Cloud computing network-based user short-time demand task scheduling optimization method
By adopting cloud computing optimization methods in short-term demand task scheduling, including requirements analysis, resource monitoring, task preprocessing and AES-CTR encryption and other technologies, the problem of insufficient security of user privacy data is solved, and efficient and secure tasks are achieved Scheduling and resource management improve system stability and user satisfaction.
Patent Information
- Application Number
- CN202411032006.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-30
- Publication Date
- 2024-11-26
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing short-term demand task scheduling methods fail to effectively protect user privacy data, resulting in insufficient security.
Adopt cloud computing-based short-term task scheduling optimization methods, including demand analysis, resource monitoring and evaluation, scheduling strategy design, task preprocessing, task scheduling and execution, result feedback and optimization, security and privacy protection, continuous monitoring and maintenance , ensure the security of user data through AES-CTR encryption, and implement strict access control and resource management.
It improves the efficiency and security of task scheduling, ensures the privacy protection of user data during transmission and storage, prevents unauthorized access and data leakage, enhances the stability and resource utilization of the system, and reduces operating costs.
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of short-term demand task scheduling, and specifically is a method for optimizing short-term demand task scheduling based on cloud computing network users. Background Art
[0002] Short-term task scheduling is a method for effectively allocating and scheduling a large number of tasks that appear in a short period of time. In modern society, with the rapid development of technologies such as artificial intelligence, the Internet of Things, and big data, it is becoming more and more common for a large number of tasks to appear in a short period of time, such as order processing during e-commerce promotions and rescue task allocation after natural disasters. Short-term task scheduling aims to solve these sudden task allocation problems and improve task completion efficiency and resource utilization. Short-term task scheduling needs to consider multiple factors such as the urgency of the task, execution time, and resource requirements, and complete the task in the shortest possible time. In practical applications, short-term task scheduling can adopt a variety of methods, such as rule-based scheduling algorithms, heuristic-based scheduling algorithms, and machine learning-based scheduling algorithms. These algorithms can be selected and optimized according to different task characteristics and requirements.
[0003] However, common optimization methods do not set up a privacy protection module during use, which makes it easy for relevant user privacy data to be leaked, resulting in insufficient security during use. Summary of the invention
[0004] The purpose of the present invention is to provide a method for optimizing task scheduling based on short-term demand of cloud computing network users in order to solve the above-mentioned problems.
[0005] The technical solution adopted by the present invention is as follows: a method for optimizing task scheduling based on short-term demand of cloud computing network users, comprising the following steps:
[0006] S1: Conduct demand analysis. First, analyze the characteristics of users' short-term needs through user surveys or historical data, such as computing volume, data transmission volume, time sensitivity, etc. Then classify user needs according to type, scale and urgency to facilitate the formulation of subsequent scheduling strategies;
[0007] S2: Perform resource monitoring and evaluation. Real-time monitoring refers to using the monitoring tools provided by the cloud computing platform to monitor resource usage in real time, including CPU usage, memory usage, network bandwidth, etc. Resource evaluation refers to evaluating the status of currently available resources based on monitoring data and predicting resource load conditions in the future;
[0008] S3: Design the scheduling strategy. First, design the goal and clarify the goal of task scheduling, such as minimizing task response time and maximizing resource utilization.
[0009] S4: Perform task preprocessing, first decompose the task into small tasks to improve the flexibility and efficiency of scheduling. Then preprocess the data required for the task, such as compression and encryption, to reduce data transmission time and improve security;
[0010] S5: Perform task scheduling and execution. First, assign tasks to appropriate resources according to the scheduling strategy through task allocation. Then, monitor the running status of tasks during task execution to ensure that tasks can be completed on time. Dynamically adjust task allocation according to resource status and task execution status to improve scheduling efficiency.
