Information management method and system based on cloud platform, terminal and storage medium

Abstract

The invention relates to an information management method and system based on a cloud platform, a terminal and a storage medium, and relates to the technical field of communication, and the method comprises the steps: responding to a received configuration request, generating a configuration model according to the configuration request, and obtaining a model configuration requirement corresponding to the configuration model; the configuration model is pruned and quantized according to the model configuration requirements, at least two quantitative configuration models are obtained, and different quantitative configuration models consume different calculated amounts; acquiring configuration information of the edge node; grouping the edge nodes according to the quantitative configuration model and the configuration information to obtain edge node groups and target quantitative configuration models in one-to-one correspondence with the edge node groups; and transmitting the target quantitative configuration model to the edge nodes in groups. The method has the effects of reducing the hardware condition influence of the edge node and improving the configurability of the model.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to an information management method, system, terminal and storage medium based on a cloud platform. Background Art

[0002] The cloud platform is a service platform based on cloud computing technology, which provides computing resources, storage resources, network resources and various application services through the Internet. The information management configuration of the cloud platform is becoming increasingly important.

[0003] When configuring user models with the same function for edge nodes, the data center will integrate the existing model architecture to obtain a unified user model. After that, the data center will send the user model to each edge node to implement the deployment of the user model.

[0004] With respect to the above-mentioned related technologies, the inventors believe that the hardware conditions of edge nodes are different. When configuring the model, a unified model will make it difficult for edge nodes with poor hardware conditions to fully use the model. Summary of the invention

[0005] In order to reduce the impact of hardware conditions on edge nodes and improve the configurability of the model, the present application provides an information management method, system, terminal and storage medium based on a cloud platform.

[0006] In the first aspect, the present application provides an information management method based on a cloud platform, which adopts the following technical solution: An information management method based on a cloud platform, comprising: In response to receiving a configuration request, generating a configuration model according to the configuration request, and obtaining a model configuration requirement corresponding to the configuration model; Pruning and quantizing the configuration model according to the model configuration requirements to obtain at least two quantized configuration models, wherein different quantized configuration models consume different amounts of computation; Get the configuration information of the edge node; The edge nodes are grouped according to the quantitative configuration model and the configuration information to obtain edge node groups and target quantitative configuration models corresponding to the edge node groups one by one; The target quantization configuration model is transmitted to the edge nodes in groups.

[0007] By adopting the above technical solution, the configuration model is pruned and quantized according to the model configuration requirements to obtain at least two quantized configuration models. And for edge nodes with different configuration information, a target quantized configuration model is allocated. This allows the target quantized configuration model to adapt to edge nodes with different configurations, reduces the impact of hardware conditions on edge nodes, and improves the configurability of the model.

[0008] Optionally, extracting device performance information from the configuration information; Acquire the network transmission performance of the edge node according to the device performance information; Predicting the transmission duration of the target quantization configuration model according to the target quantization configuration model and the network transmission performance; When the transmission duration is greater than a preset duration, a network group is added for the edge node.

[0009] By adopting the above technical solution, the transmission time of the target quantitative configuration model is predicted according to the target quantitative configuration model and network transmission performance. When the transmission time is longer than the preset time, network networking is added to the edge node to improve the network transmission performance between the data center and the edge node, and shorten the time for the data center to deploy the configuration model to the edge node.

[0010] Optionally, the target quantization configuration model is split into several functional modules; Counting the data capacity of the functional modules; According to the data capacity and the capacity threshold, classify the plurality of functional modules to obtain a first functional module and a second functional module, wherein the data capacity of the first functional module is greater than the capacity threshold, and the data capacity of the second functional module is less than the capacity threshold; Transmitting the first functional module to the edge node in groups through the network; The second functional module is transmitted to the edge node in groups through the host networking.

[0011] By adopting the above technical solution, the first functional module and the second functional module are transmitted to the edge node through different networking modes, and the transmission characteristics of different networking are fully utilized to improve the transmission efficiency of the first functional module and the second functional module.

[0012] Optionally, a group representative node is selected from the edge node group; Count the edge nodes with the same target quantitative configuration model and group them to obtain the group representative node combination; Splitting a functional module into a plurality of sub-modules, wherein the functional module includes the first functional module and the second functional module; Counting the number of management nodes corresponding to the group representative nodes; The submodules are sent to the group representative nodes according to the number of management nodes, wherein the number of submodules sent is negatively correlated with the number of management nodes, and the group representative nodes are used to obtain the functional modules by sharing the submodule combination.

