A resource management system based on cloud computing

Abstract

A resource management system based on cloud computing relates to the technical field of resource management, and comprises a cloud server, wherein the cloud server is communicatively connected with a data acquisition module, a data storage module, a traffic prediction module, a resource extension module, a resource activation module, a real-time traffic monitoring module and a data aggregation module; the data acquisition module is used to acquire data information; the data storage module is used to store data information; the traffic prediction module is used to obtain the predicted data traffic of each edge device terminal; the resource extension module is used to preset an extended elastic computing node to be activated; the resource activation module determines the estimated activation time period of the extended elastic computing node; the real-time traffic monitoring module is used to determine whether to generate a terminal to be aggregated, and determine the traffic to be aggregated of the terminal to be aggregated; the data aggregation module is used to generate an aggregated data packet and a verification list; the cloud server is used to perform integrity verification on the aggregated data packet to ensure that the cloud server provides stable, secure and efficient cloud services.

Description

Technical Field

[0001] The present invention relates to the technical field of resource management, and in particular to a resource management system based on cloud computing. Background Art

[0002] Prior art CN115904673A "Concurrent scheduling method, device, system, equipment and medium for cloud computing resources" The present invention discloses a concurrent scheduling method, device, system, equipment and medium for cloud computing resources, the method includes: centrally managing the state of cluster resources through a resource management component, the resource management component is used to provide multiple concurrent scheduler processes with a query of the state of resource nodes in the cluster; wherein multiple scheduler processes share the centrally managed cluster resource state; the job scheduling process of the scheduler process includes a screening stage and a sorting stage; the operation performed in the screening stage is: in a sequential or random manner, all or part of the candidate resource nodes that meet the constraint conditions are screened from the cluster, wherein the state of the candidate resource nodes is a normal state; the operation performed in the sorting stage is: based on all or part of the candidate resource nodes, the fitness of each candidate node is calculated; after the calculated number of candidate nodes meets the threshold, the task / task set is selected to be deployed to the node with the highest fitness. The present invention can improve the scheduling efficiency of cloud computing resources.

[0003] The prior art CN104503846A "A resource management system based on cloud computing system" includes multiple data acquisition units, which are used to obtain data information of each simulator running on a distributed computer; a data preprocessing unit, which is used to preprocess the acquired data information and send the preprocessed data information to a data aggregation relay unit; a data aggregation relay unit, which is used to receive and aggregate the preprocessed data information and send the aggregated data information to a resource adjustment unit; a resource adjustment unit, which is used to receive the data information transmitted by the data aggregation relay unit and adjust the allocation of system resources according to the preset rules of the central processing unit. The system realizes the precise adjustment of the computing volume of the entire cloud computing system and the accurate control of its effectiveness, while enhancing the level of data disaster recovery and improving the operating efficiency of the entire system.

[0004] How can a cloud computing-based resource management system reasonably allocate and schedule virtual machine instances and container instances on the cloud computing platform to achieve efficient resource utilization and load balancing, avoid over-allocation or insufficient resources, and dynamically elastically expand and shrink cloud computing resources according to actual load conditions to cope with changes in business volume, ensure that the system can meet demand and minimize resource waste, and face network congestion that may cause packet loss or errors during data transmission, thereby causing data inconsistency. How to quickly trace data to ensure that the cloud computing platform can meet the needs of different application scenarios and provide stable, secure and efficient cloud services are problems that we urgently need to solve. We now provide a cloud computing-based resource management system. Summary of the invention

[0005] In order to solve the above technical problems, the object of the present invention is to provide a resource management system based on cloud computing, including a cloud server, wherein the cloud server is communicatively connected with a data acquisition module, a data storage module, a traffic prediction module, a resource expansion module, a resource activation module, a real-time traffic monitoring module and a data aggregation module;

[0006] The data acquisition module is connected to the cloud server in a distributed manner through the Internet of Things nodes to obtain data information collected by several edge device terminals;

[0007] The data storage module is used to package the data information collected by several edge device terminals into data packets, and store the data information and collection records of the corresponding edge devices, wherein the collection records include the collection time, monitoring period and data size;

[0008] The traffic prediction module is used to build a traffic prediction model based on the historical data traffic of each edge device and obtain the predicted data traffic of each edge device terminal;

[0009] The resource expansion module is used to predictively expand each edge device terminal according to the predicted data traffic of each edge device terminal and the traffic load level of each edge device terminal, and preset the expansion elastic computing node to be activated;

