Networking configuration method, device, equipment, system, medium and product of nodes

Abstract

The present disclosure provides a networking configuration method, device, equipment, system, medium and product of a node. The method is applied to a scheduling center device, comprising: receiving work order information, the work order information at least including first work information of a supplier and second work information of a demander; extracting the first work information and the second work information to obtain node information; wherein the node information represents a privacy computing node of the supplier and a privacy computing node of the demander; determining a node networking architecture corresponding to the node information according to the number of nodes in the node information; wherein the node networking architecture is used to establish network connection for the privacy computing node of the supplier and the privacy computing node of the demander. The network connection between the privacy computing nodes is realized, the cost of multi-node interaction is reduced, and the efficiency of multi-node interaction is improved.

Description

Technical Field

[0001] This application relates to federated learning technology, specifically to a method, apparatus, device, system, medium, and product for configuring nodes in a network. Background Technology

[0002] In related technologies, privacy computing node management solutions adopt a point-to-point cooperation model, where the two parties communicate the node identity document (ID) and node information offline through methods such as telephone and email. However, this approach has the disadvantage of high cost. For example, in scenarios where the first task is executed by the first three nodes and the second task is executed by the two nodes, point-to-point cooperation is costly and inefficient. Summary of the Invention

[0003] In view of this, embodiments of this application provide a method, apparatus, device, system, medium, and product for configuring nodes in a network, aiming to effectively reduce the cost of multi-node interaction and improve the efficiency of multi-node interaction.

[0004] The technical solution of this application embodiment is implemented as follows:

[0005] This application provides a method for configuring a node's network topology, applied to a dispatch center device. The method includes:

[0006] Receive work order information, wherein the work order information includes at least the supplier's first work information and the customer's second work information;

[0007] The first working information and the second working information are extracted to obtain node information; wherein, the node information represents the privacy computing node of the supplier and the privacy computing node of the demander;

[0008] Based on the number of nodes in the node information, the node networking architecture corresponding to the node information is determined; wherein, the node networking architecture is used to establish network connections between the privacy computing nodes of the supplier and the privacy computing nodes of the demander.

[0009] In the above scheme, the node networking architecture includes a first networking architecture, a second networking architecture, and a third networking architecture; determining the node networking architecture corresponding to the node information based on the number of nodes in the node information includes:

[0010] If the number of nodes is greater than the first number threshold, then the first network architecture corresponding to the node information is determined;

[0011] If the number of nodes is greater than a second number threshold and less than a first number threshold, then the second network architecture corresponding to the node information is determined; the second number threshold is less than the first number threshold.

[0012] If the number of nodes is less than the second number threshold, then the third network architecture corresponding to the node information is determined.

[0013] In the above scheme, after determining the node network architecture corresponding to the node information based on the number of nodes in the node information, the method further includes:

[0014] If it is determined that the training information of the first node meets the exchange conditions, then the first encryption parameter of the first node is sent to the second node through the node networking architecture, and the second encryption parameter of the second node is sent to the first node through the node networking architecture.

[0015] Wherein, the first node is a local computing node; if the first node is the privacy computing node of the supplier, the second node is the privacy computing node of the demand party corresponding to the first node; if the first node is the privacy computing node of the demand party, the second node is the privacy computing node of the supplier corresponding to the first node.

[0016] In the above scheme, the method further includes:

[0017] Obtain the first time parameter corresponding to the node networking architecture; the first time parameter represents the duration of sending the second encrypted parameter of the second node to the first node;

[0018] If the first time parameter is greater than or equal to the time threshold, the received second encrypted parameter is processed based on the first node to obtain an update parameter; the update parameter is used to update the modeling model corresponding to the first node.

[0019] In the above scheme, the method further includes:

[0020] Obtain the interaction parameters between the first node and the second node;

[0021] If the interaction parameter is greater than the interaction threshold, then the second node corresponding to the interaction parameter is detected to obtain the detection result;

[0022] If the detection result indicates that the second node is in an abnormal state, then the second node is removed.

[0023] In the above scheme, the method further includes:

[0024] If the first time parameter is greater than or equal to the time threshold, obtain the second node corresponding to the first time parameter;

[0025] The second node corresponding to the first time parameter is detected to obtain the detection result;

[0026] If the detection result indicates that the second node is in an abnormal state, then the second node is removed.

[0027] In the above scheme, the method further includes:

[0028] If the second time parameter corresponding to the node networking architecture meets the time condition, then the privacy computing node of the supplier and the privacy computing node of the demander are controlled to disconnect from the network; the second time parameter represents the duration for which the privacy computing node of the supplier and the privacy computing node of the demander are controlled to establish a network connection based on the node networking architecture.

[0029] This application provides a node networking configuration device, applied to a dispatch center device, comprising:

[0030] The receiving module is used to receive work order information, which includes at least the supplier's first work information and the demander's second work information;

[0031] An extraction module is used to extract the first working information and the second working information to obtain node information; wherein the node information represents the privacy computing node of the supplier and the privacy computing node of the demander;

[0032] The first determining module is used to determine the node networking architecture corresponding to the node information based on the number of nodes in the node information; wherein, the node networking architecture is used to establish network connections between the privacy computing nodes of the supplier and the privacy computing nodes of the demander.