[0013] By adopting the above technical solution, submodules are sent to group representative nodes according to the number of management nodes, so that the group representative nodes that manage more nodes obtain fewer submodules, reducing the losses caused by information security accidents of group representative nodes. Moreover, different group representative nodes obtain different submodules, and the use of distributed deployment can also reduce the impact of model leakage.

[0014] Optionally, in response to monitoring that an abnormal condition occurs in the transmission network of the first group representative node, searching for a matching candidate group representative node in the edge node group corresponding to the first group representative node according to the configuration information of the first group representative node; Calculating the transmission distance between the candidate group representative node and the data center; Taking the candidate group representative node corresponding to the minimum value of the transmission distance as the second group representative node; Stop sending the submodule to the first group representative node, and send the submodule to the second group representative node.

[0015] By adopting the above technical solution, when an abnormal situation occurs in the transmission network of the first group representative node, the second group representative node is used to replace the first group representative node, thereby ensuring that the process of deploying the target quantitative configuration model is not interrupted and improving the data transmission efficiency and the stability of the transmission process.

[0016] Optionally, in response to receiving a node abnormality signal, counting the target group representative nodes that send the node abnormality signal, wherein the node abnormality signal is sent when a model obtained by combining the target group representative nodes is inconsistent with the target quantization configuration model; When the number of abnormal signals of the node is greater than a preset number threshold, determining an abnormal group representative node according to the target group representative node; Removing the abnormal group representative node from the abnormal edge node group in which it is located; A warning signal is sent to the abnormal group representative node.

[0017] By adopting the above technical solution, after a node abnormality signal is generated, the abnormal group representative node will be judged according to the abnormal signal sent by the target group representative node, and the abnormal group representative node will be removed to ensure the safety of the remaining edge nodes and realize a behavioral model for automatically discovering edge node vulnerabilities.

[0018] Optionally, obtaining access records of representative nodes of the abnormal group; Extracting abnormal visitors and abnormal behavior records of the abnormal visitors from the access records; Sending the abnormal visitor and the abnormal behavior record to the edge node; Receiving the search result returned by the edge node; A candidate abnormal edge node is obtained according to the search result.

[0019] By adopting the above technical solution, the abnormal visitors of the abnormal group representative node and the abnormal behavior records of the abnormal visitors are used to find out whether other edge nodes have similar problems, so as to find candidate abnormal edge nodes that may have problems for further verification by relevant personnel.

[0020] In the second aspect, the present application provides an information management system based on a cloud platform, which adopts the following technical solutions: An information management system based on a cloud platform, comprising: An acquisition module is used to obtain configuration requests, configuration information, preset duration, data capacity, capacity threshold, group representative nodes, number of management nodes, transmission distance, node abnormality signals, preset number threshold, warning signals, access records and retrieval results; A memory, used to store a program of the cloud platform-based information management method; The program in the memory can be loaded and executed by the processor to implement the cloud platform-based information management method.

[0021] By adopting the above technical solution, the configuration model is pruned and quantized according to the model configuration requirements to obtain at least two quantized configuration models. And for edge nodes with different configuration information, a target quantized configuration model is allocated. This allows the target quantized configuration model to adapt to edge nodes with different configurations, reduces the impact of hardware conditions on edge nodes, and improves the configurability of the model.

[0022] In a third aspect, the present application provides a smart terminal, which adopts the following technical solution: An intelligent terminal comprises a memory and a processor, wherein the memory stores a computer program which can be loaded by the processor and execute any one of the above-mentioned methods.

[0023] In a fourth aspect, the present application provides a computer storage medium capable of storing corresponding programs, which is convenient for reducing the impact of hardware conditions of edge nodes and improving the configurability of the model, and adopts the following technical solutions: A computer-readable storage medium stores a computer program that can be loaded by a processor and execute any of the above-mentioned cloud platform-based information management methods.

[0024] In summary, the present application includes at least one of the following beneficial technical effects: 1. Prune and quantize the configuration model according to the model configuration requirements to obtain at least two quantized configuration models. And assign the target quantized configuration model to the edge nodes with different configuration information. This allows the target quantized configuration model to adapt to edge nodes with different configurations, reduce the impact of the hardware conditions of the edge nodes, and improve the configurability of the model; 2. According to the target quantitative configuration model and network transmission performance, predict the transmission time of the target quantitative configuration model. If the transmission time is longer than the preset time, add network networking to the edge node to improve the network transmission performance between the data center and the edge node, and shorten the time for the data center to deploy the configuration model to the edge node; 3. If an abnormal situation occurs in the transmission network of the first group representative node, the second group representative node is used to replace the first group representative node to ensure that the process of deploying the target quantitative configuration model is not interrupted, thereby improving the data transmission efficiency and the stability of the transmission process. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a flow chart of a cloud platform-based information management method provided in an embodiment of the present application.