[0010] The resource activation module determines the estimated activation time period of the extended elastic computing node of each edge device terminal according to the predicted data traffic of each edge device terminal;

[0011] The real-time traffic monitoring module is used to determine whether to generate a terminal to be aggregated according to the real-time data traffic of the edge device terminal and the traffic load level related to the edge device terminal, and determine the traffic to be aggregated of the terminal to be aggregated;

[0012] The data aggregation module is used to package several groups of data information transmitted from several terminals to be aggregated to the aggregation terminal into an aggregated data packet, generate a verification list of all data information in the aggregated data packet, and send the aggregated data packet and the verification list of the aggregated data packet to the cloud server;

[0013] The cloud server is used to perform integrity verification on the data information in the aggregated data packet according to the verification list of the aggregated data packet.

[0014] Furthermore, the resource management system based on cloud computing also includes a data visualization module, and the process of constructing a data path visualization diagram according to the communication link relationship and location information of each edge device terminal and the cloud server in the physical space includes:

[0015] Obtain the longitude and latitude coordinates of each edge device terminal and cloud server in the target area by means of GIS, construct a two-dimensional plane map of the target area, and map each edge device terminal and cloud server to the two-dimensional plane map of the target area according to the longitude and latitude coordinates to obtain a location layer;

[0016] A topological network is constructed based on the location layer, and the communication link relationship between each edge device terminal and the cloud server in the physical space is obtained. Each edge device terminal and the cloud server in the location layer is used as a node of the topological network, and the communication link relationship between each edge device terminal and the cloud server in the physical space is used as the connection relationship between the nodes.

[0017] Acquire real-time data information collected by each edge device terminal, preprocess the data information in data format, generate twin data, and match the twin data with the edge device terminals in the topological network to generate a data path visualization diagram.

[0018] Furthermore, the traffic prediction module builds a traffic prediction model according to the historical data traffic of each edge device, and the process of obtaining the predicted data traffic of each edge device terminal includes:

[0019] Obtain historical collection records of each edge device terminal from the data storage module, obtain historical data traffic of each edge device at different times in several historical monitoring cycles according to the historical collection records, and use the historical data traffic as a test set and a training set;

[0020] Construct a traffic prediction model, input the training set into the traffic prediction model for training until the loss function training is stable, and save the model parameters, test the traffic prediction model through the test set until it meets the preset requirements, output the traffic prediction model, and obtain the predicted data traffic of each edge device terminal at each moment in the current monitoring period according to the traffic prediction model.

[0021] Furthermore, the resource expansion module predictively expands each edge device terminal according to the predicted data traffic of each edge device terminal and the traffic load level of each edge device terminal, and the process of presetting the expanded elastic computing node to be activated includes:

[0022] Obtaining a predicted data traffic upper limit of each edge device terminal and a traffic load level of each edge device terminal, and comparing the traffic load level of each edge device terminal with the predicted data traffic upper limit;

[0023] If the traffic load level of the edge device terminal is less than the predicted data traffic upper limit, the traffic difference between the traffic load level of the edge device terminal and the predicted data traffic upper limit is obtained, and an extended elastic computing node corresponding to the traffic load level is preset at the edge device terminal based on the traffic difference.

[0024] Furthermore, the process of the resource activation module determining the estimated activation time period of the extended elastic computing node of each edge device terminal according to the predicted data traffic of each edge device terminal includes:

[0025] Obtain the predicted data traffic of the edge device terminal at each moment in the current monitoring period, compare the predicted data traffic of the edge device terminal with the traffic load level of the edge device terminal, filter out the time period in which the predicted data traffic of the edge device terminal is greater than the traffic load level in the current monitoring period, obtain the maximum traffic difference between the predicted data traffic of the edge device terminal and the traffic load level in the time period, and obtain the estimated activation time period of the extended elastic computing node with the traffic load level corresponding to the maximum traffic difference in the current monitoring period based on the maximum traffic difference.

