Information processing method and system based on cloud computing

Abstract

The invention provides an information processing method and system based on cloud computing, and the method comprises the steps: a user submits a service request to the information processing system based on cloud computing at a client, the service request comprises the carbon emission data of the user, and a load balancer calculates the resource demand according to the node load condition; an emission source, action and transaction information and emission points are constructed into a graph, a collaborative filtering model and a method based on a knowledge graph are combined to analyze the type and content of a service request, and computing resources, storage resources and network resources are dynamically allocated; a Q-learning algorithm is used to update a carbon emission value function of the load balancer, and an optimal carbon emission data service effect is realized; and a user sends a carbon emission data processing request to the analysis scheduler, the carbon emission data is processed by the analysis scheduler and the virtual machine, and finally a processing result is returned to the client. And efficient, safe and reliable carbon emission data processing and service are provided.

Description

Technical Field

[0001] This application relates to the field of computer technology, and more specifically, to an information processing method and system based on cloud computing. Background Technology

[0002] With the increasing severity of carbon emissions, a low-carbon economy has become a crucial direction for global economic development. Cloud computing, as a highly efficient and energy-saving information technology, has been widely applied across various industries. However, traditional cloud computing technologies suffer from several problems when processing large amounts of data, such as slow data transmission speeds and low computational efficiency, which severely restrict the development of cloud computing technology. To address these issues, we propose a cloud-based information processing method and system to achieve more efficient, secure, and reliable carbon emission data processing and services. A collaborative filtering model combined with a knowledge graph-based approach, along with reinforcement learning algorithms, is employed to optimize the system's carbon emission value function, achieving the best carbon emission data service effect. The research and application of this technology are of great significance for promoting the development of a low-carbon economy and solving the carbon emission problem. Summary of the Invention

[0003] The embodiments of this application provide a cloud computing-based information processing method and system, which can, to some extent, solve the problem of carbon emission data services.

[0004] Other features and advantages of this application will become apparent from the following detailed description, or may be learned in part from practice of this application.

[0005] According to one aspect of the embodiments of this application, a cloud computing-based information processing method is provided, comprising: a user submitting a service request, including the user's carbon emission data, to a cloud computing-based information processing system on a client side; a load balancer calculating resource requirements based on node load; constructing a graph of emission sources, action and transaction information, and emission points; using a combination of collaborative filtering models and knowledge graph-based methods to analyze the service request type and content, and dynamically allocating computing resources, storage resources, and network resources; using a Q-learning algorithm to update the carbon emission value function of the load balancer to achieve optimal carbon emission data service performance; the user sending a carbon emission data processing request to a parsing scheduler, which processes the request through the parsing scheduler and virtual machines, and finally returning the processing result to the client; calculating a security score for the carbon emission data processing request, prioritizing the submission of service content to nodes with high scores for processing, and simultaneously using an ECC encryption algorithm to ensure the secure transmission and storage of sensitive data; dynamically allocating and managing resources for the carbon emission data processing request; and sending the processing result, using container orchestration technology and virtualization technology to achieve dynamic allocation and management of computing resources for the carbon emission data processing request.

[0006] In some embodiments of this application, based on the foregoing scheme, the user submits a service request to the cloud-based information processing system on the client, and the load balancer calculates the resource requirements, including: Users submit service requests to a cloud-based information processing system. These requests include user carbon emission data. With user permission, the user's carbon emission data is uploaded to the cloud storage system, including but not limited to emission sources: dietary records (types of food, time of food consumption, quantity of food consumed), travel records (driving mileage, public transportation use, walking distance), energy consumption records (household electricity consumption, water consumption, gas consumption); action and transaction information: carbon emission reduction action records (environmental protection measures, emission reduction effects, emission reduction time), carbon trading records (carbon trading projects, trading time, trading volume); emission points: buildings, locations. The load balancer calculates resource requirements based on the current load of each node: Resource requirements = S × R × T, where S is the size of the service request, R is the processing speed of the service request, and T is the duration of the service request; the larger the amount of carbon emission data submitted by the user, the larger the size of the service request S, and the higher the resource requirements. In some embodiments of this application, based on the foregoing scheme, the parsing scheduler predicts the required number of node resources and dynamically allocates resources using a knowledge graph and collaborative filtering model, including: The parsing scheduler parses the service requests submitted by users, using a combination of collaborative filtering models and knowledge graph-based methods to analyze the service request type and content; it constructs a graph of related entities and relationships, with entities including emission sources, action and transaction information, and emission points; and connections between emission sources, action and transaction information, and emission points. For example, if a user request includes "household electricity consumption," it is related to "buildings" and "locations" in the knowledge graph: Household electricity consumption × Weight of the relationship between household electricity consumption and building + Building × Weight of the relationship between building and location; The parsing scheduler parses the service requests submitted by users, matches the user requests with entities and relationships in the knowledge graph, and determines the user's needs. It uses a collaborative filtering model to learn and recommend services or resources based on user needs. By learning the user's historical service requests, it discovers the similarity between users and recommends services or resources that the current user and similar users have previously requested: Recommendation degree = Σ (similarity between the user and the current user × the user's previous service requests).