[0026] Figure 2 It is a flow chart of a method for configuring a network of edge nodes provided in an embodiment of the present application.

[0027] Figure 3 It is a flow chart of a method for transmitting a configuration model provided in an embodiment of the present application.

[0028] Figure 4 It is a flow chart of a data transmission method based on a cloud platform provided in an embodiment of the present application.

[0029] Figure 5 It is a flow chart of a cloud platform-based exception management method provided in an embodiment of the present application.

[0030] Figure 6 It is a flow chart of a cloud platform-based abnormal warning method 1 provided in an embodiment of the present application.

[0031] Figure 7 It is a flow chart of a second abnormal warning method based on a cloud platform provided in an embodiment of the present application.

[0032] Figure 8 It is a structural diagram of a cloud platform-based information management system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0033] In order to make the purpose, technical solutions and advantages of this application more clear, the following Figure 1 To Attachment Figure 8It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0034] The present application embodiment discloses a cloud platform-based information management method. The method is executed by a data center, referring to Figure 1 , the method comprising: Step S101: in response to receiving a configuration request, generating a configuration model according to the configuration request, and acquiring a model configuration requirement corresponding to the configuration model.

[0035] The cloud platform includes a data center and edge nodes. The data center is used to manage multiple edge nodes, and the data center can provide services to the edge nodes. Furthermore, the data center can provide services to the edge nodes by deploying models. Among them, if the edge node has already deployed a model, the model upgrade service can also be provided to the edge node.

[0036] The configuration request is an operation requesting the data center to deploy the model to the edge node. Optionally, the configuration request includes at least one of the model type, the deployment time limit of the model, the model number, and the model version.

[0037] The configuration model can be any type of model, and this application does not make any specific limitation on this.

[0038] Model configuration requirements refer to the hardware requirements for computer devices when deploying and configuring models on computer devices. Optionally, model configuration requirements include at least one of GPU (Graphics Processing Unit), TPU (Tensor Processing Unit), memory, and database storage.

[0039] Step S102: pruning and quantizing the configuration model according to the model configuration requirements to obtain at least two quantized configuration models, wherein different quantized configuration models consume different amounts of computation.

[0040] Pruning is used to remove unimportant parameters or neuron structures in the configuration model, thereby reducing the complexity of the configuration model while maintaining its performance.

[0041] Quantization is used to reduce the numerical precision of model parameters that configure the model, thereby reducing storage and computational overhead.

[0042] Exemplarily, a quantization model configuration requirement is generated according to the model configuration requirement, wherein the quantization model configuration requirement is lower than the model configuration requirement. The configuration model is pruned and quantized according to the quantization model configuration requirement to obtain at least two quantization configuration models. For example, the memory in the model configuration requirement is reduced to obtain the quantization model configuration requirement.

[0043] Furthermore, the parameters (which may also be the neuron structure) in the configuration model are evaluated according to the preset standards to obtain the importance score of the parameters, and the importance score is positively correlated with the importance of the parameters. The parameters are classified into necessary parameters and non-essential parameters according to the importance score. The non-essential parameters are removed in order from small to large to obtain the candidate configuration model. Compare whether the model configuration requirements of the candidate configuration model are less than the configuration requirements of the quantitative model. If so, the candidate configuration model is used as the quantitative configuration model. If not, continue to perform the step of removing non-essential parameters.

[0044] Step S103: Obtain configuration information of the edge node.

[0045] The configuration information of the edge node includes content that is consistent with the model configuration requirement, but the configuration information of the edge node has different values ​​from the model configuration requirement.

[0046] Optionally, the data center sends a configuration request to the edge node. After receiving the configuration request, the edge node sends configuration information to the data center.

[0047] Step S104: grouping edge nodes according to the quantization configuration model and the configuration information to obtain edge node groups and target quantization configuration models corresponding to the edge node groups one by one.

[0048] In some embodiments, the first model configuration requirements corresponding to the quantitative configuration model are counted. A first model configuration requirement interval is generated according to the first model configuration requirement, and the model configuration requirements within the first model configuration requirement interval are not greater than the first model configuration requirement. The first model configuration requirement interval is used as a classification standard, and the edge nodes within the same first model configuration requirement interval are divided into the same group to obtain edge node groups and target quantitative configuration models corresponding to the edge node groups one by one.

[0049] Furthermore, in order to improve the fault tolerance rate, an upper limit can be set for the number of edge nodes in the edge node group. When the number of edge nodes assigned to the quantitative configuration model reaches the upper limit of the number of edge nodes, another group of edge node groups can be set. At this time, the same quantitative configuration model will correspond to at least two groups of edge node groups.