[0026] Further, the process in which the real-time traffic monitoring module determines whether to generate a terminal to be aggregated according to the real-time data traffic of the edge device terminal and the traffic load level related to the edge device terminal, and determines the traffic to be aggregated of the terminal to be aggregated includes:

[0027] Acquire the real-time data traffic of the edge device terminal, compare the real-time data traffic with the traffic load level of each edge device terminal, and if the real-time data traffic is greater than the traffic load level of the edge device terminal, determine whether there is an extended elastic computing node for the edge device terminal;

[0028] If there is an extended elastic computing node, determine whether the traffic load level of the extended elastic computing node is greater than the traffic difference between the real-time data traffic and the traffic load level of the edge device terminal; if greater than, activate the extended elastic computing node with the corresponding traffic load level according to the traffic difference; if less than or equal to, activate the extended elastic computing node with the highest traffic load level, mark the edge device terminal as a terminal to be aggregated, obtain the sum of the traffic load level of the terminal to be aggregated and the traffic load level of the highest traffic load level of the extended elastic computing node, obtain the absolute value of the traffic difference between the sum of the traffic load levels and the real-time data traffic, and mark the absolute value of the traffic difference as the traffic to be aggregated of the terminal to be aggregated;

[0029] If there is no extended elastic computing node, the edge device terminal is marked as a terminal to be aggregated, the absolute value of the flow difference between the flow load level of the terminal to be aggregated and the real-time data flow is obtained, and the absolute value of the flow difference is marked as the flow to be aggregated of the terminal to be aggregated.

[0030] Furthermore, the process of the aggregation terminal matching module screening out the aggregation terminal from other edge device terminals according to the position relationship and data flow relationship between the plurality of terminals to be aggregated and other edge device terminals includes:

[0031] Based on the data path visual graph, the location information of several terminals to be aggregated and the cloud server in the current monitoring period is obtained, and the sum of the traffic to be aggregated of the several terminals to be aggregated is obtained, and the sum of the traffic load levels of other edge device terminals in the target area and the highest traffic load level of the corresponding extended elastic computing node is obtained, and the real-time data traffic of other edge device terminals is obtained, and the traffic difference between the sum of the traffic load levels of other edge device terminals and the corresponding real-time data traffic is marked as a callable resource;

[0032] Filter out other edge device terminals whose callable resources are greater than the sum of the traffic to be aggregated, obtain the location information of other edge device terminals based on the data path visibility diagram, and obtain the communication link path lengths between several terminals to be aggregated and other edge device terminals and the communication link path lengths between other edge device terminals and the cloud server according to the location information of several terminals to be aggregated, other edge device terminals and the cloud server;

[0033] Other edge device terminals whose sum of communication link path lengths with several terminals to be aggregated and with a cloud server is the shortest are screened out, and the other edge device terminals are marked as aggregated terminals.

[0034] Further, the data aggregation module packages several groups of data information transmitted from several terminals to be aggregated to the aggregation terminal into an aggregated data packet, generates a verification list of all data information in the aggregated data packet, and sends the aggregated data packet and the verification list of the aggregated data packet to the cloud server, including:

[0035] Packing several groups of data information transmitted by several terminals to be aggregated to the aggregation terminal into an aggregation data packet, then converting each group of data information in the aggregation data packet into binary data of a fixed length, applying a hash function to the binary data, obtaining a hash value corresponding to each group of data information, and marking the hash value corresponding to each group of data information as a leaf node;

[0036] Then, pair the adjacent leaf nodes in pairs, concatenate the hash values ​​of each pair of leaf nodes and apply the hash function again to obtain the hash value corresponding to each pair of leaf nodes, mark the hash value corresponding to each pair of leaf nodes as a branch node, then pair the adjacent branch nodes in pairs, concatenate the hash values ​​of each pair of branch nodes and apply the hash function again to obtain the hash value corresponding to each pair of branch nodes, repeat the above steps until only one hash value is generated, and mark the hash value as the main node;

[0037] Generate a verification tree of the aggregated data packet according to several leaf nodes, branch nodes and main nodes in the aggregated data packet, then obtain the node position of each group of data information in the aggregated data packet in the verification tree, and generate a verification list of all data information in the aggregated data packet according to the hash value of the leaf node corresponding to each group of data information and the node position in the verification tree;

[0038] The aggregated data packet and the verification list of the aggregated data packet are sent to the cloud server.

[0039] Furthermore, the process of the cloud server performing integrity verification on the data information in the aggregated data packet according to the verification list of the aggregated data packet includes:

[0040] The cloud server reconstructs the verification tree based on the received aggregated data packet and performs consistency matching between the received verification list and the verification tree;

[0041] If the hash value of the leaf node corresponding to each set of data information in the received verification list and the node position in the verification tree are consistent with the hash value of the leaf node corresponding to each set of data information in the verification tree and the node position in the verification tree, then mark the verification pass label on the aggregate data packet;

[0042] If the hash value of the leaf node corresponding to a certain data information in the received verification list and the node position in the verification tree are inconsistent with the hash value of the leaf node corresponding to the data information of the verification tree and the node position in the verification tree, then a certain data information loss label is marked in the aggregated data packet, and a certain data information resend instruction is sent to the aggregated terminal. After receiving the certain data information resend instruction, the aggregated terminal packages the certain data information into a data packet and sends it to the cloud server.