[0011] S6: Provide feedback and optimization, collect task execution results and user feedback to evaluate the effectiveness of the scheduling strategy, and optimize the scheduling strategy based on the feedback results to improve the effectiveness of future task scheduling;
[0012] S7: Perform security and privacy protection to ensure the security and privacy of user data during task scheduling. And implement strict access control policies to prevent unauthorized access and resource abuse;
[0013] S8: Carry out continuous monitoring and maintenance, continuously monitor the cloud computing environment and changes in user needs to identify potential problems. Regularly maintain and upgrade the scheduling system to keep it running efficiently and stably.
[0014] In a preferred embodiment, the step S1 specifically includes the following steps:
[0015] S1. Data collection: Conduct user surveys to collect information about users’ computing task requirements. Analyze historical data and extract task features, such as task execution time, resource consumption, etc.
[0016] S2. Extraction of demand characteristics: Based on the collected data, determine the main characteristic parameters of the demand. The characteristic parameters may include the amount of calculation (such as the number of CPU cycles), the amount of data transmission (such as the size of the data), time sensitivity (such as the deadline of the task), etc.
[0017] S3. Demand classification: Classify the demands into different categories based on characteristic parameters. Categories can be defined based on task type (e.g. batch processing, real-time processing), task size (e.g. small, medium, large), and urgency (e.g. low, medium, high).
[0018] S4. Demand quantification: Each demand category is quantified to facilitate mathematical modeling of scheduling strategies. For example, the amount of computation can be quantified as CPU hours, the amount of data transferred can be quantified as megabytes (MB), and the time sensitivity can be quantified as relative deadlines (such as the maximum delay that a task can tolerate).
[0019] S5. Prioritize requirements: Prioritize requirements based on their importance and urgency. You can use a weighted approach to prioritize, for example, giving higher weights to time-sensitive tasks.
[0020] In a preferred embodiment, the step S2 specifically includes:
[0021] Configure monitoring parameters: define the resource parameters that need to be monitored, such as CPU usage, memory usage, disk space, network bandwidth, etc. Set the monitoring frequency and determine the time interval for collecting data.
[0022] Implement monitoring: Deploy monitoring tools on the cloud computing platform and start collecting resource usage data. Ensure the accuracy and real-time nature of the monitoring data.
[0023] Data storage and analysis: Store the collected monitoring data in a database or data warehouse. Use data analysis tools to visualize the data to understand resource usage trends.
[0024] Resource evaluation: Evaluate the current resource status based on monitoring data, including resource idle rate, saturation, and performance bottlenecks. Predict resource load in the future to provide a basis for scheduling strategies.
[0025] In a preferred embodiment, in step S3, it is also necessary to select or design a suitable scheduling algorithm according to the design goal, such as polling scheduling, minimum connection scheduling, priority-based scheduling, etc. Then, a dynamic adjustment mechanism is designed to automatically adjust the scheduling strategy when the resource status changes.
[0026] In a preferred embodiment, in step S4, the specific process includes:
[0027] A. Task decomposition: Task analysis: Analyze the composition of large tasks and determine the independent parts that can be decomposed. Evaluate the dependencies between tasks and determine which parts can be processed in parallel.
[0028] Divide tasks: Divide large tasks into multiple small tasks based on the analysis results. Make sure each small task is a manageable and executable unit.
[0029] Define interfaces: Define clear input and output interfaces for each small task. Ensure that data and results can be exchanged between tasks through standardized interfaces.
[0030] Task scheduling: Put the decomposed small tasks into the scheduling queue and prepare for scheduling. Select the appropriate scheduling strategy based on the characteristics of the task and resource conditions.
[0031] B. Data preprocessing:
[0032] Data Identification: Determine the type and amount of data required for the task. Identify sensitive data such as personally identifiable information and trade secrets.
[0033] Data compression: compress non-sensitive data to reduce the size of data transmission. Choose an appropriate compression algorithm, such as ZIP, GZIP, Snappy, etc.