[0050] Step S105: Transmitting the target quantization configuration model to edge nodes in groups.

[0051] Optionally, the data center can transmit the target quantitative configuration model in a decentralized manner or as a whole.

[0052] By adopting the above technical solution, the configuration model is pruned and quantized according to the model configuration requirements to obtain at least two quantized configuration models. And for edge nodes with different configuration information, a target quantized configuration model is allocated. This allows the target quantized configuration model to adapt to edge nodes with different configurations, reduces the impact of hardware conditions on edge nodes, and improves the configurability of the model.

[0053] In the following embodiments, when deploying the target quantitative configuration model, it is necessary to consider that the network transmission performance of each edge node is different. In order to ensure that each edge node can complete the deployment of the model within a certain period of time, it is necessary to optimize the network status of the edge node according to the actual situation. Therefore, the embodiment of the present application discloses a method for configuring the network of edge nodes. Figure 2 , the method comprising: Step S201: extracting device performance information from configuration information.

[0054] Device performance information refers to device parameters on a computer device that affect network communication. Exemplarily, the device performance information includes at least one of CPU, memory, network card, network interface performance, and database read and write speed.

[0055] Step S202: Obtain the network transmission performance of the edge node according to the device performance information.

[0056] Network transmission performance is used to indicate the network data transmission speed between data centers and edge nodes.

[0057] In some embodiments, the network delay of the edge node is calculated based on the CPU, memory and database read and write speed. The network bandwidth between the edge node and the data center is calculated based on the network interface performance. The network delay and network bandwidth are scored to obtain a delay score and a bandwidth score. The delay score and the bandwidth score are weighted to obtain the network transmission performance.

[0058] Step S203: predicting the transmission duration of the target quantization configuration model according to the target quantization configuration model and network transmission performance.

[0059] Exemplarily, the transmission volume of the target quantitative configuration model is calculated, and the transmission duration is calculated based on the transmission volume and the network transmission performance.

[0060] Step S204: when the transmission time is longer than the preset time, adding a network group for the edge node.

[0061] Network networking refers to the virtual network implemented at the network device level (such as switches, routers, etc.), and the logical network is constructed through physical network devices. Optionally, network networking enables data packets to be encapsulated and transmitted between network devices by configuring tunneling protocols on network devices.

[0062] In some other embodiments, when the transmission duration is less than the preset duration, it means that the network transmission condition of the edge node is good and there is no need to add network networking.

[0063] By adopting the above technical solution, the transmission time of the target quantitative configuration model is predicted according to the target quantitative configuration model and network transmission performance. When the transmission time is longer than the preset time, network networking is added to the edge node to improve the network transmission performance between the data center and the edge node, and shorten the time for the data center to deploy the configuration model to the edge node.

[0064] In the following embodiments, when transmitting the configuration model, in order to improve the transmission efficiency, different modules can be transmitted for different transmission modes to improve the transmission efficiency. Therefore, the embodiment of the present application discloses a transmission method of the configuration model. Figure 3 , the method comprising: Step S301: split the target quantization configuration model into several functional modules.

[0065] Exemplarily, a functional layer of a target quantization configuration model is obtained, where the functional layer includes at least one of a fully connected layer, a convolutional layer, a recurrent layer, and an output layer. According to the functions of each functional layer, functional layers with the same functions are split into the same functional module.

[0066] Step S302: Count the data capacity of the functional modules.

[0067] Data capacity is used to represent the space occupied by the model's modules in the computer device.

[0068] Exemplarily, the number of parameters and data types in the function module are counted and calculated. According to the number of parameters and data types, the data capacity of the function module is calculated. Different data types correspond to different basic data capacities. For example, if a parameter of data type A corresponds to a data capacity, then b data of data type A need to consume a*b data capacity.

[0069] Step S303: According to the data capacity and the capacity threshold, several functional modules are classified to obtain a first functional module and a second functional module, the data capacity of the first functional module is greater than the capacity threshold, and the data capacity of the second functional module is less than the capacity threshold.

[0070] The capacity threshold is a preset empirical value, and relevant personnel can adjust the specific value of the capacity threshold according to actual needs.

[0071] Exemplarily, if the data capacity of the submodule is greater than the capacity threshold, the submodule is classified as a first functional module. If the data capacity of the submodule is less than the capacity threshold, the submodule is classified as a second functional module.

[0072] Step S304: Transmitting the first functional module in groups to the edge node through the network.

[0073] It should be noted that there is no order in which step S304 and step S305 are executed. Step S304 may be executed first, or step S305 may be executed first, or step S304 and step S305 may be executed simultaneously.

[0074] Step S305: Transmit the second functional module to the edge node in groups through the host networking.