[0043] Compared with the prior art, the present invention has the following beneficial effects:

[0044] 1. By predicting the expansion of each edge device terminal according to the predicted data traffic and traffic load level of each edge device terminal, presetting the extended elastic computing nodes to be activated, and reasonably allocating and scheduling the computing nodes on the edge of the cloud computing platform, efficient resource utilization and load balancing can be achieved to avoid over-allocation or insufficient resources.

[0045] 2. By determining the estimated activation time period of the extended elastic computing nodes of each edge device terminal based on the predicted data traffic of each edge device terminal, the cloud computing resources are elastically expanded and contracted dynamically according to the actual load conditions to cope with changes in business volume, ensuring that the system can meet demand and minimize resource waste.

[0046] 3. When the resource management network encounters network congestion, the edge device terminals whose real-time data traffic is greater than the load traffic level are marked as terminals to be aggregated, and edge device terminals with idle resources in the surrounding area are selected as aggregation terminals. Multiple groups of data information are packaged into an aggregated data packet to avoid constructing a data packet for each group of data information, saving bandwidth resources of the resource management network and improving the throughput of the resource management network.

[0047] 4. Send the aggregated data packet and the verification list of the aggregated data packet to the cloud server. The cloud server is used to verify the integrity of the data information in the aggregated data packet according to the verification list of the aggregated data packet. When tracing the missing data information in the aggregated data packet, it is only necessary to match the verification list with the verification tree of the aggregated data packet to determine the specific missing data information. There is no need for the aggregation terminal to transmit and verify the entire aggregated data packet to the cloud server again, thereby ensuring that the cloud server provides stable, secure and efficient cloud services. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 This is a schematic diagram of a resource management system based on cloud computing according to an embodiment of the present application. DETAILED DESCRIPTION

[0049] The following is a clear and complete description of the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0050] like Figure 1 As shown, a resource management system based on cloud computing includes a cloud server, wherein the cloud server is communicatively connected with a data acquisition module, a data storage module, a traffic prediction module, a resource expansion module, a resource activation module, a real-time traffic monitoring module and a data aggregation module;

[0051] The data acquisition module is connected to the cloud server in a distributed manner through the Internet of Things nodes to obtain data information collected by several edge device terminals;

[0052] The data storage module is used to package the data information collected by several edge device terminals into data packets, and store the data information and collection records of the corresponding edge devices, wherein the collection records include the collection time, monitoring period and data size;

[0053] The traffic prediction module is used to build a traffic prediction model based on the historical data traffic of each edge device and obtain the predicted data traffic of each edge device terminal;

[0054] The resource expansion module is used to predictively expand each edge device terminal according to the predicted data traffic of each edge device terminal and the traffic load level of each edge device terminal, and preset the expansion elastic computing node to be activated;

[0055] The resource activation module determines the estimated activation time period of the extended elastic computing node of each edge device terminal according to the predicted data traffic of each edge device terminal;

[0056] The real-time traffic monitoring module is used to determine whether to generate a terminal to be aggregated according to the real-time data traffic of the edge device terminal and the traffic load level related to the edge device terminal, and determine the traffic to be aggregated of the terminal to be aggregated;

[0057] The data aggregation module is used to package several groups of data information transmitted from several terminals to be aggregated to the aggregation terminal into an aggregated data packet, generate a verification list of all data information in the aggregated data packet, and send the aggregated data packet and the verification list of the aggregated data packet to the cloud server;

[0058] The cloud server is used to perform integrity verification on the data information in the aggregated data packet according to the verification list of the aggregated data packet.

[0059] It should be further explained that, in the specific implementation process, the resource management system based on cloud computing also includes a data visualization module, and the process of the data visualization module constructing a data path visualization diagram according to the communication link relationship and location information of each edge device terminal and the cloud server in the physical space includes:

[0060] Obtain the longitude and latitude coordinates of each edge device terminal and cloud server in the target area by means of GIS, construct a two-dimensional plane map of the target area, and map each edge device terminal and cloud server to the two-dimensional plane map of the target area according to the longitude and latitude coordinates to obtain a location layer;

[0061] A topological network is constructed based on the location layer, and the communication link relationship between each edge device terminal and the cloud server in the physical space is obtained. Each edge device terminal and the cloud server in the location layer is used as a node of the topological network, and the communication link relationship between each edge device terminal and the cloud server in the physical space is used as the connection relationship between the nodes.