[0034] Data encryption: Encrypt sensitive data to protect data security. Select the appropriate AES encryption algorithm;
[0035] Data transmission optimization: Optimize data transmission protocols to reduce data transmission time. Use data caching and pre-fetching technology to increase data access speed.
[0036] Data Verification: Data integrity verification is performed before and after data transmission to ensure that the data is not damaged during transmission. Verification is performed using technologies such as checksums and digital signatures.
[0037] The flexibility and efficiency of task scheduling can be significantly improved through task decomposition and data preprocessing. Task decomposition allows a large number of tasks to be executed in parallel on different resources, thereby shortening the overall execution time. Data preprocessing reduces the overhead of data transmission, improves the efficiency of data processing, and ensures data security. These steps lay the foundation for the subsequent task scheduling and execution stages.
[0038] In a preferred embodiment, in step S5, the specific process method includes:
[0039] A. Task allocation:
[0040] Scheduling strategy selection: Select an appropriate task scheduling strategy based on the results of demand analysis and resource evaluation. Possible strategies include shortest job first, highest response ratio first, priority scheduling, and round-robin scheduling.
[0041] Resource matching: Match the most appropriate resources based on the resource requirements of the task (such as CPU, memory, storage, network bandwidth) and resource monitoring data. Consider resource utilization, performance, location and other factors for matching.
[0042] Task mapping: Map tasks to selected resources and create task instances. Set the environment and necessary parameters for task execution.
[0043] Task start: Start the task instance and start executing the task. Ensure that the task is started correctly on the resources and record the start time of the task.
[0044] B.Execution Monitoring:
[0045] Status tracking: Real-time monitoring of the running status of tasks, including progress, resource usage, performance indicators, etc. Use monitoring tools to collect data during task execution.
[0046] Anomaly detection: Detect anomalies in task execution, such as insufficient resources, execution errors, deadlocks, etc. Set up an alarm mechanism to notify the system administrator immediately once an anomaly is detected.
[0047] Performance analysis: Analyze the performance data of task execution to evaluate whether the task is performed as expected. Compare actual performance with expected performance to determine whether resource allocation needs to be adjusted.
[0048] C. Dynamic Adjustment: Resource reallocation: Dynamically adjust task allocation based on real-time resource status and task execution status. Migrate tasks from resources with higher loads to resources with lower loads to balance the load. Priority adjustment: Adjust the priority of tasks based on their importance and urgency. It may be necessary to rearrange the order of tasks in the task queue.
[0049] Scheduling strategy optimization: Optimize the scheduling strategy based on historical data and feedback from task execution. Regularly evaluate and adjust the scheduling algorithm to improve scheduling efficiency.
[0050] Through task allocation, execution monitoring, and dynamic adjustment, the task scheduling system can ensure the efficient execution of tasks on cloud computing resources. These steps help improve resource utilization, shorten task response time, and ensure stable system operation.
[0051] In a preferred embodiment, in step S6, the specific process method is:
[0052] A. Result Collection:
[0053] Execution result record: record the results of each task execution, including task completion time, resource consumption, output data, etc. Collect system performance indicators such as response time, throughput, resource utilization, etc.
[0054] User feedback acquisition: Collect user satisfaction with task execution through questionnaires, user interviews, online feedback, etc.
[0055] Get direct feedback and suggestions from users on task scheduling strategies.
[0056] B. Analysis and Evaluation:
[0057] Performance analysis: Analyze task execution results and system performance indicators to evaluate the effectiveness of scheduling strategies and identify bottlenecks and problems in the scheduling process.
[0058] User satisfaction evaluation: Evaluate user satisfaction with task execution based on user feedback.
[0059] Analyze the problems and needs mentioned in user feedback.
[0060] C. Strategy Optimization:
[0061] Problem diagnosis: Diagnose problems in the scheduling strategy based on the analysis results. Determine whether the problem is caused by improper resource allocation, inappropriate scheduling algorithm, or other reasons.