[0007] Based on the results of the collaborative filtering model and knowledge graph, the parsing scheduler calculates the computing, storage, and network resources required for the service request, and calculates the number of node resources required. When dynamically allocating computing, storage, and network resources, the parsing scheduler dynamically allocates resources based on the node resource requirements of the computing and the API calls of the cloud computing platform, including but not limited to requests to create virtual machines and requests for storage resources.

[0008] In some embodiments of this application, based on the aforementioned scheme, a Q-learning algorithm is used to update the carbon emission value function of the load balancer to achieve effective scheduling and load balancing of computing resources. This includes: using a Q-learning algorithm to update the carbon emission value function, Q(s,a)=(1-α)×Q(s,a)+α×(r+γ×max(Q(s',a'))), where s represents different records of the carbon emission data service; a represents different operations related to carbon emission calculation and analysis; Q(s,a) represents the carbon emission value of taking action a in state s; and α is the learning rate. 0.1-0.9 is used to control the update degree of Q value; r is the immediate reward obtained by taking action a in the current state s, which depends on the contribution of action a to carbon emission data service; γ is a discount factor used to measure the degree to which the current decision considers future rewards, and γ is 0.1-0.9; max(Q(s',a')) means taking the action a' with the highest value among all possible actions in the next state s' to maximize the carbon emission value after the state transition. The value function Q is continuously updated through the Q-learning algorithm, and the action with the highest value is selected, thereby achieving effective scheduling and load balancing of computing resources.

[0009] In some embodiments of this application, based on the foregoing scheme, the user sends a carbon emission data processing request to the parsing scheduler, which processes the data and the virtual machine, and finally returns the processing result to the client, including: Users send carbon emission analysis and processing requests to the parsing scheduler, specifying the data time range, data type and quantity, processing type (cleaning, summarizing, mining, prediction), and purpose (statistics, analysis, monitoring). Upon receiving the request, the parsing scheduler selects the appropriate compute instance type and quantity based on the request content and Service Level Agreement (SLA), and allocates resources to the request. The parsing scheduler then sends a response to the client, containing information such as the virtual machine's IP address and port for subsequent access and querying. The parsing scheduler employs a load balancing strategy combining weighted round-robin and least-connections algorithms, distributing requests based on the virtual machine's weight ratio while monitoring its load. If a virtual machine's load is too high, the least-connections algorithm distributes requests to the virtual machine with the fewest connections, preventing overloaded machines from handling more requests. When a virtual machine's connection count or IP address changes, the scheduler updates the virtual machine's parameters. The system calculates the number of virtual machines (VMs) to ensure balanced load for subsequent requests. The scheduler starts the corresponding VMs on the cloud computing platform, obtaining information such as VM type, quantity, and specifications. The scheduler maintains a mapping table between service requests and computing instances, recording the computing instance information used by each service request for subsequent resource release and billing operations. Users send data processing tasks to the VMs, including computational requirements, data location, algorithm selection, and parameters. Upon receiving the user's request, the VM selects an appropriate algorithm and data structure based on the content of the request and the complexity of the computational task. For example, if the user request requires large-scale data mining and analysis, the VM selects the distributed computing framework Hadoop to process the data; if the user request requires real-time data processing and analysis, the VM selects the streaming computing engine Flink to process the data and uses the corresponding algorithms and data structures for computation and analysis. The VM returns the processing results to the client, which then performs data processing and usage operations.

[0010] In some embodiments of this application, based on the aforementioned scheme, the calculation of the security score for carbon emission data processing requests prioritizes submitting service content to nodes with high scores for processing. Simultaneously, an ECC encryption algorithm is employed to ensure the secure transmission and storage of sensitive data, including: Data encryption: food and beverage records, transportation records, energy consumption records, carbon reduction action records, carbon trading records, emission points; access control: transportation records, energy consumption records, carbon reduction action records, carbon trading records, emission points; auditing: energy consumption records, carbon reduction action records, carbon trading records, emission points; system user interface, using the system user interface for security control: carbon reduction action records; system isolation: carbon trading records, emission points; The overall security score of the service content is calculated as: Score = weight w1 × attribute value F1 + weight w2 × attribute value F2 + ... + weight wn × attribute value Fn; where the attribute value Fn corresponds to the security mechanism adopted, including data encryption, access control, auditing, system user interface, and system isolation. Different mechanisms are assigned weights wn according to their contribution to ensuring security. The ECC encryption algorithm is used to ensure secure data transmission and storage. Appropriate elliptic curves and base points are selected: P-256, P-384, and P-521. A public key and private key pair are generated using a random number generation algorithm; the public key is a point, and the private key is a number. Plaintext data m is converted into point P, and P is encrypted using the recipient's public key Q to obtain ciphertext C: C = P x Q. The recipient decrypts C using the private key d to obtain plaintext P: P = C x d. Finally, plaintext P is restored to the original data m.