[0062] Acquire real-time data information collected by each edge device terminal, preprocess the data information in data format, generate twin data, and match the twin data with the edge device terminals in the topological network to generate a data path visualization diagram.

[0063] It should be further explained that, in the specific implementation process, the traffic prediction module builds a traffic prediction model according to the historical data traffic of each edge device, and the process of obtaining the predicted data traffic of each edge device terminal includes:

[0064] Obtain historical collection records of each edge device terminal from the data storage module, obtain historical data traffic of each edge device at different times in several historical monitoring cycles according to the historical collection records, and use the historical data traffic as a test set and a training set;

[0065] Construct a traffic prediction model, input the training set into the traffic prediction model for training until the loss function training is stable, and save the model parameters, test the traffic prediction model through the test set until it meets the preset requirements, output the traffic prediction model, and obtain the predicted data traffic of each edge device terminal at each moment in the current monitoring period according to the traffic prediction model.

[0066] It should be further explained that, in the specific implementation process, the resource expansion module predictively expands each edge device terminal according to the predicted data traffic of each edge device terminal and the traffic load level of each edge device terminal, and the process of presetting the expanded elastic computing node to be activated includes:

[0067] Obtaining a predicted data traffic upper limit of each edge device terminal and a traffic load level of each edge device terminal, and comparing the traffic load level of each edge device terminal with the predicted data traffic upper limit;

[0068] If the traffic load level of the edge device terminal is less than the predicted data traffic upper limit, the traffic difference between the traffic load level of the edge device terminal and the predicted data traffic upper limit is obtained, and an extended elastic computing node corresponding to the traffic load level is preset at the edge device terminal based on the traffic difference.

[0069] It should be further explained that, in a specific implementation process, the process in which the resource activation module determines the estimated activation time period of the extended elastic computing node of each edge device terminal according to the predicted data traffic of each edge device terminal includes:

[0070] Obtain the predicted data traffic of the edge device terminal at each moment in the current monitoring period, compare the predicted data traffic of the edge device terminal with the traffic load level of the edge device terminal, filter out the time period in which the predicted data traffic of the edge device terminal is greater than the traffic load level in the current monitoring period, obtain the maximum traffic difference between the predicted data traffic of the edge device terminal and the traffic load level in the time period, and obtain the estimated activation time period of the extended elastic computing node with the traffic load level corresponding to the maximum traffic difference in the current monitoring period based on the maximum traffic difference.

[0071] It should be further explained that, in a specific implementation process, the real-time traffic monitoring module determines whether to generate a terminal to be aggregated according to the real-time data traffic of the edge device terminal and the traffic load level related to the edge device terminal, and determines the process of the traffic to be aggregated of the terminal to be aggregated, including:

[0072] Acquire the real-time data traffic of the edge device terminal, compare the real-time data traffic with the traffic load level of each edge device terminal, and if the real-time data traffic is greater than the traffic load level of the edge device terminal, determine whether there is an extended elastic computing node for the edge device terminal;

[0073] If there is an extended elastic computing node, determine whether the traffic load level of the extended elastic computing node is greater than the traffic difference between the real-time data traffic and the traffic load level of the edge device terminal; if greater than, activate the extended elastic computing node with the corresponding traffic load level according to the traffic difference; if less than or equal to, activate the extended elastic computing node with the highest traffic load level, mark the edge device terminal as a terminal to be aggregated, obtain the sum of the traffic load level of the terminal to be aggregated and the traffic load level of the highest traffic load level of the extended elastic computing node, obtain the absolute value of the traffic difference between the sum of the traffic load levels and the real-time data traffic, and mark the absolute value of the traffic difference as the traffic to be aggregated of the terminal to be aggregated;

[0074] If there is no extended elastic computing node, the edge device terminal is marked as a terminal to be aggregated, the absolute value of the flow difference between the flow load level of the terminal to be aggregated and the real-time data flow is obtained, and the absolute value of the flow difference is marked as the flow to be aggregated of the terminal to be aggregated.