[0062] Policy adjustment: Adjust the scheduling policy based on the diagnosis results, such as modifying algorithm parameters, adjusting resource allocation rules, etc. It may be necessary to redesign the scheduling algorithm or introduce a new scheduling mechanism.
[0063] Optimization implementation: Deploy the optimized scheduling strategy to the task scheduling system. Monitor the effect of the strategy adjustment to ensure that the improvement is positive.
[0064] Continuous iteration: Take result feedback and optimization as an ongoing process, collect data and analyze results regularly, and continue to optimize scheduling strategies based on new data and feedback.
[0065] In a preferred embodiment, in step S7, an AES-CTR variant is selected to ensure the security and privacy of user data during the task scheduling process. The encryption process of the AES-CTR variant is:
[0066] Choose an initial value IV (usually a random number). Encrypt the IV using AES to get the first ciphertext block. XOR the plaintext with the ciphertext block to get the ciphertext. Increment the IV (can be a counter or other method) and repeat the above steps. The decryption process of the AES-CTR variant is: Use the same IV and key to encrypt the ciphertext block using AES. XOR the ciphertext with the ciphertext block to get the plaintext. Increment the IV and repeat the above steps.
[0067] In a preferred embodiment, in step S7, the encryption process of the AES-GCM variant is:
[0068] S1. Select an initial value IV.
[0069] S2. Use AES to encrypt the plaintext to obtain the ciphertext.
[0070] S3. Calculate the GCM tag (GMAC) for data integrity verification.
[0071] S4. Transmit the IV, ciphertext and GCM tag together.
[0072] The decryption process of S5.AES-GCM variant is:
[0073] S6. Separate the IV, ciphertext, and GCM tag from the received data.
[0074] S7. Use AES to decrypt the ciphertext to obtain the plaintext.
[0075] S8. Recalculate the GCM tag and compare it with the received GCM tag to ensure data integrity.
[0076] In step S4, the AES encryption formula is:
[0077] C=AES(K,P)
[0078] Among them, C is the ciphertext, K is the key, and P is the plaintext.
[0079] The AES decryption formula is:
[0080] P=AES^(-1)(K,C)
[0081] Among them, P is plaintext, K is the key, and C is the ciphertext;
[0082] The formula for the XOR operation in the AES-CTR variant is:
[0083] C_i=P_iXORAES(K,IV_i)
[0084] Where C_i is the i-th ciphertext block, P_i is the i-th plaintext block, and IV_i is the i-th IV value.
[0085] The GMAC calculation formula for the AES-GCM variant is:
[0086] GMAC=AES_GCM_MAC(K,A,C)
[0087] Where GMAC is the GCM tag, K is the key, A is the Additional Authentication Data (AAD), and C is the ciphertext.
[0088] In a preferred embodiment, in step S8, the maintenance work includes regularly upgrading the task scheduling system, patching security holes, improving the stability and performance of the system, and regularly backing up the configuration data and monitoring data of the task scheduling system, and formulating a disaster recovery plan. Performance optimization is also an important part of continuous monitoring and maintenance. By regularly evaluating the performance of the system, identifying and solving performance bottlenecks, optimizing database queries, caching strategies and system architecture, the system response speed can be improved.
[0089] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:
[0090] 1. In the present invention, AES-CTR uses an initial value IV and a counter. Each encryption operation generates a new ciphertext block. Even if two plaintext blocks are the same, different ciphertext blocks will be generated, which increases the security of encryption because the diversity of ciphertext blocks makes it difficult for attackers to analyze plaintext through ciphertext.
[0091] Since AES-CTR uses a counter as part of the key, different counters can encrypt data independently of other counters, which allows multiple tasks to be encrypted in parallel, thereby improving the efficiency of encryption and the flexibility of task scheduling. AES-CTR is relatively simple and does not require complex settings. It only requires a random IV and a counter, which makes it relatively easy to integrate AES-CTR encryption in a task scheduling system.