[0011] In some embodiments of this application, based on the foregoing scheme, the dynamic allocation and management of resources for carbon emission data processing requests includes: Collect request types, request nature, and information. Request types include: emission calculation tasks, emission data storage tasks, and emission data transmission tasks. Request nature includes: data volume, sensitivity, importance, and real-time requirements. Information includes: node security configuration and historical security performance. A custom deep neural network is used to learn states and make decisions about actions. The deep neural network is represented as f_θ: S→A, where θ represents the network parameters. The parsing scheduler uses a custom deep reinforcement learning (DRL) to predict the safety performance of nodes. The agent in the reinforcement learning framework is represented as a quadruple (S_t, A_t, R_{t+1}, S_{t+1}). The safety performance of a node is taken as the current state S_t, resource allocation is taken as the action A_t, and the reward obtained by improving safety performance through resource allocation is taken as the reward R_{t+1} obtained by the agent after executing the action. S_{t+1} represents the next state of the agent after executing the action A_t. The goal is to learn a policy π: S→A through interaction with the environment to maximize the safety performance of the system. The required security level is determined based on the task's score and sensitivity. Among the nodes that meet the score requirements, the node with the higher security level and better geographical location is selected for resource allocation. If no node can meet the score requirements at present, the node with the highest security level and lowest latency is selected for temporary execution.

[0012] In some embodiments of this application, based on the foregoing scheme, the sending of processing results utilizes container orchestration and virtualization technologies to dynamically allocate and manage computing resources for carbon emission data processing requests, including: Use communication protocols, including but not limited to HTTP, TCP / IP, and WebSocket; Data on diet, transportation, and energy consumption needs to be returned to users promptly, using the HTTP protocol to enable communication between the client and server; data on carbon reduction actions needs to maintain long-term connections and be transmitted in real time, using the WebSocket protocol for bidirectional communication; while data on carbon trading records and emission points requires large-scale data transmission and processing, using message queue technology for asynchronous communication and task distribution. For the dynamic allocation and management of computing resources, automated technologies such as container orchestration and virtualization are employed. Specifically, container orchestration is used to automatically manage the lifecycle of containers, enabling dynamic allocation and management of computing resources. Simultaneously, virtualization is used to isolate and manage resources, ensuring the security and stability of service content. Security factor = number of virtual machines / number of physical servers; a higher number indicates a higher degree of resource isolation and better security. Stability = actual resources acquired by the virtual machine / configured resources of the virtual machine; the closer this value is to 1, the better the resource acquisition and configuration match, and the higher the stability.

[0013] By combining the management tools and APIs of cloud computing platforms, dynamic allocation and management of computing resources can be achieved, including but not limited to Kubernetes and OpenStac.

[0014] The Kubernetes scheduler can dynamically create and manage Pods (container groups) through the Kubernetes API to dynamically allocate and schedule computing resources: the dynamic allocation capability of computing resources = the number of Pods that Kubernetes can create and destroy per second; Through OpenStack API requests, the scheduler can create and delete virtual machines, thereby enabling dynamic allocation of computing resources: Dynamic allocation capability of computing resources = number of virtual machines that OpenStack can create and destroy per second; The parser scheduler manages containers and pods by calling Kubernetes APIs and virtual machines by calling OpenStack APIs.

[0015] A cloud computing-based information processing system includes: The user submits a service request module, where users submit service requests to the cloud-based information processing system. The service request includes user data, and with the user's permission, the user data is uploaded to the cloud storage system. The load balancer module calculates resource requirements based on the current load status of each node. The parsing scheduler module parses the service requests submitted by users, matches the user requests with the entities and relationships in the knowledge graph, determines the user's needs, and calculates the computing resources, storage resources, and network resources required for the service requests. The carbon emission value function update module uses reinforcement learning algorithms to update the carbon emission value function to achieve the best carbon emission data service effect. The virtual machine processing module allows users to send data processing tasks to the virtual machine, which then returns the processing results to the client. The security mechanism module uses the ECC encryption algorithm to ensure the secure transmission and storage of sensitive data. At the same time, it employs different security mechanisms to protect different service content, including data encryption, access control, auditing, system user interface, and system isolation.

[0016] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows: In the scheme of this application: 1. Integrate knowledge graph and collaborative filtering models to predict the required number of node resources and achieve dynamic resource allocation; 2. Use the Q-learning algorithm to update the carbon emission value function of the load balancer to achieve the best carbon emission data service effect; use security score calculation to prioritize the submission of service content to nodes with high scores for processing, and use ECC encryption algorithm to ensure the secure transmission and storage of sensitive data; 3. A load balancing strategy using weighted round-robin and least connections algorithms is employed to distribute requests based on the weight ratio of virtual machines, while monitoring the load of virtual machines to prevent overloaded virtual machines from handling more requests. 4. The parsing scheduler maintains a mapping table between service requests and compute instances, recording the compute instance information used by each service request for subsequent resource release and billing operations. Attached Figure Description

[0018] Figure 1 A flowchart illustrating a cloud computing-based information processing method according to an embodiment of this application is shown schematically. Figure 2 The illustration shows a schematic diagram of a cloud-based information processing system according to an embodiment of this application. Detailed Implementation

[0019] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided to make this application more comprehensive and complete, and to fully convey the concept of the exemplary embodiments to those skilled in the art.