[0075] Host networking refers to a virtual network implemented at the host level (such as a virtual machine or container), which creates a logical network between hosts through software. For example, a virtual network agent runs on the edge node to encapsulate and transmit data packets through tunneling technology.

[0076] Among them, network networking has better network processing efficiency than host networking and is suitable for transmitting large volumes of data.

[0077] By adopting the above technical solution, the first functional module and the second functional module are transmitted to the edge node through different networking modes, and the transmission characteristics of different networking are fully utilized to improve the transmission efficiency of the first functional module and the second functional module.

[0078] In the following embodiments, when transmitting data from a data center to an edge node, it is necessary to consider how to reduce the adverse effects caused by data leakage. The present application embodiment discloses a data transmission method based on a cloud platform. Figure 4 , the method comprising: Step S401: Select a group representative node from the edge node group.

[0079] The group representative node may be any edge node in the edge node group.

[0080] Exemplarily, the edge node with the best configuration information in the edge node group is selected as the group representative node. Alternatively, the edge nodes in the edge node group vote to elect the group representative node. Alternatively, the edge node with the best network transmission performance in the edge node group is selected as the group representative node.

[0081] Step S402: Count the edge node groups with the same target quantization configuration model to obtain group representative node combinations.

[0082] The group representative node combination is composed of group representative nodes, and the group representative nodes in the same group representative node combination correspond to the same target quantitative configuration model.

[0083] Exemplarily, edge node groups having the same target quantization configuration model are counted, and group representative nodes in these edge node groups are extracted to form a group representative node combination.

[0084] Step S403: split the functional module into several sub-modules, where the functional module includes a first functional module and a second functional module.

[0085] Furthermore, this step includes two situations: splitting the first functional module into a plurality of sub-modules and splitting the second functional module into a plurality of sub-modules.

[0086] For example, taking the fully connected layer as the functional module, the fully connected layer is split into several sub-layers according to the output dimension of the fully connected layer to obtain several sub-modules. For example, if the output dimension of the fully connected layer is m, the fully connected layer can be split into two sub-layers, and the output dimension of each sub-layer is m / 2. Alternatively, if the weight matrix of the fully connected layer is an n×m matrix, it is decomposed into the product of several small matrices through matrix decomposition technology to achieve the splitting of the fully connected layer.

[0087] Furthermore, the difference between the data capacities of the sub-modules obtained by splitting is not less than the preset data capacity, so as to ensure that the data capacity of each sub-module is close, which is beneficial to the subsequent allocation and sending of the sub-modules.

[0088] Step S404: Count the number of management nodes corresponding to the group representative nodes.

[0089] The number of management nodes refers to the total number of edge nodes in the edge node group where the group representative node is located.

[0090] Exemplarily, the data center sends working query information to the edge nodes in the edge node group, and the working query information is used to query whether the edge nodes are in working state. The data center counts the working state information returned by the edge nodes to obtain the number of management nodes.

[0091] Exemplarily, the data center sends node quantity query information to the group representative node. After receiving the data query information, the group representative node sends work query information to the edge nodes in the same edge node group. The group representative node counts the work status information returned by the edge nodes to obtain the number of management nodes. The group representative node sends the number of management nodes to the data center.

[0092] Step S405: Submodules are sent to group representative nodes according to the number of management nodes, wherein the number of submodules sent is negatively correlated with the number of management nodes, and the group representative nodes are used to obtain functional modules by combining shared submodules.

[0093] Optionally, the group representative nodes are sorted from small to large according to the number of management nodes to obtain the group representative node sorting. The number of submodules sent to the group representative nodes is allocated according to the group representative node sorting. On the one hand, when the total number of group representative nodes is greater than the number of submodules, the number of submodules is recorded as a, and the total number of group representative nodes is recorded as b, where b is greater than a. Randomly take out a-1 submodules from the submodule. Assign a-1 submodules to the 1st group representative node to the a-1th group representative node in the group representative node sorting, so that there is a one-to-one correspondence between each submodule and the group representative node. Assign the remaining submodules after taking out a-1 submodules to the ath group representative node to the bth group representative node. On the other hand, when the total number of group representative nodes is less than or equal to the number of submodules, the division of the number of submodules and the total number of group representative nodes is calculated to obtain the quotient and remainder, and the quotient is recorded as c and the remainder is recorded as d. If the remainder is 0, randomly take out b groups of submodules from the submodules, and each group of submodules includes c-1 submodules. Evenly distribute b groups of submodules to the group representative nodes. Randomly take out p submodules from the remaining b submodules and assign them to the first group representative node, p=[b / 2]+1, where [b / 2] means rounding down b / 2. Distribute the remaining bp submodules to the second group representative node to the bth group representative node. If the remainder is not 0, randomly take out b groups of submodules from the module, each group of submodules includes c submodules. Randomly take out q submodules from the remaining d submodules and assign them to the first group representative node, q=[d / 2]+1. Distribute the remaining dq submodules to the second group representative node to the bth group representative node.