[0092] Privacy protection: By using AES-CTR to encrypt task data, the privacy of user data during transmission and storage can be ensured to prevent unauthorized access and data leakage. Due to the high efficiency of AES-CTR encryption and decryption operations, it can reduce the use of computing resources while maintaining data security, which is an important advantage for task scheduling systems because cloud computing resources can be used more efficiently.
[0093] 2. In the present invention, the design of the task scheduling strategy and the implementation of execution monitoring ensure that the tasks can be responded to and completed quickly, thereby shortening the execution time of the tasks. The task decomposition and dynamic adjustment mechanism enable the system to flexibly respond to different types of tasks and resource changes, thereby improving the adaptability and flexibility of the system. Continuous monitoring and maintenance ensure the stable operation of the system, reduce the risk of failure and performance degradation, and improve the reliability and stability of the system. By optimizing the scheduling strategy and performance analysis, the method can provide more efficient and stable services, thereby improving user satisfaction and trust. The method can make rapid adjustments according to changes in user needs and resource conditions, and support the expansion and adaptability of the system. By optimizing resource allocation and improving resource utilization, the method can reduce unnecessary resource consumption and reduce operating costs. DETAILED DESCRIPTION
[0094] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.
[0095] Example:
[0096] A method for optimizing task scheduling based on short-term demand of cloud computing network users: the method comprises the following steps:
[0097] S1: Conduct demand analysis. First, analyze the characteristics of users' short-term needs through user surveys or historical data, such as computing volume, data transmission volume, time sensitivity, etc. Then classify user needs according to type, scale and urgency to facilitate the formulation of subsequent scheduling strategies;
[0098] S2: Perform resource monitoring and evaluation. Real-time monitoring refers to using the monitoring tools provided by the cloud computing platform to monitor resource usage in real time, including CPU usage, memory usage, network bandwidth, etc. Resource evaluation refers to evaluating the status of currently available resources based on monitoring data and predicting resource load conditions in the future;
[0099] S3: Design the scheduling strategy. First, design the goal and clarify the goal of task scheduling, such as minimizing task response time and maximizing resource utilization.
[0100] S4: Perform task preprocessing, first decompose the task into small tasks to improve the flexibility and efficiency of scheduling. Then preprocess the data required for the task, such as compression and encryption, to reduce data transmission time and improve security;
[0101] S5: Perform task scheduling and execution. First, assign tasks to appropriate resources according to the scheduling strategy through task allocation. Then, monitor the running status of tasks during task execution to ensure that tasks can be completed on time. Dynamically adjust task allocation according to resource status and task execution status to improve scheduling efficiency.
[0102] S6: Provide feedback and optimization, collect task execution results and user feedback to evaluate the effectiveness of the scheduling strategy, and optimize the scheduling strategy based on the feedback results to improve the effectiveness of future task scheduling;
[0103] S7: Perform security and privacy protection to ensure the security and privacy of user data during task scheduling. And implement strict access control policies to prevent unauthorized access and resource abuse;
[0104] S8: Carry out continuous monitoring and maintenance, continuously monitor the cloud computing environment and changes in user needs to identify potential problems. Regularly maintain and upgrade the scheduling system to keep it running efficiently and stably.
[0105] Step S1 specifically includes the following steps:
[0106] S1. Data collection: Conduct user surveys to collect information about users’ computing task requirements. Analyze historical data and extract task features, such as task execution time, resource consumption, etc.
[0107] S2. Extraction of demand characteristics: Based on the collected data, determine the main characteristic parameters of the demand. The characteristic parameters may include the amount of calculation (such as the number of CPU cycles), the amount of data transmission (such as the size of the data), time sensitivity (such as the deadline of the task), etc.
[0108] S3. Demand classification: Classify the demands into different categories based on characteristic parameters. Categories can be defined based on task type (e.g. batch processing, real-time processing), task size (e.g. small, medium, large), and urgency (e.g. low, medium, high).