[0075] It should be further explained that, in a specific implementation process, the process of the aggregation terminal matching module screening out the aggregation terminal from other edge device terminals according to the position relationship and data flow relationship between the plurality of terminals to be aggregated and other edge device terminals includes:

[0076] Based on the data path visual graph, the location information of several terminals to be aggregated and the cloud server in the current monitoring period is obtained, and the sum of the traffic to be aggregated of the several terminals to be aggregated is obtained, and the sum of the traffic load levels of other edge device terminals in the target area and the highest traffic load level of the corresponding extended elastic computing node is obtained, and the real-time data traffic of other edge device terminals is obtained, and the traffic difference between the sum of the traffic load levels of other edge device terminals and the corresponding real-time data traffic is marked as a callable resource;

[0077] Filter out other edge device terminals whose callable resources are greater than the sum of the traffic to be aggregated, obtain the location information of other edge device terminals based on the data path visibility diagram, and obtain the communication link path lengths between several terminals to be aggregated and other edge device terminals and the communication link path lengths between other edge device terminals and the cloud server according to the location information of several terminals to be aggregated, other edge device terminals and the cloud server;

[0078] Other edge device terminals whose sum of communication link path lengths with several terminals to be aggregated and with a cloud server is the shortest are screened out, and the other edge device terminals are marked as aggregated terminals.

[0079] It should be further explained that, in the specific implementation process, the data aggregation module packages several groups of data information transmitted by several terminals to be aggregated to the aggregation terminal into an aggregated data packet, generates a verification list of all data information in the aggregated data packet, and sends the aggregated data packet and the verification list of the aggregated data packet to the cloud server. The process includes:

[0080] Packing several groups of data information transmitted by several terminals to be aggregated to the aggregation terminal into an aggregation data packet, then converting each group of data information in the aggregation data packet into binary data of a fixed length, applying a hash function to the binary data, obtaining a hash value corresponding to each group of data information, and marking the hash value corresponding to each group of data information as a leaf node;

[0081] Then, pair the adjacent leaf nodes in pairs, concatenate the hash values ​​of each pair of leaf nodes and apply the hash function again to obtain the hash value corresponding to each pair of leaf nodes, mark the hash value corresponding to each pair of leaf nodes as a branch node, then pair the adjacent branch nodes in pairs, concatenate the hash values ​​of each pair of branch nodes and apply the hash function again to obtain the hash value corresponding to each pair of branch nodes, repeat the above steps until only one hash value is generated, and mark the hash value as the main node;

[0082] Generate a verification tree of the aggregated data packet according to several leaf nodes, branch nodes and main nodes in the aggregated data packet, then obtain the node position of each group of data information in the aggregated data packet in the verification tree, and generate a verification list of all data information in the aggregated data packet according to the hash value of the leaf node corresponding to each group of data information and the node position in the verification tree;

[0083] The aggregated data packet and the verification list of the aggregated data packet are sent to the cloud server.

[0084] It should be further explained that, in the specific implementation process, the process of the cloud server verifying the integrity of the data information in the aggregated data packet according to the verification list of the aggregated data packet includes:

[0085] The cloud server reconstructs the verification tree based on the received aggregated data packet and performs consistency matching between the received verification list and the verification tree;

[0086] If the hash value of the leaf node corresponding to each set of data information in the received verification list and the node position in the verification tree are consistent with the hash value of the leaf node corresponding to each set of data information in the verification tree and the node position in the verification tree, then mark the verification pass label on the aggregate data packet;

[0087] If the hash value of the leaf node corresponding to a certain data information in the received verification list and the node position in the verification tree are inconsistent with the hash value of the leaf node corresponding to the data information of the verification tree and the node position in the verification tree, then a certain data information loss label is marked in the aggregated data packet, and a certain data information resend instruction is sent to the aggregated terminal. After receiving the certain data information resend instruction, the aggregated terminal packages the certain data information into a data packet and sends it to the cloud server.