[0094] Furthermore, after completing the transmission of the submodules, the group representative node will share the submodules it holds with other group representative nodes. At this point, the group representative node can obtain all the complete functional modules by combining the submodules, and combine the functional modules into a complete target quantitative configuration model. Afterwards, the group representative node deploys the target quantitative configuration model to other edge nodes in the edge node group.

[0095] Furthermore, before transmitting the submodule, a check code may be added to the submodule. The group representative node may determine whether the target quantitative configuration model is complete through the check code.

[0096] By adopting the above technical solution, submodules are sent to group representative nodes according to the number of management nodes, so that the group representative nodes that manage more nodes obtain fewer submodules, reducing the losses caused by information security accidents of group representative nodes. Moreover, different group representative nodes obtain different submodules, and the use of distributed deployment can also reduce the impact of model leakage.

[0097] In the following embodiments, when a problem is detected in the transmission to the group representative node, other edge nodes can be selected from the edge node group to replace the group representative node to improve the stability and anti-interference ability of the transmission. Therefore, the embodiment of the present application discloses an abnormality management method based on a cloud platform. Figure 5 , the method comprising: Step S501: In response to monitoring that an abnormal condition occurs in the transmission network of the first group representative node, searching for a matching candidate group representative node in the edge node group corresponding to the first group representative node according to configuration information of the first group representative node.

[0098] The abnormal conditions include at least one of communication interruption, network delay and jitter, data synchronization failure, data tampering or loss, and network attack.

[0099] Exemplarily, the configuration difference is obtained based on the configuration information of the first group representative node and the configuration information of other edge nodes in the edge node group, and the edge nodes with a configuration difference less than a preset difference threshold are selected as candidate group representative nodes.

[0100] Step S502: Calculate the transmission distance between the candidate group representative node and the data center.

[0101] The transmission distance is determined based on the information transmission method between the candidate group representative node and the data center. For example, when the candidate group representative node and the data center communicate via optical fiber, the transmission distance refers to the physical distance between the candidate group representative node and the data center. When the candidate group representative node and the data center communicate via a wireless network, the transmission distance refers to the network logical distance between the candidate group representative node and the data center.

[0102] Step S503: taking the candidate group representative node corresponding to the minimum value of the transmission distance as the second group representative node.

[0103] In some embodiments, if there are at least two candidate group representative nodes corresponding to the minimum value in the transmission distance, a second group representative node is randomly selected.

[0104] Step S504: stop sending the submodule to the first group representative node, and send the submodule to the second group representative node.

[0105] In some embodiments, if the abnormal condition is any one of communication interruption, network delay and jitter, or data synchronization failure, after stopping sending the sub-module to the first group representative node, the first group representative node sends the sub-module that the data center has transmitted to the first group representative node to the second group representative node.

[0106] In some embodiments, the data center may send the completed submodule to the second group representative.

[0107] By adopting the above technical solution, when an abnormal situation occurs in the transmission network of the first group representative node, the second group representative node is used to replace the first group representative node, thereby ensuring that the process of deploying the target quantitative configuration model is not interrupted and improving the data transmission efficiency and the stability of the transmission process.

[0108] In the following embodiments, the edge node can also detect whether there is a security problem in the group representative node by judging the integrity of the target quantitative configuration model. Therefore, the embodiment of the present application discloses a cloud platform-based abnormal warning method 1. Figure 6 , the method comprising: Step S601: In response to receiving a node abnormality signal, counting the target group representative nodes that send the node abnormality signal, the node abnormality signal is sent when the model obtained by combining the target group representative nodes is inconsistent with the target quantitative configuration model.

[0109] Exemplarily, before transmitting the submodule, the data center will add a check code to the submodule. The group representative node can use the check code to determine whether the target quantitative configuration model is complete. Furthermore, the edge node can verify the check code of each submodule to determine whether the submodule is available.

[0110] The node abnormality signal includes the sub-model number where the abnormality occurs.

[0111] Step S602: when the number of abnormal node signals is greater than a preset number threshold, determine the abnormal group representative node according to the target group representative node.

[0112] The preset quantity threshold is a preset experience value, and relevant personnel can adjust the value of the preset quantity threshold according to actual needs.

[0113] Exemplarily, the abnormal sub-model number is determined according to the node abnormality signal, and the group representative node corresponding to the abnormal sub-model number is determined to obtain the abnormal group representative node.