[0020] Furthermore, the described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. Numerous specific details are provided in the following description to give a thorough understanding of embodiments of this application. However, those skilled in the art will recognize that the technical solutions of this application can be practiced without one or more of the specific details, or other methods, components, apparatuses, steps, etc., can be employed. In other instances, well-known methods, apparatuses, implementations, or operations are not shown or described in detail to avoid obscuring various aspects of this application.

[0021] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0022] The flowcharts shown in the accompanying drawings are merely illustrative and do not necessarily include all content and operations / steps, nor do they necessarily have to be performed in the described order. For example, some operations / steps can be broken down, while others can be combined or partially combined; therefore, the actual execution order may change depending on the specific circumstances.

[0023] Example 1 The implementation details of the technical solutions in the embodiments of this application are described in detail below: Figure 1 A flowchart of a cloud computing-based information processing method according to an embodiment of this application is shown. The cloud computing-based information processing method includes at least steps S101 to S107, which are described in detail below: In step S101, the user submits a service request to the cloud-based information processing system on the client, which includes the user's carbon emission data. The load balancer calculates the resource requirements based on the node load. Users submit service requests to a cloud-based information processing system. These requests include user data. With user permission, the system uploads the user's carbon emission data to the cloud storage system. This data includes, but is not limited to, emission sources: dietary records (types of food, time of food consumption, quantity of food consumed), travel records (driving mileage, public transportation use, walking distance), energy consumption records (household electricity consumption, water consumption, gas consumption); action and transaction information: carbon emission reduction action records (environmental protection measures, emission reduction effects, emission reduction time), carbon trading records (carbon trading projects, trading time, trading volume); and emission points: buildings and locations. The load balancer calculates resource requirements based on the current load of each node: Resource requirements = S × R × T, where S is the size of the service request, R is the processing speed of the service request, and T is the duration of the service request; the larger the amount of carbon emission data submitted by the user, the larger the size of the service request S, and the higher the resource requirements. In step S102, emission sources, action and transaction information and emission points are constructed into a graph. A collaborative filtering model and a knowledge graph-based method are combined to analyze the type and content of carbon emission data service requests and dynamically allocate computing resources, storage resources and network resources. The parsing scheduler parses the service requests submitted by users, using a combination of collaborative filtering models and knowledge graph-based methods to analyze the service request type and content; it constructs a graph of related entities and relationships, with entities including emission sources, action and transaction information, and emission points; and connections between emission sources, action and transaction information, and emission points. The parsing scheduler parses the service requests submitted by users, matches the user requests with entities and relationships in the knowledge graph, and determines the user's needs. It uses a collaborative filtering model to learn and recommend services based on user needs. Based on the results of the collaborative filtering model and the knowledge graph, the parsing scheduler calculates the computing, storage, and network resources required for the service request, and calculates the required number of node resources. When dynamically allocating computing, storage, and network resources, the parsing scheduler dynamically allocates resources based on the node resource requirements and the API calls of the cloud computing platform, including but not limited to requests to create virtual machines and requests for storage resources.

[0024] In one embodiment of this application, the user request includes "household electricity consumption", which is related to "building" and "location" in the knowledge graph: household electricity consumption × the relationship weight between household electricity consumption and building + building × the relationship weight between building and location; The parsing scheduler parses the service requests submitted by users, matches the user requests with entities and relationships in the knowledge graph, and determines the user's needs. It uses a collaborative filtering model to learn and recommend services or resources based on user needs. By learning the user's historical service requests, it discovers the similarity between users and recommends services or resources that the current user and similar users have previously requested: Recommendation degree = Σ (similarity between the user and the current user × the user's previous service requests).

[0025] Based on the results of the collaborative filtering model and knowledge graph, the parsing scheduler calculates the computing, storage, and network resources required for the service request, and calculates the number of node resources required. When dynamically allocating computing, storage, and network resources, the parsing scheduler dynamically allocates resources based on the node resource requirements of the computing and the API calls of the cloud computing platform, including but not limited to requests to create virtual machines and requests for storage resources.

[0026] In step S103, the Q-learning algorithm is used to update the carbon emission value function of the load balancer to achieve the optimal carbon emission data service effect; The Q-learning algorithm is used to update the carbon emission value function, Q(s,a)=(1-α)×Q(s,a)+α×(r+γ×max(Q(s',a'))), where s represents different records of the carbon emission data service; a represents different operations related to carbon emission calculation and analysis; Q(s,a) represents the carbon emission value of taking action a in state s; α is the learning rate of 0.1-0.9, used to control the degree of Q-value update; r is the immediate reward obtained by taking action a in the current state s, which depends on the contribution of action a to the carbon emission data service; γ is a discount factor used to measure the degree of consideration of future rewards in the current decision, and γ is 0.1-0.9; max(Q(s',a')) means taking the action a' with the highest value among all possible actions in the next state s' to maximize the carbon emission value after the state transition. By continuously updating the value function Q through the Q-learning algorithm and selecting the action with the highest value, the efficient scheduling and load balancing of computing resources can be achieved.