[0109] S4. Demand quantification: Each demand category is quantified to facilitate mathematical modeling of scheduling strategies. For example, the amount of computation can be quantified as CPU hours, the amount of data transferred can be quantified as megabytes (MB), and the time sensitivity can be quantified as relative deadlines (such as the maximum delay that a task can tolerate).
[0110] S5. Prioritize requirements: Prioritize requirements based on their importance and urgency. You can use a weighted approach to prioritize, for example, giving higher weights to time-sensitive tasks.
[0111] Step S2 specifically includes:
[0112] Configure monitoring parameters: define the resource parameters that need to be monitored, such as CPU usage, memory usage, disk space, network bandwidth, etc. Set the monitoring frequency and determine the time interval for collecting data.
[0113] Implement monitoring: Deploy monitoring tools on the cloud computing platform and start collecting resource usage data. Ensure the accuracy and real-time nature of the monitoring data.
[0114] Data storage and analysis: Store the collected monitoring data in a database or data warehouse. Use data analysis tools to visualize the data to understand resource usage trends.
[0115] Resource evaluation: Evaluate the current resource status based on monitoring data, including resource idle rate, saturation, and performance bottlenecks. Predict resource load in the future to provide a basis for scheduling strategies.
[0116] In step S3, it is also necessary to select or design a suitable scheduling algorithm according to the design goal, such as polling scheduling, minimum connection scheduling, priority-based scheduling, etc. Then, a dynamic adjustment mechanism is designed to automatically adjust the scheduling strategy when the resource status changes.
[0117] In step S4, the specific process includes:
[0118] A. Task decomposition: Task analysis: Analyze the composition of large tasks and determine the independent parts that can be decomposed. Evaluate the dependencies between tasks and determine which parts can be processed in parallel.
[0119] Divide tasks: Divide large tasks into multiple small tasks based on the analysis results. Make sure each small task is a manageable and executable unit.
[0120] Define interfaces: Define clear input and output interfaces for each small task. Ensure that data and results can be exchanged between tasks through standardized interfaces.
[0121] Task scheduling: Put the decomposed small tasks into the scheduling queue and prepare for scheduling. Select the appropriate scheduling strategy based on the characteristics of the task and resource conditions.
[0122] B. Data preprocessing:
[0123] Data Identification: Determine the type and amount of data required for the task. Identify sensitive data such as personally identifiable information and trade secrets.
[0124] Data compression: compress non-sensitive data to reduce the size of data transmission. Choose an appropriate compression algorithm, such as ZIP, GZIP, Snappy, etc.
[0125] Data encryption: Encrypt sensitive data to protect data security. Select the appropriate AES encryption algorithm;
[0126] Data transmission optimization: Optimize data transmission protocols to reduce data transmission time. Use data caching and pre-fetching technology to increase data access speed.
[0127] Data Verification: Data integrity verification is performed before and after data transmission to ensure that the data is not damaged during transmission. Verification is performed using technologies such as checksums and digital signatures.
[0128] The flexibility and efficiency of task scheduling can be significantly improved through task decomposition and data preprocessing. Task decomposition allows a large number of tasks to be executed in parallel on different resources, thereby shortening the overall execution time. Data preprocessing reduces the overhead of data transmission, improves the efficiency of data processing, and ensures data security. These steps lay the foundation for the subsequent task scheduling and execution stages.
[0129] In step S5, the specific process method includes:
[0130] A. Task allocation:
[0131] Scheduling strategy selection: Select an appropriate task scheduling strategy based on the results of demand analysis and resource evaluation. Possible strategies include shortest job first, highest response ratio first, priority scheduling, and round-robin scheduling.
[0132] Resource matching: Match the most appropriate resources based on the resource requirements of the task (such as CPU, memory, storage, network bandwidth) and resource monitoring data. Consider resource utilization, performance, location and other factors for matching.