[0088] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A resource management system based on cloud computing, comprising a cloud server, characterized in that: The cloud server is communicatively connected with a data acquisition module, a data storage module, a traffic prediction module, a resource expansion module, a resource activation module, a real-time traffic monitoring module and a data aggregation module; The data acquisition module is connected to the cloud server in a distributed manner through the IoT nodes to obtain data information collected by several edge device terminals; The data storage module is used to package the data information collected by several edge device terminals into data packets, and store the data information and collection records of the corresponding edge devices, wherein the collection records include the collection time, monitoring period and data size; The traffic prediction module is used to build a traffic prediction model based on the historical data traffic of each edge device and obtain the predicted data traffic of each edge device terminal; The resource expansion module is used to predictively expand each edge device terminal according to the predicted data traffic of each edge device terminal and the traffic load level of each edge device terminal, and preset the expansion elastic computing node to be activated; The resource activation module determines the estimated activation time period of the extended elastic computing node of each edge device terminal according to the predicted data traffic of each edge device terminal; The real-time traffic monitoring module is used to determine whether to generate a terminal to be aggregated according to the real-time data traffic of the edge device terminal and the traffic load level related to the edge device terminal, and determine the traffic to be aggregated of the terminal to be aggregated; The data aggregation module is used to package several groups of data information transmitted from several terminals to be aggregated to the aggregation terminal into an aggregated data packet, generate a verification list of all data information in the aggregated data packet, and send the aggregated data packet and the verification list of the aggregated data packet to the cloud server; The cloud server is used to verify the integrity of the data information in the aggregated data packet according to the verification list of the aggregated data packet; The process of determining the estimated activation time period of the extended elastic computing node of each edge device terminal according to the predicted data traffic of each edge device terminal by the resource activation module includes: Obtain the predicted data traffic of the edge device terminal at each moment in the current monitoring period, compare the predicted data traffic of the edge device terminal with the traffic load level of the edge device terminal, screen out the time period in which the predicted data traffic of the edge device terminal is greater than the traffic load level in the current monitoring period, obtain the maximum traffic difference between the predicted data traffic of the edge device terminal and the traffic load level in the time period, and obtain the estimated activation time period of the extended elastic computing node with the traffic load level corresponding to the maximum traffic difference in the current monitoring period according to the maximum traffic difference; The data aggregation module packages several groups of data information transmitted from several terminals to be aggregated to the aggregation terminal into an aggregated data packet, generates a verification list of all data information in the aggregated data packet, and sends the aggregated data packet and the verification list of the aggregated data packet to the cloud server. The process includes: Packing several groups of data information transmitted by several terminals to be aggregated to the aggregation terminal into an aggregation data packet, then converting each group of data information in the aggregation data packet into binary data of a fixed length, applying a hash function to the binary data, obtaining a hash value corresponding to each group of data information, and marking the hash value corresponding to each group of data information as a leaf node; Then, pair the adjacent leaf nodes in pairs, concatenate the hash values ​​of each pair of leaf nodes and apply the hash function again to obtain the hash value corresponding to each pair of leaf nodes, mark the hash value corresponding to each pair of leaf nodes as a branch node, then pair the adjacent branch nodes in pairs, concatenate the hash values ​​of each pair of branch nodes and apply the hash function again to obtain the hash value corresponding to each pair of branch nodes, repeat the above steps until only one hash value is generated, and mark the hash value as the main node; Generate a verification tree of the aggregated data packet according to several leaf nodes, branch nodes and main nodes in the aggregated data packet, then obtain the node position of each group of data information in the aggregated data packet in the verification tree, and generate a verification list of all data information in the aggregated data packet according to the hash value of the leaf node corresponding to each group of data information and the node position in the verification tree; The aggregated data packet and the verification list of the aggregated data packet are sent to the cloud server.

2. A cloud computing-based resource management system according to claim 1, characterized in that: It also includes a data visualization module, and the process of constructing a data path visualization diagram according to the communication link relationship and location information of each edge device terminal and the cloud server in the physical space includes: Obtain the longitude and latitude coordinates of each edge device terminal and cloud server in the target area by means of GIS, construct a two-dimensional plane map of the target area, and map each edge device terminal and cloud server to the two-dimensional plane map of the target area according to the longitude and latitude coordinates to obtain a location layer; A topological network is constructed based on the location layer, and the communication link relationship between each edge device terminal and the cloud server in the physical space is obtained. Each edge device terminal and the cloud server in the location layer is used as a node of the topological network, and the communication link relationship between each edge device terminal and the cloud server in the physical space is used as the connection relationship between the nodes. Acquire real-time data information collected by each edge device terminal, preprocess the data information in data format, generate twin data, and match the twin data with the edge device terminals in the topological network to generate a data path visualization diagram.

3. A cloud computing-based resource management system according to claim 2, characterized in that: The traffic prediction module builds a traffic prediction model based on the historical data traffic of each edge device, and the process of obtaining the predicted data traffic of each edge device terminal includes: Obtain historical collection records of each edge device terminal from the data storage module, obtain historical data traffic of each edge device at different times in several historical monitoring cycles according to the historical collection records, and use the historical data traffic as a test set and a training set; Construct a traffic prediction model, input the training set into the traffic prediction model for training until the loss function training is stable, and save the model parameters, test the traffic prediction model through the test set until it meets the preset requirements, output the traffic prediction model, and obtain the predicted data traffic of each edge device terminal at each moment in the current monitoring period according to the traffic prediction model.

