security policy non-disruptive synchronization unit 104 is compatible with the policy format of traditional isolation devices, enabling the import of existing security policies without modification and the synchronization of the entire cluster without difference. Through the three-level layered architecture and automatic conflict resolution mechanism, it adapts to the complex scenario of multiple security zones in the power industry, eliminates policy conflicts and security vulnerabilities, and policy changes only need to be configured once to take effect on the entire cluster without changing the user's original security protection rules, thus avoiding the security risks brought about by expansion.

[0072] As an example, the status monitoring and disaster recovery scheduling unit 105 is equipped with monitoring probes deployed on each isolation vector device node. It collects operational data such as dynamic isolation vector values, CPU utilization, memory usage, throughput, link connectivity, and isolation rule matching success rate of each isolation vector device node at a period of 10ms. The monitoring data is aggregated to the main management node in real time, and the main management node performs load balancing and disaster recovery scheduling.

[0073] As an example, the status monitoring and disaster recovery scheduling unit 105 uses the dynamic isolation vector values ​​generated by each isolation vector device node through the power-specific isolation vector quantization model as the core scheduling basis to perform load balancing and disaster recovery switching.

[0074] As an example, the horizontal performance expansion system of the isolation vector device also includes a cross-platform secure communication component deployed in the cluster collaborative management and control unit 102 and each access service terminal. This component uploads the access behavior data and security status data of the access service terminals to the power system boundary access trust evaluation and risk control system in real time. It receives and executes risk control policies issued by the system, restricting or blocking access permissions for terminals with risks, thus achieving dynamic risk management of the access terminals. Optionally, the access service terminals include new energy power plant monitoring terminals, energy storage management systems (EMS) systems, virtual power plant dispatch terminals, and power grid dispatch service terminals, enabling terminal authentication and access control.

[0075] As an example, the cross-platform secure communication component includes a national cryptographic authentication module, a credential acquisition module, and a terminal identity verification module, among which:

[0076] The national cryptographic authentication module is used to achieve two-way identity authentication and communication link encryption between cluster collaborative management and control units 102 and between cluster collaborative management and control units 102 and access service terminals.

[0077] The credential collection module is used to collect device fingerprints, operating system integrity, security patch status, and access behavior characteristics of access service terminals.

[0078] The terminal identity verification module is used to implement integrity verification and authorized access control of access service terminals, and only access service terminals that pass the verification are allowed to access the cluster collaborative management and control unit 102.

[0079] As an example, each isolation vector device node is equipped with a built-in national cryptographic SM2 / SM3 / SM4 encryption component to achieve encryption and integrity verification of data transmission, which complies with the cryptographic application specifications of the power system.

[0080] Optionally, all cluster collaborative management and control unit 102 control signaling uses the national cryptographic standard SM4 encryption, SM3 integrity verification, and SM2 signature verification throughout the entire process.

[0081] As an example, the cluster collaborative management unit 102 supports hot-swappable expansion of isolated vector device nodes. After a new isolated vector device node is powered on, it must pass dual trusted verification based on the national cryptographic system, including national cryptographic SM2 two-way identity authentication and full integrity verification of the trusted computing module. Only terminals that pass the integrity verification and identity authentication can access the cluster collaborative management unit 102 according to the preset authorization rules. After the verification is passed, the cluster registration, configuration synchronization and isolated vector value benchmark test are automatically completed. After reaching the minimum threshold of the cluster, it is automatically included in the resource pool without interrupting the original business operation.

[0082] As an example, when a source-network-load-storage business terminal accesses the cluster system, the credential collection module of the cross-platform secure communication component automatically collects information such as the terminal's device fingerprint, operating system version, security patch status, and access behavior characteristics. The terminal identity verification module verifies the integrity of the terminal and simultaneously implements two-way identity authentication between the terminal and the cluster through the national cryptographic SM2 algorithm. Only terminals that pass the integrity verification and identity authentication can access the cluster system according to the preset authorization rules.

[0083] Meanwhile, the cross-platform secure communication component uploads terminal access behavior data and security status data to the power system boundary access trust evaluation and risk control system in real time, receives and executes risk control policies issued by the system, and performs access restriction or blocking operations on terminals with risks, thereby realizing dynamic risk control of access terminals.

[0084] The above example employs a primary-multiple-standby cluster management architecture, supporting hot-swappable node expansion. Newly added nodes can be automatically managed without interrupting services. Through real-time session state synchronization and lossless migration mechanisms, seamless switching of faulty nodes is achieved, with service interruption time <200ms. Zero packet loss and zero interruption are achieved for control-related services, improving the availability and reliability of the boundary isolation system and meeting the continuous operation requirements of high-frequency interaction scenarios involving power grid, load, and storage. Based on dual node access verification using national cryptographic standards, a trusted management and control system for the entire node lifecycle is constructed. A hardware-level multi-queue bandwidth hard isolation mechanism reserves dedicated bandwidth for power grid control services, preventing large-volume data from crowding out core business resources.