[0133] Task mapping: Map tasks to selected resources and create task instances. Set the environment and necessary parameters for task execution.
[0134] Task start: Start the task instance and start executing the task. Ensure that the task is started correctly on the resources and record the start time of the task.
[0135] B.Execution monitoring:
[0136] Status tracking: Real-time monitoring of the running status of tasks, including progress, resource usage, performance indicators, etc. Use monitoring tools to collect data during task execution.
[0137] Anomaly detection: Detect anomalies in task execution, such as insufficient resources, execution errors, deadlocks, etc. Set up an alarm mechanism to notify the system administrator immediately once an anomaly is detected.
[0138] Performance analysis: Analyze the performance data of task execution to evaluate whether the task is performed as expected. Compare actual performance with expected performance to determine whether resource allocation needs to be adjusted.
[0139] C. Dynamic Adjustment: Resource reallocation: Dynamically adjust task allocation based on real-time resource status and task execution status. Migrate tasks from resources with higher loads to resources with lower loads to balance the load. Priority adjustment: Adjust the priority of tasks based on their importance and urgency. It may be necessary to rearrange the order of tasks in the task queue.
[0140] Scheduling strategy optimization: Optimize the scheduling strategy based on historical data and feedback from task execution. Regularly evaluate and adjust the scheduling algorithm to improve scheduling efficiency.
[0141] Through task allocation, execution monitoring, and dynamic adjustment, the task scheduling system can ensure the efficient execution of tasks on cloud computing resources. These steps help improve resource utilization, shorten task response time, and ensure stable system operation.
[0142] In step S6, the specific process method is:
[0143] A. Result Collection:
[0144] Execution result record: record the results of each task execution, including task completion time, resource consumption, output data, etc. Collect system performance indicators such as response time, throughput, resource utilization, etc.
[0145] User feedback acquisition: Collect user satisfaction with task execution through questionnaires, user interviews, online feedback, etc.
[0146] Get direct feedback and suggestions from users on task scheduling strategies.
[0147] B. Analysis and Evaluation:
[0148] Performance analysis: Analyze task execution results and system performance indicators to evaluate the effectiveness of scheduling strategies and identify bottlenecks and problems in the scheduling process.
[0149] User satisfaction evaluation: Evaluate user satisfaction with task execution based on user feedback.
[0150] Analyze the problems and needs mentioned in user feedback.
[0151] C. Strategy Optimization:
[0152] Problem diagnosis: Diagnose problems in the scheduling strategy based on the analysis results. Determine whether the problem is caused by improper resource allocation, inappropriate scheduling algorithm, or other reasons.
[0153] Policy adjustment: Adjust the scheduling policy based on the diagnosis results, such as modifying algorithm parameters, adjusting resource allocation rules, etc. It may be necessary to redesign the scheduling algorithm or introduce a new scheduling mechanism.
[0154] Optimization implementation: Deploy the optimized scheduling strategy to the task scheduling system. Monitor the effect of the strategy adjustment to ensure that the improvement is positive.
[0155] Continuous iteration: Take result feedback and optimization as an ongoing process, collect data and analyze results regularly, and continue to optimize scheduling strategies based on new data and feedback.
[0156] In step S7, the AES-CTR variant is selected to ensure the security and privacy of user data during the task scheduling process. The encryption process of the AES-CTR variant is:
[0157] Choose an initial value IV (usually a random number). Encrypt the IV using AES to get the first ciphertext block. XOR the plaintext with the ciphertext block to get the ciphertext. Increment the IV (can be a counter or other method) and repeat the above steps. The decryption process of the AES-CTR variant is: Use the same IV and key to encrypt the ciphertext block using AES. XOR the ciphertext with the ciphertext block to get the plaintext. Increment the IV and repeat the above steps.
[0158] In step S7, the encryption process of the AES-GCM variant is:
[0159] S1. Select an initial value IV.