4. A cloud computing-based resource management system according to claim 3, characterized in that: The resource expansion module predictively expands each edge device terminal according to the predicted data traffic of each edge device terminal and the traffic load level of each edge device terminal, and the process of presetting the expanded elastic computing node to be activated includes: Obtaining a predicted data traffic upper limit of each edge device terminal and a traffic load level of each edge device terminal, and comparing the traffic load level of each edge device terminal with the predicted data traffic upper limit; If the traffic load level of the edge device terminal is less than the predicted data traffic upper limit, the traffic difference between the traffic load level of the edge device terminal and the predicted data traffic upper limit is obtained, and an extended elastic computing node corresponding to the traffic load level is preset at the edge device terminal based on the traffic difference.

5. A cloud computing-based resource management system according to claim 4, characterized in that: The process in which the real-time traffic monitoring module determines whether to generate a terminal to be aggregated according to the real-time data traffic of the edge device terminal and the traffic load level related to the edge device terminal, and determines the traffic to be aggregated of the terminal to be aggregated includes: Acquire the real-time data traffic of the edge device terminal, compare the real-time data traffic with the traffic load level of each edge device terminal, and if the real-time data traffic is greater than the traffic load level of the edge device terminal, determine whether there is an extended elastic computing node for the edge device terminal; If there is an extended elastic computing node, determine whether the traffic load level of the extended elastic computing node is greater than the traffic difference between the real-time data traffic and the traffic load level of the edge device terminal; if greater than, activate the extended elastic computing node with the corresponding traffic load level according to the traffic difference; if less than or equal to, activate the extended elastic computing node with the highest traffic load level, mark the edge device terminal as a terminal to be aggregated, obtain the sum of the traffic load level of the terminal to be aggregated and the traffic load level of the highest traffic load level of the extended elastic computing node, obtain the absolute value of the traffic difference between the sum of the traffic load levels and the real-time data traffic, and mark the absolute value of the traffic difference as the traffic to be aggregated of the terminal to be aggregated; If there is no extended elastic computing node, the edge device terminal is marked as a terminal to be aggregated, the absolute value of the flow difference between the flow load level of the terminal to be aggregated and the real-time data flow is obtained, and the absolute value of the flow difference is marked as the flow to be aggregated of the terminal to be aggregated.

6. A cloud computing-based resource management system according to claim 5, characterized in that: The process of the aggregation terminal matching module screening out an aggregation terminal from other edge device terminals according to the position relationship and data flow relationship between a plurality of terminals to be aggregated and other edge device terminals includes: Based on the data path visual graph, the location information of several terminals to be aggregated and the cloud server in the current monitoring period is obtained, and the sum of the traffic to be aggregated of the several terminals to be aggregated is obtained, and the sum of the traffic load levels of other edge device terminals in the target area and the highest traffic load level of the corresponding extended elastic computing node is obtained, and the real-time data traffic of other edge device terminals is obtained, and the traffic difference between the sum of the traffic load levels of other edge device terminals and the corresponding real-time data traffic is marked as a callable resource; Filter out other edge device terminals whose callable resources are greater than the sum of the traffic to be aggregated, obtain the location information of other edge device terminals based on the data path visibility diagram, and obtain the communication link path lengths between several terminals to be aggregated and other edge device terminals and the communication link path lengths between other edge device terminals and the cloud server according to the location information of several terminals to be aggregated, other edge device terminals and the cloud server; Other edge device terminals whose sum of communication link path lengths with several terminals to be aggregated and with the cloud server is the shortest are screened out, and the other edge device terminals are marked as aggregated terminals.

7. A cloud computing-based resource management system according to claim 6, characterized in that: The process of the cloud server performing integrity verification on the data information in the aggregated data packet according to the verification list of the aggregated data packet includes: The cloud server reconstructs the verification tree based on the received aggregated data packet and performs consistency matching between the received verification list and the verification tree; If the hash value of the leaf node corresponding to each set of data information in the received verification list and the node position in the verification tree are consistent with the hash value of the leaf node corresponding to each set of data information in the verification tree and the node position in the verification tree, then mark the verification pass label on the aggregate data packet; If the hash value of the leaf node corresponding to a certain data information in the received verification list and the node position in the verification tree are inconsistent with the hash value of the leaf node corresponding to the data information of the verification tree and the node position in the verification tree, then a certain data information loss label is marked in the aggregated data packet, and a certain data information resend instruction is sent to the aggregated terminal. After receiving the certain data information resend instruction, the aggregated terminal packages the certain data information into a data packet and sends it to the cloud server.

Images