[0027] In step S104, the user sends a carbon emission data processing request to the parsing scheduler, which processes the data through the parsing scheduler and the virtual machine, and finally returns the processing result to the client. Users send carbon emission analysis and processing requests to the parsing scheduler, specifying the data time range, data type and quantity, processing type (cleaning, summarizing, mining, prediction), and purpose (statistics, analysis, monitoring). Upon receiving the request, the parsing scheduler selects the appropriate compute instance type and quantity based on the request content and Service Level Agreement (SLA), and allocates resources to the request. The parsing scheduler then sends a response to the client, containing information such as the virtual machine's IP address and port for subsequent access and querying. The parsing scheduler employs a load balancing strategy combining weighted round-robin and least-connections algorithms, distributing requests based on the virtual machine's weight ratio while monitoring its load. If a virtual machine's load is too high, the least-connections algorithm distributes requests to the virtual machine with the fewest connections, preventing overloaded machines from handling more requests. When a virtual machine's connection count or IP address changes, the scheduler updates the virtual machine's parameters. The system calculates the number of virtual machines (VMs) to ensure balanced load for subsequent requests. The scheduler starts the corresponding VMs on the cloud computing platform, obtaining information such as VM type, quantity, and specifications. The scheduler maintains a mapping table between service requests and computing instances, recording the computing instance information used by each service request for subsequent resource release and billing operations. Users send data processing tasks to the VMs, including computational requirements, data location, algorithm selection, and parameters. Upon receiving the user's request, the VM selects an appropriate algorithm and data structure based on the content of the request and the complexity of the computational task. For example, if the user request requires large-scale data mining and analysis, the VM selects the distributed computing framework Hadoop to process the data; if the user request requires real-time data processing and analysis, the VM selects the streaming computing engine Flink to process the data and uses the corresponding algorithms and data structures for computation and analysis. The VM returns the processing results to the client, which then performs data processing and usage operations.

[0028] In step S105, the security score of the carbon emission data processing request is calculated, and the service content is preferentially submitted to the node with the high score for processing. At the same time, the ECC encryption algorithm is used to ensure the secure transmission and storage of sensitive data. Data encryption: food and beverage records, transportation records, energy consumption records, carbon reduction action records, carbon trading records, emission points; access control: transportation records, energy consumption records, carbon reduction action records, carbon trading records, emission points; auditing: energy consumption records, carbon reduction action records, carbon trading records, emission points; system user interface, using the system user interface for security control: carbon reduction action records; system isolation: carbon trading records, emission points; The overall security score of the service content is calculated as: Score = weight w1 × attribute value F1 + weight w2 × attribute value F2 + ... + weight wn × attribute value Fn; where the attribute value Fn corresponds to the security mechanism adopted, which includes data encryption, access control, auditing, system user interface and system isolation. Different mechanisms are assigned weight wn according to their contribution to ensuring security. The ECC encryption algorithm is used to ensure secure data transmission and storage. Suitable elliptic curves and base points are selected: P-256, P-384, and P-521. A public and private key pair is generated using a random number generation algorithm; the public key is a point, and the private key is a number. The plaintext data m is converted into point P, and P is encrypted using the recipient's public key Q to obtain ciphertext C: C = P x Q. The recipient decrypts C using the private key d to obtain plaintext P: P = C x d. Finally, the plaintext P is restored to the original data m.

[0029] In step S106, resources for carbon emission data processing requests are dynamically allocated and managed; Collect request types, request nature, and information. Request types include: emission calculation tasks, emission data storage tasks, and emission data transmission tasks. Request nature includes: data volume, sensitivity, importance, and real-time requirements. Information includes: node security configuration and historical security performance. A custom deep neural network is used to learn states and make decisions about actions. The deep neural network is represented as f_θ: S→A, where θ represents the network parameters. The parsing scheduler uses a custom deep reinforcement learning (DRL) to predict the safety performance of nodes. The agent in the reinforcement learning framework is represented as a quadruple (S_t, A_t, R_{t+1}, S_{t+1}). The safety performance of a node is taken as the current state S_t, resource allocation is taken as the action A_t, and the reward obtained by improving safety performance through resource allocation is taken as the reward R_{t+1} obtained by the agent after executing the action. S_{t+1} represents the next state of the agent after executing the action A_t. The goal is to learn a policy π: S→A through interaction with the environment to maximize the safety performance of the system. The required security level is determined based on the task's score and sensitivity. Among the nodes that meet the score requirements, the node with the higher security level and better geographical location is selected for resource allocation. If no node can meet the score requirements at present, the node with the highest security level and lowest latency is selected for temporary execution.

[0030] In step S107, the processing result is sent, and the computing resources for carbon emission data processing requests are dynamically allocated and managed through container orchestration and virtualization technologies.