[0033] This application provides a scheduling center device, including: a processor and a memory for storing computer programs capable of running on the processor, wherein...

[0034] The processor, when running a computer program, performs the steps of the method according to any one of claims 1 to 8.

[0035] This application provides a node networking configuration system, including a data sharing device, the aforementioned scheduling center device, and node devices; wherein,

[0036] The data sharing device is used for:

[0037] Generate work order information, which includes at least the supplier's first work information and the demander's second work information; determine the dispatch center device based on the first scheduling information in the first work information and the second scheduling information in the second work information; and send the work order information to the dispatch center device.

[0038] The node device is used to control the privacy computing node to obtain training information.

[0039] In the above scheme, the data sharing device determines the scheduling center device based on the first scheduling information in the first working information and the second scheduling information in the second working information, including:

[0040] If it is determined that the first scheduling information and the second scheduling information are consistent, then the first scheduling device corresponding to the first scheduling information or the second scheduling device corresponding to the second scheduling information is determined as the scheduling center device;

[0041] If it is determined that the first scheduling information and the second scheduling information are inconsistent, then a third scheduling device is determined and the third scheduling device is determined as the scheduling center device, wherein the third scheduling center is connected to the first scheduling device and the second scheduling device respectively.

[0042] This application provides a storage medium storing a computer program, which, when executed by a processor, implements the steps of the network configuration method for the node.

[0043] This application provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the network configuration method for the node.

[0044] This application provides a method, apparatus, device, system, medium, and product for configuring nodes in a network. The method is applied to a dispatch center device and includes: receiving work order information, the work order information including at least a first work information of a supplier and a second work information of a demander; extracting node information from the first work information and the second work information to obtain node information; wherein the node information represents the privacy computing nodes of the supplier and the privacy computing nodes of the demander; determining the node network architecture corresponding to the node information based on the number of nodes in the node information; wherein the node network architecture is used to establish network connections between the privacy computing nodes of the supplier and the privacy computing nodes of the demander. By adopting the technical solution of this application, node information is determined through the first work information of the supplier and the second work information of the demander included in the work order information, and the node network architecture is determined through the number of nodes in the node information, thereby realizing network connections between privacy computing nodes, reducing the cost of multi-node interaction, and improving the efficiency of multi-node interaction. Attached Figure Description

[0045] Figure 1 This is a schematic diagram illustrating the implementation process of the network configuration method for nodes in an embodiment of this application.

[0046] Figure 2 This is a schematic diagram of the composition structure of the networking configuration device for nodes in an embodiment of this application;

[0047] Figure 3 This is a schematic diagram of a hardware entity structure of a networking configuration device for a node in an embodiment of this application;

[0048] Figure 4 This is a schematic diagram of the network configuration system for nodes in an embodiment of this application;

[0049] Figure 5 This is a schematic diagram illustrating the implementation process of the network configuration method for nodes in the application example of Embodiment 1 of this application.

[0050] Figure 6 This is a schematic diagram of a hybrid topology architecture in an application example of Embodiment 1 of this application;

[0051] Figure 7 This is a schematic diagram of a mesh topology architecture in an application example of Embodiment 1 of this application. Detailed Implementation

[0052] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of this application.

[0053] In the federated learning training process of related technology one: During task execution, each node trains its local model using local data. After reaching the iterative condition of exchanging model parameters with other nodes, it sends its encrypted local parameters to the collaborating nodes and simultaneously receives encrypted parameters from the collaborating nodes. The local model is then updated after the local and collaborating parameters are federated and averaged.

[0054] The drawback of the aforementioned related technology is that if a node exits or the network is poor, the task cannot be executed.

[0055] The privacy computing node management solution in related technology 2 adopts a point-to-point cooperation model. The node ID and node information are communicated offline by both parties through telephone, email and other means. When connecting to the network, the network connection cannot be controlled according to the cooperation period and needs to be manually disconnected.

[0056] The disadvantages of the aforementioned related technology 2 include: high cost. For example, in a scenario where the first task is executed by the first three nodes and the second task is executed by the two nodes, the point-to-point cost is high and the efficiency is low.

[0057] It is understandable that when the demand side selects data from partners on the trading platform, it is not a one-off transaction. There may be inconsistencies in the data used by nodes in each federated learning task, inconsistencies in the number of partner nodes, node exits during the collaboration process, or abnormal node status. In related technologies, data circulation is carried out using a point-to-point project-based model, which cannot dynamically adjust the network architecture according to actual needs, and issues such as node exits during operation may arise.

[0058] This application provides a method for configuring a node's network topology, applied to a dispatch center device, such as... Figure 1 As shown, the method includes:

[0059] Step 101: Receive work order information. The work order information includes at least the supplier's first work information and the customer's second work information.

[0060] For example, the dispatch center device can be connected to the data sharing device, and the dispatch center device can receive work order information sent by the data sharing device.

[0061] For example, the first working information may be the supplier's scheduling center name information, the supplier's scheduling center address information, the privacy computing node ID information, node address information, node port information, and node number information participating in the task.

[0062] For example, the second work information may include work order name information, demander scheduling center name information, demander scheduling center address information, demander privacy computing node ID information, node address information, node port information, number of nodes information, computing type information (federated learning, hidden query, joint statistics, privacy intersection), resource name information, cooperation time information, etc.