[0114] In some other embodiments, during the data transmission process, accidental events such as missing data or garbled data may occur. Therefore, when the number of abnormal node signals is less than a preset threshold, the above-mentioned accidental events are likely to have occurred, and therefore the step of determining the abnormal group representative node based on the target group representative node is not performed.

[0115] Step S603: remove the abnormal group representative node from the abnormal edge node group in which it is located.

[0116] Exemplarily, the communication connection between the data center and the abnormal group representative node is disconnected. Alternatively, the data center stops receiving data or information from the abnormal group representative.

[0117] Step S604: Send a warning signal to the representative node of the abnormal group.

[0118] Warning signals are used to notify abnormal groupings that nodes represent potential safety risks.

[0119] In some other embodiments, the user terminal corresponding to the abnormal group representative node is determined, and a warning signal is sent to the user terminal.

[0120] By adopting the above technical solution, after a node abnormality signal is generated, the abnormal group representative node will be judged according to the abnormal signal sent by the target group representative node, and the abnormal group representative node will be removed to ensure the safety of the remaining edge nodes and realize a behavioral model for automatically discovering edge node vulnerabilities.

[0121] In the following embodiments, the security risks existing in other edge nodes can also be found through the abnormal group representative node. Therefore, the embodiment of the present application discloses a second abnormal warning method based on the cloud platform. Figure 7 , the method comprising: Step S701: Obtain access records of representative nodes of abnormal groups.

[0122] Access records include login records, file access records, network access records, and service access records. Among them, login records are used to record detailed information about users or systems logging into a node, and login records include at least one of login time, login IP address, login user, login method, and login success or failure status. File access records are used to record read and write operations on files on a node. File access records include at least one of file path, operation user, operation time, and operation type. Network access records are used to record network communications between abnormal group representative nodes and other devices, and network access records include at least one of source IP, target IP, communication port, communication protocol, transmission time, and connection status. Service access records are used to record the access behavior of services running on abnormal group representative nodes, and service access records include at least one of access time, request path, client IP, response status code, and response time.

[0123] Step S702: extracting abnormal visitors and abnormal behavior records of abnormal visitors from the access records.

[0124] Exemplarily, abnormal behavior is set. Abnormal visitors matching abnormal behavior are extracted from the access records, and abnormal behavior records corresponding to the abnormal visitors are obtained. The abnormal behavior may include at least one of high-frequency access, access during unconventional time periods, frequent login failures, access through unconventional paths, and abnormal IP addresses.

[0125] Step S703: Send abnormal visitor and abnormal behavior records to the edge node.

[0126] Furthermore, after the edge node receives the abnormal visitor and abnormal behavior record, the edge node searches its own access record according to the abnormal visitor and abnormal behavior record to obtain the search result. If the abnormal visitor or abnormal behavior record is retrieved, the search result has a risk. If the abnormal visitor and abnormal behavior record are not retrieved, the search result has no risk.

[0127] Step S704: receiving the search result returned by the edge node.

[0128] Step S705: Obtain candidate abnormal edge nodes according to the search results.

[0129] Exemplarily, risk retrieval results for which no risk exists are extracted, and edge nodes corresponding to the risk retrieval results are determined to obtain candidate abnormal edge nodes.

[0130] By adopting the above technical solution, the abnormal visitors of the abnormal group representative node and the abnormal behavior records of the abnormal visitors are used to find out whether other edge nodes have similar problems, so as to find candidate abnormal edge nodes that may have problems for further verification by relevant personnel.

[0131] Based on the same inventive concept, the present application embodiment provides an information management system based on a cloud platform, please refer to Figure 8 , the system comprises: The acquisition module 801 is used to acquire configuration requests, configuration information, preset duration, data capacity, capacity threshold, group representative node, number of management nodes, transmission distance, node abnormality signal, preset number threshold, warning signal, access record and retrieval result; Memory 802, used to store the program of the cloud platform-based information management method; Processor 803, the program in the memory can be loaded and executed by the processor to implement the cloud platform-based information management method.

[0132] By adopting the above technical solution, the configuration model is pruned and quantized according to the model configuration requirements to obtain at least two quantized configuration models. And for edge nodes with different configuration information, a target quantized configuration model is allocated. This allows the target quantized configuration model to adapt to edge nodes with different configurations, reduces the impact of hardware conditions on edge nodes, and improves the configurability of the model.

[0133] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device and unit described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0134] An embodiment of the present application provides a computer-readable storage medium storing a computer program that can be loaded by a processor and execute a cloud platform-based information management method.

[0135] Computer storage media include, for example, various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks or optical disks.