[0031] Use communication protocols, including but not limited to HTTP, TCP / IP, and WebSocket; Data on diet, transportation, and energy consumption needs to be returned to users promptly, using the HTTP protocol to enable communication between the client and server; data on carbon reduction actions needs to maintain long-term connections and be transmitted in real time, using the WebSocket protocol for bidirectional communication; while data on carbon trading records and emission points requires large-scale data transmission and processing, using message queue technology for asynchronous communication and task distribution. For the dynamic allocation and management of computing resources, automated technologies such as container orchestration and virtualization are employed. Specifically, container orchestration is used to automatically manage the lifecycle of containers, enabling the dynamic allocation and management of computing resources. At the same time, virtualization is used to achieve resource isolation and management, ensuring the security and stability of service content. By combining the management tools and APIs of cloud computing platforms, dynamic allocation and management of computing resources can be achieved, including but not limited to Kubernetes and OpenStac.

[0032] The Kubernetes scheduler can dynamically create and manage Pods (container groups) through the Kubernetes API to dynamically allocate and schedule computing resources: the dynamic allocation capability of computing resources = the number of Pods that Kubernetes can create and destroy per second; Through OpenStack API requests, the scheduler can create and delete virtual machines, thereby enabling dynamic allocation of computing resources: Dynamic allocation capability of computing resources = number of virtual machines that OpenStack can create and destroy per second; The parser scheduler manages containers and pods by calling Kubernetes APIs and virtual machines by calling OpenStack APIs.

[0033] In practical applications, Using the OpenStack API: Creating a virtual machine: The scheduler calls the OpenStack API to create a virtual machine with 500GB of memory, 10 virtual CPUs, and a 100GB system disk to run the carbon emission calculation application service; virtual machines are chosen instead of containers to ensure data security; Create a floating IP address and security group: Use the OpenStack API to dynamically assign a floating IP address and security group to the virtual machine, and configure it to allow access only from specified IP addresses and ports to ensure network security; Create block storage: Call the OpenStack API to dynamically allocate a 1TB block storage for the virtual machine and persistently store the carbon emission data submitted by the user; Resource expansion: If the user data volume increases to 700GB, call the OpenStack API to expand the virtual machine memory to 700GB, the CPU to 15, and the block storage to 1.5TB to cope with the increase in resource demand; Resource shrinkage: If the user data volume is reduced to 200GB, call the OpenStack API to shrink the virtual machine memory to 256GB, reduce the CPU to 6, and shrink the block storage to 0.5TB, releasing excess resources.

[0034] To run a containerized carbon emissions calculation service using Kubernetes, you can use the Kubernetes API to manage it on a virtual machine as follows: Create a Deployment and start 5 service Pods on the virtual machine; Pod expansion: If the amount of user data increases, expand the number of Pods in the Deployment to 8; Pod shrinkage: If the amount of user data decreases, shrink the number of Pods in the Deployment to 3; Pod restart: If any Pod fails, Kubernetes will automatically restart the Pod to ensure high availability of the service.

[0035] The above methods can ensure the provision of efficient, safe, and reliable carbon emission data processing and services.

[0036] The following describes an apparatus embodiment of this application, which can be used to execute the cloud computing-based information processing method in the above embodiments of this application. It is understood that the apparatus can be a computer program (including program code) running on a computer device, for example, the apparatus is application software; the apparatus can be used to execute the corresponding steps in the method provided in the embodiments of this application. For details not disclosed in the apparatus embodiments of this application, please refer to the embodiments of the cloud computing-based information processing method described above.

[0037] Example 2 Figure 2 A block diagram of a cloud computing-based information processing system according to an embodiment of this application is shown.

[0038] Reference Figure 2 As shown, a cloud computing-based information processing system 200 according to an embodiment of this application includes: User service request module 210: The user submits a service request to the cloud-based information processing system. The service request includes user data. With the user's permission, the user data is uploaded to the cloud storage system. Load balancer module 220: The load balancer calculates the resource requirements based on the current load status of each node. The parsing scheduler module 230 parses the service requests submitted by users, matches the user requests with the entities and relationships in the knowledge graph, determines the user's needs, and calculates the computing resources, storage resources and network resources required for the service requests. The carbon emission value function update module 240 uses a reinforcement learning algorithm to update the carbon emission value function to achieve the best carbon emission data service effect. The virtual machine processing module 250 allows users to send data processing tasks to the virtual machine, and the virtual machine to return the processing results to the client. Security mechanism module 260 uses ECC encryption algorithm to ensure the secure transmission and storage of sensitive data, and adopts different security mechanisms to protect different service content, including data encryption, access control, auditing, system user interface and system isolation.

[0039] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. Each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. Furthermore, each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0040] The units described in the embodiments of this application can be implemented in software or hardware, and the described units can also be located in a processor. The names of these units do not necessarily limit the specific unit itself.

[0041] According to one aspect of this application, a computer program product or computer program is provided, comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the methods provided in the various alternative implementations described above.

[0042] It should be noted that although several modules or units for the device used to perform actions have been mentioned in the detailed description above, this division is not mandatory. In fact, according to the embodiments of this application, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0043] Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, touch terminal, or network device, etc.) to execute the methods according to the embodiments of this application.

[0044] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the embodiments disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein.