[0063] Step 102: Extract the first working information and the second working information to obtain node information; wherein, the node information represents the privacy computing nodes of the supplier and the privacy computing nodes of the demander.

[0064] For example, the dispatch center device may include a delivery center unit, which is used to receive work order information and extract node information from the work order information. The node information includes information such as the node name, node address, node port, and number of nodes of the privacy computing node.

[0065] In this embodiment, nodes can be managed in a networked manner. The cooperating nodes for each task are confirmed through work order information, and the networking method is selected based on the number of local nodes and cooperating nodes.

[0066] Step 103: Determine the node networking architecture corresponding to the node information based on the number of nodes in the node information; wherein, the node networking architecture is used to establish network connections between the privacy computing nodes of the supplier and the privacy computing nodes of the demander.

[0067] For example, the number of nodes can be determined within a certain range, and the node networking architecture corresponding to the node information can be determined based on the preset relationship and the number range. It can be understood that the preset relationship represents that different number ranges correspond to different node networking architectures.

[0068] In one application example, the node networking architecture includes a first networking architecture, a second networking architecture, and a third networking architecture; based on the number of nodes in the node information, the node networking architecture corresponding to the node information is determined, including:

[0069] If the number of nodes is greater than the first threshold, then the first network architecture corresponding to the node information is determined.

[0070] If the number of nodes is greater than the second threshold and less than the first threshold, then the second network architecture corresponding to the node information is determined; the second threshold is less than the first threshold.

[0071] If the number of nodes is less than the second threshold, then the third network architecture corresponding to the node information is determined.

[0072] For example, the first quantity threshold can be a relatively large threshold b, and the first network architecture can be determined according to the actual situation, without limitation here. As an example, the first network architecture can be a hybrid topology architecture. If the number of nodes is greater than the threshold b, then the network architecture of the node information is determined to be a hybrid topology architecture. For the hybrid topology architecture, the nodes are grouped, and a stack node is selected from each group, with the scheduling center device serving as the central coordination node. In each iteration, the stack nodes merge the sum of their parameters and send it to the central coordination node. The central coordination node calculates the average and feeds it back to the stack nodes, as well as to other nodes through the stack nodes.

[0073] For example, the second quantity threshold can be a threshold a that is less than threshold b. The second network architecture can be determined according to the actual situation and is not limited here. As an example, the second network architecture can be a star network architecture. If the number of nodes is greater than threshold a but less than threshold b, then the network architecture of the node information is determined to be a star network architecture. For the star architecture, the scheduling center device is selected as the coordination node.

[0074] For example, the third network architecture can be determined based on the actual situation and is not limited here. As an example, the third network architecture can be a mesh topology. If the number of nodes is less than a threshold 'a', then the network architecture for the node information is determined to be a mesh topology. For the mesh topology, in each iteration, the coordinating nodes for the next round can be determined based on the collected node network quality information.

[0075] In this embodiment, the network architecture can be selected based on the number of nodes cooperating in each task, and the central node of each modeling task can be determined based on the network architecture.

[0076] In one application example, after determining the node network architecture corresponding to the node information based on the number of nodes in the node information, the method further includes:

[0077] If it is determined that the training information of the first node meets the exchange conditions, then the first encrypted parameter of the first node is sent to the second node through the node networking architecture, and the second encrypted parameter of the second node is sent to the first node through the node networking architecture.

[0078] Wherein, the first node is a local computing node; if the first node is the supplier's privacy computing node, the second node is the demander's privacy computing node corresponding to the first node; if the first node is the demander's privacy computing node, the second node is the supplier's privacy computing node corresponding to the first node.

[0079] For example, the second node can be a collaborative computing node corresponding to a local computing node. It is understood that if the supplier's privacy computing node is a local computing node, then the demander's privacy computing node corresponding to the supplier's privacy computing node is a collaborative computing node; if the demander's privacy computing node is a local computing node, then the supplier's privacy computing node corresponding to the demander's privacy computing node is a collaborative computing node.

[0080] For example, the training information of the first node can be the training parameters obtained by training a local model using local data during the modeling task. The exchange condition can be the iterative condition for the first node to exchange model parameters with other nodes. The first encrypted parameter can be the parameter obtained after encrypting the training parameters.

[0081] In one application example, the method also includes:

[0082] Obtain the first time parameters corresponding to the node network architecture; the first time parameters represent the duration for the second encrypted parameters of the second node to be sent to the first node;

[0083] If the first time parameter is greater than or equal to the time threshold, the received second encrypted parameter is processed based on the first node to obtain the update parameter; the update parameter is used to update the modeling model corresponding to the first node.

[0084] For example, the time threshold can be determined according to the actual situation, and is not limited here. As an example, the time threshold can be a preset reception duration for the first node to receive the second encrypted parameter sent by the second node, and the time threshold can be less than or equal to the preset reception duration t1.

[0085] For example, the updated parameters are obtained by processing the received second encrypted parameters based on the first node. This can be achieved by performing a federated average of the local training parameters and the second encrypted parameters from the collaborator. The modeling model can be the local model corresponding to the first node. It should be noted that if the first time parameter is greater than or equal to the time threshold, the first node will no longer receive the second encrypted parameters and will only process the second encrypted parameters that have already been received.