[0085] Optionally, the usage process for a system that laterally scales up the performance of an isolated vector device includes:

[0086] Preparation before deployment: Before deploying the horizontal performance expansion system of the isolation vector device, those skilled in the art should sort out and divide the security policy, source-grid-load-storage business traffic model, access terminal list, and power business priority of the original horizontal isolation device, confirm the number of nodes and performance requirements of the cluster, and complete the network planning, isolation vector value weight configuration, and bandwidth reservation ratio configuration for the cluster deployment.

[0087] Cluster initialization deployment: By mounting the hardware of the isolation vector device node and connecting it, configuring the basic parameters of the main control node, importing the existing security policies, and completing the initial deployment of the cluster collaborative control unit 102 and the configuration of the three-level policy architecture.

[0088] Node management and expansion: Power on the remaining isolated vector device nodes, complete automatic registration, dual trusted verification, isolated vector value benchmark testing and management, expand the number of cluster collaborative management and control units 102 as needed, and achieve smooth performance improvement.

[0089] Service Deployment and Traffic Switching: After completing the functional testing, performance testing, security policy verification, and service priority testing of the cluster collaborative management unit 102, the source-network-load-storage service traffic will be gradually switched to the cluster collaborative management unit 102 to complete the deployment and deployment.

[0090] Routine Operation and Management: Through the global management interface of the main management node, users can view the operating status, isolation vector value distribution, performance data, and security events of the cluster collaborative management unit 102, and perform maintenance operations such as policy changes, node expansion, and firmware upgrades. The system automatically completes the configuration synchronization and status monitoring of the entire cluster collaborative management unit 102. Optionally, the full-link audit and traceability module can be used to complete the full-process traceability query of business traffic, meeting the compliance and security audit requirements of the power industry.

[0091] The isolation vector device horizontal performance extension system of this application can be deployed at the horizontal boundary between the production control area and the management information area of ​​the provincial power grid dispatch center in the new power system to realize the secure isolation between the two areas and the large-volume data exchange of source, grid, load and storage businesses.

[0092] The aforementioned horizontal performance expansion system for isolation vector devices includes an isolation vector device cluster unit 101, comprising at least two isolation vector device nodes with dedicated power horizontal isolation capabilities; a cluster collaborative management unit 102, communicatively connected to the isolation vector device nodes, used for lifecycle management of cluster nodes, dynamic topology discovery, management of extended nodes, and global coordination of cluster operation status; a data distribution and parallel processing unit 103, located between the cluster collaborative management unit 102 and each isolation vector device node, used to distribute the power big data exchange traffic to be processed to the isolation vector device nodes for parallel isolation forwarding processing, and to complete the orderly reassembly of processed packets; a security policy non-disruptive synchronization unit 104, connected to the cluster collaborative management unit 102, used to synchronize existing security policies under the original boundary isolation scenario to the isolation vector device nodes; and a status monitoring and tolerance unit. The disaster dispatching unit 105 is communicatively connected to the isolation vector device cluster unit 101 and the cluster collaborative management and control unit 102. It is used to collect the operating data of each isolation vector device node and perform load balancing scheduling and traffic switching of faulty nodes within the cluster. In this way, it overcomes the problems of traditional technologies, such as the difficulty of clustered scheduling to fully adapt to the compliance requirements of power secondary system security protection and the easy occurrence of session timing disorder in mixed transmission scenarios. It takes into account high security, high reliability and business adaptability, meets the application requirements of the new power system source-grid-load-storage interaction scenario, and is compatible with the deployment architecture, security policies and business processes of existing power system horizontal isolation devices. It can be deployed and launched without modifying existing business systems or interrupting existing business, and does not require adaptation and modification of existing business terminals, thus reducing the upgrade and transformation cost and deployment cycle of the new power system horizontal security protection system.

[0093] It should be noted that all electrical devices involved in this application can be powered by a storage battery or an external power source, and this application does not make any specific limitations.