[0045] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A cloud computing-based information processing method and system, characterized in that, include: S101. The user submits a service request to the cloud-based information processing system on the client, which includes the user's carbon emission data. The load balancer calculates the resource requirements based on the node load. S102. Construct a graph of emission sources, actions and transactions, and emission points, and use a combination of collaborative filtering models and knowledge graph-based methods to analyze the type and content of carbon emission data service requests, and dynamically allocate computing resources, storage resources, and network resources. S103. Use the Q-learning algorithm to update the carbon emission value function of the load balancer to achieve the best carbon emission data service effect; S104. The user sends a carbon emission data processing request to the parsing scheduler, which processes the data through the parsing scheduler and the virtual machine, and finally returns the processing result to the client. S105. Calculate the security score of carbon emission data processing requests, prioritize submitting service content to nodes with high scores for processing, and use ECC encryption algorithm to ensure the secure transmission and storage of sensitive data. S106. Dynamically allocate and manage resources for carbon emission data processing requests; S107. Send the processing results. The computing resources for carbon emission data processing requests are dynamically allocated and managed through container orchestration and virtualization technologies.

2. The information processing method based on cloud computing according to claim 1, characterized in that, S101 includes: Users submit service requests to the cloud-based information processing system. The service requests include user data. With the user's permission, the user's carbon emission data is uploaded to the cloud storage system. The user's carbon emission data includes emission sources, action and transaction information, and emission points. Emission sources include food records, transportation records, and energy consumption records; food records include food type, time of eating, and quantity of food; transportation records include driving mileage, public transportation usage, and walking distance; energy consumption records include household electricity consumption, water consumption, and gas consumption. Action and Trading Information: Carbon emission reduction action records, including environmental protection measures, emission reduction effects, and emission reduction time; carbon trading records, including carbon trading projects, trading time, and trading volume; Emission point: building, location; The load balancer calculates resource requirements based on the current load of each node: Resource requirement = S × R × T, where S is the size of the service request, R is the processing speed of the service request, and T is the duration of the service request; the larger the amount of carbon emission data submitted by the user, the larger the size of the service request S, and the higher the resource requirement.

3. The information processing method based on cloud computing according to claim 1, characterized in that, S102 includes: The parsing scheduler parses the service requests submitted by users, using a combination of collaborative filtering models and knowledge graph-based methods to analyze the service request type and content; it constructs a graph of related entities and relationships, with entities including emission sources, action and transaction information, and emission points; and connections between emission sources, action and transaction information, and emission points. The parsing scheduler parses the service requests submitted by users, matches the user requests with entities and relationships in the knowledge graph, and determines the user's needs. It uses a collaborative filtering model to learn and recommend services based on user needs. Based on the results of the collaborative filtering model and the knowledge graph, the parsing scheduler calculates the computing, storage, and network resources required for the service request, and calculates the required number of node resources. When dynamically allocating computing, storage, and network resources, the parsing scheduler dynamically allocates resources based on the node resource requirements and the API calls of the cloud computing platform, including but not limited to requests to create virtual machines and requests for storage resources.

4. The information processing method based on cloud computing according to claim 1, characterized in that, S103 includes: The Q-learning algorithm is used to update the carbon emission value function, Q(s,a)=(1-α)×Q(s,a)+α×(r+γ×max(Q(s',a'))), where s represents different records of the carbon emission data service; a represents different operations related to carbon emission calculation and analysis; Q(s,a) represents the carbon emission value of taking action a in state s; α is the learning rate of 0.1-0.9, used to control the degree of Q-value update; r is the immediate reward obtained by taking action a in the current state s, which depends on the contribution of action a to the carbon emission data service; γ is a discount factor used to measure the degree of consideration of future rewards in the current decision, and γ is 0.1-0.9; max(Q(s',a')) means taking the action a' with the highest value among all possible actions in the next state s' to maximize the carbon emission value after the state transition. By continuously updating the value function Q through the Q-learning algorithm and selecting the action with the highest value, the efficient scheduling and load balancing of computing resources can be achieved.

5. The information processing method based on cloud computing according to claim 1, characterized in that, S104 includes: Users send carbon emission analysis and processing requests to the parsing scheduler, including the data time range, data type and quantity, processing type and purpose of use; After receiving a user's request, the parser scheduler selects the appropriate compute instance type and quantity based on the content of the request and the service level agreement (SLA), and allocates resources to the request. The parser scheduler sends a response to the client regarding the processing request. The response contains the virtual machine's IP address and port information for subsequent access and queries by the user. The load balancing strategy of the parser scheduler combines the weighted round-robin algorithm and the least connections algorithm. It distributes requests according to the weight ratio of virtual machines and monitors the load of virtual machines. If the load of a virtual machine is too high, the least connections algorithm is used to distribute requests to the virtual machine with the fewest connections to avoid the overloaded virtual machine handling more requests. When the connection number or IP address of a virtual machine changes, the virtual machine parameters are updated to ensure balanced load for the next request. The parsing scheduler starts the corresponding virtual machines on the cloud computing platform and obtains the specified virtual machine type, quantity, and specification information; the parsing scheduler maintains a mapping table between service requests and computing instances, and records the computing instance information used by each service request for subsequent resource release and billing operations. Users send data processing tasks to the virtual machine, including computational requirements, data location, algorithm selection, and parameters; After receiving a user's request, the virtual machine selects an appropriate algorithm and data structure based on the content of the request and the complexity of the computation task. For example, if the user's request requires large-scale data mining and analysis tasks, the virtual machine selects the distributed computing framework Hadoop to process the data; if the user's request requires real-time data processing and analysis, the virtual machine selects the streaming computing engine Flink to process the data and uses the corresponding algorithms and data structures for computation and analysis. The virtual machine returns the processing results to the client, which then performs data processing and usage operations.