[0086] In one application example, the method also includes:

[0087] Get the interaction parameters between the first node and the second node;

[0088] If the interaction parameter is greater than the interaction threshold, then the second node corresponding to the interaction parameter is detected to obtain the detection result;

[0089] If the detection result indicates that the second node is in an abnormal state, then the second node is removed.

[0090] For example, the process of obtaining interaction parameters can be determined according to the actual situation and is not limited here. As an example, interaction parameters can be obtained by processing the interaction information between the first node and the second node, where the interaction parameters can be distance parameters. For example, the interaction threshold can be a distance threshold T. For example, obtaining interaction parameters by processing interaction information can involve obtaining first interaction information when the first node and the second node perform a first round of interaction; obtaining second interaction information when the first node and the second node perform a second round of interaction; and determining the distance parameter between the first interaction information and the second interaction information.

[0091] For example, the interaction information includes node name information, central processing unit (CPU) information, memory information, bandwidth information between nodes, latency information between nodes, jitter information between nodes, packet loss information between nodes, local node sending parameter time information, and cooperative node receiving parameter time information, etc.

[0092] For example, the interaction information also includes the name information of the cooperative node that did not receive the parameter; the second node corresponding to the interaction parameter can be the aforementioned cooperative node that did not receive the parameter; the second node corresponding to the interaction parameter is detected, and the detection result is obtained, which can be to obtain the heartbeat status of the second node corresponding to the interaction parameter, where the heartbeat status is the node interconnection status, and to determine whether the heartbeat status of the second node is abnormal.

[0093] In one application example, the method also includes:

[0094] If the first time parameter is greater than or equal to the time threshold, obtain the second node corresponding to the first time parameter;

[0095] The second node corresponding to the first time parameter is detected to obtain the detection result;

[0096] If the detection result indicates that the second node is in an abnormal state, then the second node is removed.

[0097] Understandably, if the first time parameter is greater than or equal to the time threshold, and the second node fails to successfully send its second encrypted parameter to the first node, then the heartbeat status of the second node is obtained. The heartbeat status is the node interconnection status, and it is determined whether there is an anomaly in the heartbeat status of the second node.

[0098] In this embodiment, compared with the federated modeling method in related technologies, a node anomaly monitoring strategy is added. Specifically, this strategy involves adding a timed task unit and a task monitor unit. During each round of aggregating local and cooperative parameters, based on receiving parameters only from cooperative nodes within a specified range, and considering the network status of the received data in this round and the names of unmerged nodes, the node status and abnormal nodes are managed uniformly.

[0099] In one application example, the method also includes:

[0100] If the second time parameter corresponding to the node networking architecture meets the time condition, the privacy computing nodes of the supplier and the privacy computing nodes of the demand side will disconnect from the network. The second time parameter represents the duration for establishing a network connection between the privacy computing nodes of the supplier and the privacy computing nodes of the demand side based on the node networking architecture.

[0101] For example, the time condition characterizes the time taken for privacy computing nodes to establish network connections in the node networking architecture to exceed the cooperation duration; wherein, the cooperation duration can be determined based on the cooperation time information in the second work information of the work order information, and the cooperation time information can characterize the cooperation duration. In this embodiment of the application, after confirming the networking architecture, a timed node connection policy is configured, and the node automatically disconnects after the cooperation duration is exceeded.

[0102] This application provides a node networking configuration device, applied to dispatch center equipment, such as... Figure 2 As shown, the network configuration device 200 for this node includes: a receiving module 201, an extraction module 202, and a first determining module 203; wherein,

[0103] The receiving module 201 is used to receive work order information, which includes at least the first work information of the supplier and the second work information of the demander;

[0104] Extraction module 202 is used to extract the first working information and the second working information to obtain node information; wherein, the node information represents the privacy computing node of the supplier and the privacy computing node of the demander;

[0105] The first determining module 203 is used to determine the node networking architecture corresponding to the node information based on the number of nodes in the node information; wherein the node networking architecture is used to establish network connections between the privacy computing nodes of the supplier and the privacy computing nodes of the demander.

[0106] In some embodiments, the node networking architecture includes a first networking architecture, a second networking architecture, and a third networking architecture; the first determining module 203 is further configured to: if the number of nodes is greater than a first quantity threshold, determine the first networking architecture corresponding to the node information; if the number of nodes is greater than a second quantity threshold and less than the first quantity threshold, determine the second networking architecture corresponding to the node information; the second quantity threshold is less than the first quantity threshold; if the number of nodes is less than the second quantity threshold, determine the third networking architecture corresponding to the node information.

[0107] In some embodiments, the first determining module 203 is further configured to, if it is determined that the training information of the first node satisfies the exchange conditions, send the first encrypted parameter of the first node to the second node through the node networking architecture, and send the second encrypted parameter of the second node to the first node through the node networking architecture; wherein, the first node is a local computing node, if the first node is the privacy computing node of the supplier, the second node is the privacy computing node of the demand party corresponding to the first node, and if the first node is the privacy computing node of the demand party, the second node is the privacy computing node of the supplier corresponding to the first node.