[0094] Those skilled in the art will understand that Figure 1 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0095] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0096] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A system for laterally extending the performance of an isolated vector device, characterized in that, include: The isolated vector device cluster unit, the cluster collaborative management and control unit, the data diversion and parallel processing unit, the security policy non-disruptive synchronization unit, and the status monitoring and disaster recovery scheduling unit are all included. The isolation vector device cluster unit includes at least two isolation vector device nodes with dedicated power lateral isolation capabilities; The cluster collaborative management and control unit is communicatively connected to the isolation vector device node, and is used to realize the life cycle management of cluster nodes, dynamic topology discovery, expansion node management, and global coordination of cluster operation status. The data splitting and parallel processing unit is located between the cluster collaborative management and control unit and each of the isolation vector device nodes. It is used to distribute the power big data exchange traffic to be processed to the isolation vector device nodes to perform isolation forwarding processing in parallel, and to complete the orderly reassembly of the processed packets. The security policy non-disruptive synchronization unit is connected to the cluster collaborative management and control unit and is used to synchronize the existing security policies under the original boundary isolation scenario to the isolation vector device node. The status monitoring and disaster recovery scheduling unit is communicatively connected to the isolation vector device cluster unit and the cluster collaborative management and control unit, and is used to collect the operating data of each isolation vector device node and perform load balancing scheduling and traffic switching of faulty nodes within the cluster.

2. The system according to claim 1, characterized in that, The isolation vector device node is also equipped with a standardized data interaction interface and a dual-host dual-ferry isolation hardware architecture; the standardized data interaction interface is used to build a boundary isolation resource pool; the dual-host dual-ferry isolation hardware architecture includes an inner host and an outer host, and a physically isolated one-way ferry channel for business data and a one-way ferry channel for cluster control are set between the inner host and the outer host.

3. The system according to claim 2, characterized in that, The cluster collaborative management and control unit is embedded in the dual-host dual-ferry isolation hardware architecture of each isolation vector device node. The cluster collaborative management and control unit has a built-in power-specific isolation vector quantization model to realize the quantitative scheduling of global isolation resources of the cluster.

4. The system according to claim 3, characterized in that, The data offloading and parallel processing unit incorporates a power protocol-aware service characteristic and network 5-tuple dual-dimensional session consistency offloading algorithm to perform power service session-level consistent traffic distribution; and / or The status monitoring and disaster recovery scheduling unit uses the dynamic isolation vector values ​​generated by each isolation vector device node through the power-specific isolation vector quantization model as the core scheduling basis to perform load balancing and disaster recovery switching.

5. The system according to claim 3, characterized in that, The cluster collaborative management and control unit adopts a redundant architecture with one master and multiple backups, including a master management node and backup management nodes. The master control node is used for the global configuration and scheduling of the cluster collaborative control unit; The backup management node is used to synchronize the configuration data and operating status of the primary management node in real time, and to seamlessly switch over to take over management authority when the primary management node fails.

6. The system according to claim 1, characterized in that, The security policy non-disruptive synchronization unit incorporates a policy format conversion subunit, a policy consistency verification subunit, and a policy atomic synchronization subunit. The policy format conversion subunit is compatible with the security policy format of power forward and reverse isolation devices, enabling the import of existing policies without modification. The policy consistency verification subunit is used to verify in real time whether the security policies of all the isolation vector device nodes in the cluster are completely consistent, and automatically triggers synchronization correction when deviations occur. The policy atomic synchronization subunit ensures that policy changes take effect simultaneously throughout the entire cluster, and policy changes only need to be configured once at the global entry point to achieve synchronized effect across the entire cluster.

7. The system according to claim 1, characterized in that, The operational data collected by the status monitoring and disaster recovery scheduling unit for each isolation vector device node includes: CPU utilization, memory usage, throughput, bandwidth usage, number of session connections, link connectivity status, and the success rate of matching isolation forwarding rules.

8. The system according to claim 1, characterized in that, Each of the aforementioned isolation vector device nodes has a built-in dedicated isolation data exchange chip, a trusted computing module, and a national cryptographic encryption component. The trusted computing module is used to realize trusted startup, version integrity verification, and trusted supply chain management of the isolation vector device nodes. The dedicated isolation data exchange chip is equipped with multiple physically isolated hardware ferry queues of different priorities, which correspond one-to-one with the power business priorities. The bandwidth of the hardware ferry queues is hard isolated and does not crowd out each other. The cluster collaborative management and control unit establishes a global resource pool for hardware ferry queues of different priorities and hard-reserves the preset total cluster bandwidth for the highest priority hardware ferry queue.

9. The system according to claim 1, characterized in that, The security policy non-disruptive synchronization unit adopts a three-tiered architecture consisting of a global basic policy, a scenario-specific policy, and a node-specific policy.

10. The system according to claim 1, characterized in that, It also includes a cross-platform secure communication component, deployed between the cluster collaborative management and control unit and each access service terminal. The cross-platform secure communication component includes: The national cryptographic authentication module is used to realize two-way identity authentication and communication link encryption between the cluster collaborative management and control units and between the cluster collaborative management and control units and the access service terminals. The credential collection module is used to collect the device fingerprint, operating system integrity, security patch status, and access behavior characteristics of the access service terminal. The terminal identity verification module is used to implement integrity verification and authorized access control of the access service terminal, and only allows access service terminals that pass the verification to access the cluster collaborative management and control unit.

Images