6. The information processing method based on cloud computing according to claim 1, characterized in that, S105 includes: Data encryption: dietary records, transportation records, energy consumption records, carbon reduction action records, carbon trading records, emission points; Access control: travel records, energy consumption records, carbon reduction action records, carbon trading records, emission points; Audit: Energy consumption records, carbon reduction action records, carbon trading records, emission points; The system user interface is used for security control: carbon emission reduction action recording. System isolation: carbon trading records, emission points; The security score for carbon emission data processing requests is calculated as: Score = weight w1 × attribute value F1 + weight w2 × attribute value F2 + ... + weight wn × attribute value Fn; where the attribute value Fn corresponds to the security mechanism adopted, which includes data encryption, access control, auditing, system user interface, and system isolation. Different mechanisms are assigned weight wn according to their contribution to ensuring security. The ECC encryption algorithm is used to ensure secure data transmission and storage; suitable elliptic curves and base points are selected, P-256, P-384, and P-521; a public key and a private key pair are generated using a random number generation algorithm, where the public key is a point and the private key is a number; The plaintext data m is converted into point P, which is then encrypted using the receiver's public key Q to obtain ciphertext C: C = P x Q. The receiver decrypts C using their private key d to obtain plaintext P: P = C x d. Finally, the plaintext P is restored to the original data m.

7. The information processing method based on cloud computing according to claim 1, characterized in that, S106 includes: Collect request types, request nature, and information. Request types include: emission calculation tasks, emission data storage tasks, and emission data transmission tasks. Request nature includes: data volume, sensitivity, importance, and real-time requirements. Information includes: node security configuration and historical security performance. A custom deep neural network is used to learn states and make decisions about behaviors. The deep neural network is represented as f_θ:S→A, where θ represents the network parameters. The parsing scheduler uses a custom deep reinforcement learning (DRL) to predict node safety performance, representing the agent in the reinforcement learning framework as a quadruple (S_t, A_t, R_{t+1}, S_{t+1}). The security performance of a node is taken as the current state S_t, resource allocation is taken as the action A_t, and the reward obtained by improving security performance through resource allocation is taken as the reward R_{t+1} obtained by the agent after executing the action. S_{t+1} represents the next state of the agent after executing the action A_t. The goal is to learn a policy π: S→A through interaction with the environment, so as to maximize the security performance of the system. The required security level is determined based on the task's score and sensitivity. Among the nodes that meet the score requirements, the node with the higher security level and better geographical location is selected for resource allocation. If no node can meet the score requirements at present, the node with the highest security level and lowest latency is selected for temporary execution.

8. The information processing method based on cloud computing according to claim 1, characterized in that, S107 includes: Use communication protocols, including HTTP, TCP / IP, and WebSocket; Data on diet, transportation, and energy consumption needs to be returned to users promptly, using the HTTP protocol to enable communication between the client and server; data on carbon reduction actions needs to maintain long-term connections and be transmitted in real time, using the WebSocket protocol for bidirectional communication; while data on carbon trading records and emission points requires large-scale data transmission and processing, using message queue technology for asynchronous communication and task distribution. For the dynamic allocation and management of computing resources, automation technologies such as container orchestration and virtualization are employed. Specifically, container orchestration is used to automatically manage the lifecycle of containers, enabling dynamic allocation and management of computing resources. At the same time, virtualization is used to achieve resource isolation and management, ensuring the security and stability of service content. It combines the management tools and APIs of cloud computing platforms to achieve dynamic allocation and management of computing resources, including Kubernetes and OpenStac.

9. An information processing system based on cloud computing, characterized in that, include: The user submits a service request module, where users submit service requests to the cloud-based information processing system. The service request includes user data, and with the user's permission, the user data is uploaded to the cloud storage system. The load balancer module calculates resource requirements based on the current load status of each node. The parsing scheduler module parses the service requests submitted by users, matches the user requests with the entities and relationships in the knowledge graph, determines the user's needs, and calculates the computing resources, storage resources, and network resources required for the service requests. The carbon emission value function update module uses reinforcement learning algorithms to update the carbon emission value function to achieve the best carbon emission data service effect. The virtual machine processing module allows users to send data processing tasks to the virtual machine, which then returns the processing results to the client. The security mechanism module uses the ECC encryption algorithm to ensure the secure transmission and storage of sensitive data. It also employs different security mechanisms to protect different service content, including data encryption, access control, auditing, system user interface, and system isolation.

Images