[0108] In some embodiments, the network configuration device 200 for the node further includes: a first acquisition module and a processing module; wherein,

[0109] The first acquisition module is used to acquire the first time parameter corresponding to the node networking architecture; the first time parameter represents the duration of sending the second encryption parameter of the second node to the first node;

[0110] The processing module is configured to process the received second encrypted parameter based on the first node to obtain an update parameter if the first time parameter is greater than or equal to a time threshold; the update parameter is used to update the modeling model corresponding to the first node.

[0111] In some embodiments, the network configuration device 200 for the node further includes: a second acquisition module, a first detection module, and a first rejection module; wherein,

[0112] The second acquisition module is used to acquire the interaction parameters between the first node and the second node;

[0113] The first detection module is used to detect the second node corresponding to the interaction parameter if the interaction parameter is greater than the interaction threshold, and obtain the detection result.

[0114] The elimination module is used to eliminate the second node if the detection result indicates that the second node is in an abnormal state.

[0115] In some embodiments, the network configuration device 200 for the node further includes: a third acquisition module, a second detection module, and a second rejection module; wherein,

[0116] The third acquisition module is used to acquire the second node corresponding to the first time parameter if the first time parameter is greater than or equal to the time threshold.

[0117] The second detection module is used to detect the second node corresponding to the first time parameter and obtain the detection result;

[0118] The second elimination module is used to eliminate the second node if the detection result indicates that the second node is in an abnormal state.

[0119] In some embodiments, the network configuration device 200 of the node further includes: a control module, configured to control the privacy computing node of the supplier and the privacy computing node of the demander to disconnect the network connection if the second time parameter corresponding to the node network architecture meets the time condition; the second time parameter represents the duration for which the privacy computing node of the supplier and the privacy computing node of the demander establish a network connection based on the node network architecture.

[0120] It should be noted that the node networking configuration device provided in the above embodiments is only illustrated by the division of the above program modules. In actual applications, the above processing can be assigned to different program modules as needed, that is, the internal structure of the device can be divided into different program modules to complete all or part of the processing described above. In addition, the node networking configuration device provided in the above embodiments and the aforementioned node networking configuration method embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0121] Based on the hardware implementation of the above program modules, and in order to implement the method of the embodiments of this application, the embodiments of this application also provide a networking configuration device for nodes. Figure 3 This only shows an exemplary structure of the network configuration device for this node, not the entire structure. Implementation can be carried out as needed. Figure 3 The structure shown may be part or all of the structure.

[0122] like Figure 3As shown, the node networking configuration device 300 provided in this application embodiment includes: at least one processor 301, a memory 302, and a user interface 303. The various components in the node networking configuration device 300 are coupled together through a bus system 304. It can be understood that the bus system 304 is used to implement communication between these components. In addition to a data bus, the bus system 304 also includes a power bus, a control bus, and a status signal bus. However, for clarity, in... Figure 3 The general designated all buses as Bus System 304.

[0123] The user interface 303 may include a monitor, keyboard, mouse, trackball, click wheel, buttons, touchpad, or touch screen.

[0124] The memory 302 in this embodiment is used to store various types of data to support the operation of the control device. Examples of such data include any computer program used to operate on the control device.

[0125] The node networking configuration method disclosed in this application can be applied to or implemented by the processor 301. The processor 301 may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the node networking configuration method can be completed by the integrated logic circuitry in the hardware of the processor 301 or by instructions in software form. The processor 301 can be a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor 301 can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor can be a microprocessor or any conventional processor, etc. The steps of the method disclosed in the embodiments of this application can be directly manifested as execution by a hardware decoding processor, or execution by a combination of hardware and software modules in the decoding processor. The software modules can be located in a storage medium, specifically in memory 302. The processor 301 reads information from memory 302 and, in conjunction with its hardware, completes the steps of the node networking configuration method provided in the embodiments of this application.

[0126] In an exemplary embodiment, the networking configuration device of a node may be implemented by one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), general-purpose processors, controllers, microcontrollers (MCUs), microprocessors, or other electronic components to perform the aforementioned method.

[0127] It is understood that memory 302 can be volatile memory or non-volatile memory, or both. Non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), ferromagnetic random access memory (FRAM), flash memory, magnetic surface memory, optical disc, or compact disc read-only memory (CD-ROM); magnetic surface memory can be disk storage or magnetic tape storage. Volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Synchronous Static Random Access Memory (SSRAM), Dynamic Random Access Memory (DRAM), Synchronous Dynamic Random Access Memory (SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDRSDRAM), Enhanced Synchronous Dynamic Random Access Memory (ESDRAM), SyncLink Dynamic Random Access Memory (SLDRAM), and Direct Rambus Random Access Memory (DRRAM).The memories described in the embodiments of this application are intended to include, but are not limited to, these and any other suitable types of memories.

[0128] This application provides a network configuration system for nodes, such as... Figure 4 As shown, the network configuration system of this node includes a data sharing device 401, the aforementioned dispatch center device 402, and a node device 403; wherein,

[0129] Data sharing device 401 is used for:

[0130] Generate work order information, which includes at least the supplier's first work information and the demander's second work information; determine the dispatch center equipment based on the first scheduling information in the first work information and the second scheduling information in the second work information; and send the work order information to the dispatch center equipment.

[0131] Node device 403 is used to control the privacy computing node to obtain training information.