[0136] Based on the same inventive concept, an embodiment of the present application provides a smart terminal, including a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executes a cloud platform-based information management method.

[0137] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device and unit described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0138] The above are all preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Any feature disclosed in this specification (including the abstract and drawings), unless otherwise stated, can be replaced by other equivalent or alternative features with similar purposes. That is, unless otherwise stated, each feature is only an example of a series of equivalent or similar features.

Claims

1. A cloud platform-based information management method, characterized in that: The cloud platform includes a data center and an edge node. The method is performed by the data center. The method includes: In response to receiving a configuration request, generating a configuration model according to the configuration request, and obtaining a model configuration requirement corresponding to the configuration model; Pruning and quantizing the configuration model according to the model configuration requirements to obtain at least two quantized configuration models, wherein different quantized configuration models consume different amounts of computation; Get the configuration information of the edge node; The edge nodes are grouped according to the quantitative configuration model and the configuration information to obtain edge node groups and target quantitative configuration models corresponding to the edge node groups one by one; The target quantization configuration model is transmitted to the edge nodes in groups.

2. The cloud platform-based information management method according to claim 1, characterized in that: Before transmitting the target quantization configuration model to the edge nodes in groups, the method further includes: extracting device performance information from the configuration information; Acquire the network transmission performance of the edge node according to the device performance information; Predicting the transmission duration of the target quantization configuration model according to the target quantization configuration model and the network transmission performance; When the transmission duration is greater than a preset duration, a network group is added for the edge node.

3. The cloud platform-based information management method according to claim 2, characterized in that: The transmitting the target quantization configuration model to the edge nodes in groups includes: Splitting the target quantization configuration model into several functional modules; Counting the data capacity of the functional modules; According to the data capacity and the capacity threshold, classify the plurality of functional modules to obtain a first functional module and a second functional module, wherein the data capacity of the first functional module is greater than the capacity threshold, and the data capacity of the second functional module is less than the capacity threshold; Transmitting the first functional module to the edge node in groups through the network; The second functional module is transmitted to the edge node in groups through the host networking.

4. The cloud platform-based information management method according to claim 3, characterized in that: The method further comprises: Selecting a group representative node from the edge node group; Count the edge nodes with the same target quantitative configuration model and group them to obtain the group representative node combination; Splitting a functional module into a plurality of sub-modules, wherein the functional module includes the first functional module and the second functional module; Counting the number of management nodes corresponding to the group representative nodes; The submodules are sent to the group representative nodes according to the number of management nodes, wherein the number of submodules sent is negatively correlated with the number of management nodes, and the group representative nodes are used to obtain the functional modules by sharing the submodule combination.

5. The cloud platform-based information management method according to claim 4, characterized in that: The method further comprises: In response to monitoring that an abnormal condition occurs in the transmission network of the first group representative node, searching for a matching candidate group representative node in the edge node group corresponding to the first group representative node according to the configuration information of the first group representative node; Calculating the transmission distance between the candidate group representative node and the data center; Taking the candidate group representative node corresponding to the minimum value of the transmission distance as the second group representative node; Stop sending the submodule to the first group representative node, and send the submodule to the second group representative node.

6. The cloud platform-based information management method according to claim 4, characterized in that: The method further comprises: In response to receiving a node abnormality signal, counting the target group representative nodes that send the node abnormality signal, wherein the node abnormality signal is sent when a model obtained by combining the target group representative nodes is inconsistent with the target quantization configuration model; When the number of abnormal signals of the node is greater than a preset number threshold, determining an abnormal group representative node according to the target group representative node; Removing the abnormal group representative node from the abnormal edge node group in which it is located; A warning signal is sent to the abnormal group representative node.

7. The cloud platform-based information management method according to claim 6, characterized in that: The method further comprises: Obtaining access records of representative nodes of the abnormal group; Extracting abnormal visitors and abnormal behavior records of the abnormal visitors from the access records; Sending the abnormal visitor and the abnormal behavior record to the edge node; Receiving the search result returned by the edge node; A candidate abnormal edge node is obtained according to the search result.

8. An information management system based on a cloud platform, characterized in that: For executing the information management method based on a cloud platform as claimed in any one of claims 1 to 7, the system comprises: An acquisition module is used to obtain configuration requests, configuration information, preset duration, data capacity, capacity threshold, group representative nodes, number of management nodes, transmission distance, node abnormality signals, preset number threshold, warning signals, access records and retrieval results; A memory, used to store a program of the cloud platform-based information management method; The program in the memory can be loaded and executed by the processor to implement the cloud platform-based information management method.

9. An intelligent terminal, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: A computer program is stored which can be loaded by a processor and execute the method according to any one of claims 1 to 7.

Images