[0132] For example, the data sharing device 401 includes a work order management unit, which is used to generate work orders based on first work information and second work information. The first scheduling information may be information about a first scheduling device of the supplier, which may be a first scheduling center; the second scheduling information may be information about a second scheduling device of the demander, which may be a second scheduling center.

[0133] For example, the process of generating work order information can be determined according to the actual situation and is not limited here. As an example, the first terminal device of the demand side can create initial work order information based on the first work information and push the initial work order information to the second terminal device of the supplier. The second terminal device determines the work order information based on the second work information and the initial work order information.

[0134] In some embodiments, before generating work order information, the method further includes: a data sharing device 401 storing a product resource catalog of a first terminal device; receiving selection information based on the product resource catalog sent by a second terminal device; generating contract information based on the selection information; and determining order information based on the contract information, wherein the order information includes order number information, resource name information, supplier name information, buyer name information, order placement time information, computation type information (federated learning, hidden query, joint statistics, privacy intersection), cooperation time information, and other information.

[0135] In some embodiments, the data sharing device 401 determines the scheduling center device based on the first scheduling information in the first working information and the second scheduling information in the second working information, including:

[0136] If the first scheduling information and the second scheduling information are determined to be consistent, then the first scheduling device corresponding to the first scheduling information or the second scheduling device corresponding to the second scheduling information is determined as the scheduling center device.

[0137] If the first scheduling information and the second scheduling information are determined to be inconsistent, then a third scheduling device is determined and designated as the scheduling center device. The third scheduling center is connected to both the first scheduling device and the second scheduling device.

[0138] For example, the first scheduling information can be the name information of a first scheduling device; the second scheduling information can be the name information of a second scheduling device; and the third scheduling device can be a scheduling center connected to the first and second scheduling devices respectively. Specifically, if the name information of the first scheduling device is consistent with the name information of the second scheduling device, then the first scheduling device or the second scheduling device is identified as a scheduling center device to realize the network interconnection of the supplier and the demander through the scheduling center.

[0139] The following example illustrates the node networking configuration method of this application embodiment. Specifically, it can be a method for joint modeling of networked privacy computing nodes, applied to a system for joint modeling of networked privacy computing nodes.

[0140] like Figure 5 As shown, this networked privacy computing node-based joint modeling system comprises three devices: a data sharing device, a scheduling center device, and node devices. The data sharing device includes a resource catalog unit, an order management unit, a work order management unit, and a scheduling center network configuration unit. The scheduling center device includes a delivery center unit, a privacy computing node management unit, a node network configuration unit, and a node monitoring center unit. The node devices primarily provide privacy computing functions, including a task monitor unit and a scheduled task unit. It should be noted that nodes need to be deployed in advance before any task begins.

[0141] like Figure 5 As shown, the joint modeling method based on networked privacy computing nodes includes the following steps:

[0142] Step 501: Product ordering and contract signing.

[0143] Step 502: Generate an order.

[0144] For example, in steps 501 and 502, after the supply and demand parties complete the resource selection and contract signing based on the resource catalog unit of the product in the data sharing device, an order is generated in the order management unit. The order information includes the order number, resource name, supplier name, demander name, order placement time, calculation type (federated learning, hidden query, joint statistics, privacy intersection), cooperation time, etc.

[0145] Step 503: Generate a work order.

[0146] For example, the demander creates a work order in the work order management unit. The work order information includes the work order name, the demander's scheduling center name, the demander's scheduling center address, the demander's privacy computing node ID, node address, node port, number of nodes, computing type (federated learning, hidden query, joint statistics, privacy intersection), resource name, and collaboration time. After creation, it is pushed to the supplier in sequence.

[0147] In the work order management unit, the supplier edits the pushed work order information, adding the supplier's dispatch center name, dispatch center address, privacy computing node ID, node address, node port, number of nodes, etc., participating in the task. After completion, the work order information is pushed to the dispatch center equipment for network configuration.

[0148] Step 504: Interconnection of dispatch center networks.

[0149] For example, the specific method for configuring the interconnection between the supplier and demand dispatch centers is as follows: First, determine whether the dispatch center names of the supplier and demand are consistent. If they are consistent, push the work order to the delivery center unit of the dispatch center equipment. If they are inconsistent, identify a third dispatch center, activate the dispatch center network policy for the third dispatch center for both the supplier and demand, complete the network interconnection, and simultaneously configure the task monitoring unit according to the cooperation time. The network policies of both parties will automatically expire after the cooperation time expires. After completing the network configuration, push the work order to the delivery center pushed to the dispatch center.

[0150] Step 505: Receive work order and extract node information.

[0151] For example, after receiving a work order, the delivery center unit of the dispatch center equipment extracts node information, which includes node name, node address, node port, number of nodes, etc.

[0152] Step 506: Confirm the node network architecture.

[0153] For example, the node information is pushed to the privacy computing node management unit. After receiving the node information, the privacy computing node management unit first determines the number of nodes and then determines the corresponding network architecture based on the number of nodes.

[0154] Step 1: If the value is greater than the threshold b, a hybrid topology architecture is adopted. The hybrid topology architecture is as follows: Figure 6As shown, the nodes are randomly divided into k piles, and a node is randomly selected as the pile center node in each pile. Computational nodes within each pile are interconnected with pile nodes, and pile nodes are interconnected with each other. k can be determined based on the number of cooperating nodes; specifically, it can be divided into piles of 10 nodes each. The optimal pile node can also be selected based on factors such as memory, CPU, and network performance. Each time a modeling task is run, the pile nodes first sum the results, send the sums to the scheduling center for averaging, and then return the average sums to the pile nodes, which then synchronize the data with other nodes.

[0155] The second step is to adopt a star network architecture if the number of nodes is greater than threshold a but less than threshold b. The scheduling center device acts as the central node, and the computing nodes only need to communicate with the central node through the network. The central node sums and averages the results each time and then feeds them back to the computing nodes.

[0156] Step 3: If the value is below the threshold 'a', a mesh topology is adopted, as shown in the mesh topology diagram. Figure 7 As shown, 'a' can be 3-5, and 'b' can be 10 or higher. In the mesh topology, computing nodes are interconnected. During task execution, in each round, based on the network quality information fed back by the node monitoring center unit, the computing node with the best network information is selected as the coordinating node for the next round.

[0157] Step 507: Nodes connect periodically.

[0158] For example, after the privacy computing node management unit determines the network architecture, it pushes the network architecture to the node network configuration unit. The node network configuration unit opens virtual private networks between nodes according to the cooperation time, completing the network connection between nodes. At the same time, a timed task unit is set up so that the node connection is automatically disconnected when the cooperation time expires. After completing the network connection of the nodes, the node network configuration unit pushes a work order to the privacy computing node.

[0159] Step 508: Start the modeling task.

[0160] For example, when a privacy computing node starts a modeling task, the engine's local node interacts with cooperating nodes to exchange encrypted parameters while executing the task. To prevent nodes from temporarily exiting or the network from becoming unavailable during the modeling process, each privacy computing node adds a timer task unit and a task monitor unit to execute the task.

[0161] Step 509: Swap parameters in round t.

[0162] For example, during task execution, each node trains its local model using local data. Once the iterative condition for exchanging model parameters with other nodes is met, it sends its encrypted local parameters to the collaborating nodes and simultaneously receives encrypted parameters from the collaborating nodes. After the local and collaborating parameters are federated and averaged, the local model is updated.

[0163] Step 510: Update the local model in round t.

[0164] For example, before starting the task, the maximum waiting time t1 is set in the timed task unit for each round of parameter interaction. During the task operation, after the local node reaches the iteration condition for exchanging model parameters with other nodes, when other cooperative nodes send parameters to the local node, after the parameter receiving time t1 has expired, no more parameters are received. Instead, the received parameters are directly merged and summarized for federated averaging to update the local model.

[0165] Step 511: Collect node network information.

[0166] For example, when monitoring each round of parameter interaction through the task monitor, information such as node name, CPU, memory, bandwidth between nodes, latency between nodes, jitter between nodes, packet loss between nodes, time of parameter sending by local nodes, name of cooperating nodes that did not receive parameters, and time of cooperating nodes that received parameters is collected, and the above information is transmitted to the node monitoring center unit.

[0167] Step 512: Compare the differences with the previous round.

[0168] Step 513: Check the nodes and remove abnormal nodes.

[0169] For example, in steps 512 and 513, before starting the task, a threshold T is set in the task monitor unit. The node monitoring center unit compares the information from the previous round and calculates the distance. Nodes with a distance greater than T and those that have not provided timely network information are reported to the node network configuration unit of the scheduling center device to detect the node interconnection status (e.g., heartbeat detection). If the status is abnormal, the abnormal node is removed during the next round of node interaction, and only normal nodes are allowed to participate in task execution. Multiple iterations are performed until the node local model is stable, at which point the task execution stops.

[0170] Step 514: Swap parameters in round t+1.

[0171] Step 515: Update the local model in round t+1.

[0172] In an exemplary embodiment, this application also provides a storage medium, namely a computer storage medium, specifically a computer-readable storage medium, such as a memory 302 that stores a computer program. This computer program can be executed by the processor 301 of the node's network configuration device 300 to complete the steps described in the method of this application embodiment. The computer-readable storage medium can be a ROM, PROM, EPROM, EEPROM, Flash Memory, magnetic surface memory, optical disc, or CD-ROM, etc.

[0173] In an exemplary embodiment, this application also provides a computer program product, including a computer program that can be executed by the processor 301 of the networking configuration device 300 of a node to perform the steps described in any of the foregoing methods.

[0174] It should be noted that terms such as "first" and "second" are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0175] Furthermore, the technical solutions described in the embodiments of this application can be combined arbitrarily without conflict.

[0176] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for configuring nodes in a network, characterized in that, Applied to dispatch center equipment, the method includes: Receive work order information, wherein the work order information includes at least the supplier's first work information and the customer's second work information; The first working information and the second working information are extracted to obtain node information; wherein, the node information represents the privacy computing node of the supplier and the privacy computing node of the demander; Based on the number of nodes in the node information, the node networking architecture corresponding to the node information is determined; wherein, the node networking architecture is used to establish network connections between the privacy computing nodes of the supplier and the privacy computing nodes of the demander; the node networking architecture includes a first networking architecture, a second networking architecture, and a third networking architecture; the first networking architecture is a hybrid topology architecture, the second networking architecture is a star network architecture, and the third networking architecture is a mesh topology architecture; If the second time parameter corresponding to the node networking architecture meets the time condition, then the privacy computing node of the supplier and the privacy computing node of the demander are controlled to disconnect from the network; wherein, the second time parameter represents the duration for which the privacy computing node of the supplier and the privacy computing node of the demander establish a network connection based on the node networking architecture; the time condition represents that the duration for which the privacy computing node establishes a network connection in the node networking architecture exceeds the cooperation duration; the cooperation duration is determined according to the cooperation time information of the second work information in the work order information, and the cooperation time information represents the cooperation duration.

2. The method according to claim 1, characterized in that, The step of determining the node network architecture corresponding to the node information based on the number of nodes in the node information includes: If the number of nodes is greater than the first number threshold, then the first network architecture corresponding to the node information is determined; If the number of nodes is greater than a second number threshold and less than a first number threshold, then the second network architecture corresponding to the node information is determined; the second number threshold is less than the first number threshold. If the number of nodes is less than the second number threshold, then the third network architecture corresponding to the node information is determined.

3. The method according to claim 1, characterized in that, After determining the node network architecture corresponding to the node information based on the number of nodes in the node information, the method further includes: If it is determined that the training information of the first node meets the exchange conditions, then the first encryption parameter of the first node is sent to the second node through the node networking architecture, and the second encryption parameter of the second node is sent to the first node through the node networking architecture. Wherein, the first node is a local computing node; if the first node is the privacy computing node of the supplier, the second node is the privacy computing node of the demand party corresponding to the first node; if the first node is the privacy computing node of the demand party, the second node is the privacy computing node of the supplier corresponding to the first node.

4. The method according to claim 3, characterized in that, The method further includes: Obtain the first time parameter corresponding to the node networking architecture; the first time parameter represents the duration of sending the second encrypted parameter of the second node to the first node; If the first time parameter is greater than or equal to the time threshold, the received second encrypted parameter is processed based on the first node to obtain an update parameter; the update parameter is used to update the modeling model corresponding to the first node.

5. The method according to claim 3, characterized in that, The method further includes: Obtain the interaction parameters between the first node and the second node; If the interaction parameter is greater than the interaction threshold, then the second node corresponding to the interaction parameter is detected to obtain the detection result; If the detection result indicates that the second node is in an abnormal state, then the second node is removed.

6. The method according to claim 4, characterized in that, The method further includes: If the first time parameter is greater than or equal to the time threshold, obtain the second node corresponding to the first time parameter; The second node corresponding to the first time parameter is detected to obtain the detection result; If the detection result indicates that the second node is in an abnormal state, then the second node is removed.

7. A node networking configuration device, applied to dispatch center equipment, characterized in that, include: The receiving module is used to receive work order information, which includes at least the supplier's first work information and the demander's second work information; An extraction module is used to extract the first working information and the second working information to obtain node information; wherein the node information represents the privacy computing node of the supplier and the privacy computing node of the demander; The first determining module is used to determine the node network architecture corresponding to the node information based on the number of nodes in the node information; wherein, the node network architecture is used to establish network connections between the privacy computing nodes of the supplier and the privacy computing nodes of the demander; the node network architecture includes a first network architecture, a second network architecture, and a third network architecture; the first network architecture is a hybrid topology architecture, the second network architecture is a star network architecture, and the third network architecture is a mesh topology architecture; The control module is configured to disconnect the network connection between the privacy computing node of the supplier and the privacy computing node of the demander if the second time parameter corresponding to the node networking architecture meets the time condition; wherein, the second time parameter represents the duration for which the privacy computing node of the supplier and the privacy computing node of the demander establish a network connection based on the node networking architecture; the time condition represents that the duration for which the privacy computing node establishes a network connection in the node networking architecture exceeds the cooperation duration; the cooperation duration is determined according to the cooperation time information of the second work information in the work order information, and the cooperation time information represents the cooperation duration.

8. A dispatch center device, characterized in that, include: The processor and memory for storing computer programs that can run on the processor, wherein, The processor, when running a computer program, performs the steps of the method according to any one of claims 1 to 6.

9. A node networking configuration system, characterized in that, Includes data sharing equipment, the scheduling center equipment as described in claim 8, and node equipment; wherein, The data sharing device is used for: Generate work order information, which includes at least the supplier's first work information and the demander's second work information; determine the dispatch center device based on the first scheduling information in the first work information and the second scheduling information in the second work information; and send the work order information to the dispatch center device. The node device is used to control the privacy computing node to obtain training information.

10. The system according to claim 9, characterized in that, The data sharing device determines the scheduling center device based on the first scheduling information in the first working information and the second scheduling information in the second working information, including: If it is determined that the first scheduling information and the second scheduling information are consistent, then the first scheduling device corresponding to the first scheduling information or the second scheduling device corresponding to the second scheduling information is determined as the scheduling center device; If it is determined that the first scheduling information and the second scheduling information are inconsistent, then a third scheduling device is determined and the third scheduling device is identified as the scheduling center device, wherein the third scheduling device is connected to both the first scheduling device and the second scheduling device.

11. A storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

12. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images