[0160] S2. Use AES to encrypt the plaintext to obtain the ciphertext.
[0161] S3. Calculate the GCM tag (GMAC) for data integrity verification.
[0162] S4. Transmit the IV, ciphertext and GCM tag together.
[0163] The decryption process of S5.AES-GCM variant is:
[0164] S6. Separate the IV, ciphertext, and GCM tag from the received data.
[0165] S7. Use AES to decrypt the ciphertext to obtain the plaintext.
[0166] S8. Recalculate the GCM tag and compare it with the received GCM tag to ensure data integrity.
[0167] In step S4, the AES encryption formula is:
[0168] C=AES(K,P)
[0169] Among them, C is the ciphertext, K is the key, and P is the plaintext.
[0170] The AES decryption formula is:
[0171] P=AES^(-1)(K,C)
[0172] Among them, P is plaintext, K is the key, and C is the ciphertext;
[0173] The formula for the XOR operation in the AES-CTR variant is:
[0174] C_i=P_iXORAES(K,IV_i)
[0175] Where C_i is the i-th ciphertext block, P_i is the i-th plaintext block, and IV_i is the i-th IV value.
[0176] The GMAC calculation formula for the AES-GCM variant is:
[0177] GMAC=AES_GCM_MAC(K,A,C)
[0178] Where GMAC is the GCM tag, K is the key, A is the Additional Authentication Data (AAD), and C is the ciphertext.
[0179] In step S8, maintenance work includes regularly upgrading the task scheduling system, patching security vulnerabilities, improving system stability and performance, and regularly backing up configuration data and monitoring data of the task scheduling system, and formulating a disaster recovery plan. Performance optimization is also an important part of continuous monitoring and maintenance. By regularly evaluating the performance of the system, identifying and solving performance bottlenecks, and optimizing database queries, cache strategies, and system architecture, the system response speed can be improved.
[0180] In the present invention, AES-CTR uses an initial value IV and a counter, and each encryption operation generates a new ciphertext block. Even if two plaintext blocks are the same, different ciphertext blocks will be generated, which increases the security of encryption because the diversity of ciphertext blocks makes it difficult for attackers to analyze plaintext through ciphertext.
[0181] Since AES-CTR uses a counter as part of the key, different counters can encrypt data independently of other counters, which allows multiple tasks to be encrypted in parallel, thereby improving the efficiency of encryption and the flexibility of task scheduling. AES-CTR is relatively simple and does not require complex settings. It only requires a random IV and a counter, which makes it relatively easy to integrate AES-CTR encryption in a task scheduling system.
[0182] Privacy protection: By using AES-CTR to encrypt task data, the privacy of user data during transmission and storage can be ensured to prevent unauthorized access and data leakage. Due to the high efficiency of AES-CTR encryption and decryption operations, it can reduce the use of computing resources while maintaining data security, which is an important advantage for task scheduling systems because cloud computing resources can be used more efficiently.
[0183] In the present invention, the design of task scheduling strategy and the implementation of execution monitoring ensure that tasks can be responded to and completed quickly, thereby shortening the execution time of tasks. Task decomposition and dynamic adjustment mechanism enable the system to flexibly respond to different types of tasks and resource changes, thereby improving the adaptability and flexibility of the system. Continuous monitoring and maintenance ensure the stable operation of the system, reduce the risk of failure and performance degradation, and improve the reliability and stability of the system. By optimizing scheduling strategies and performance analysis, the method can provide more efficient and stable services, thereby improving user satisfaction and trust. The method can make rapid adjustments according to changes in user needs and resource conditions, and support the expansion and adaptability of the system. By optimizing resource allocation and improving resource utilization, the method can reduce unnecessary resource consumption and reduce operating costs.
[0184] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, the essence of these modified or replaced technical solutions deviates from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A method for optimizing task scheduling based on short-term demand of cloud computing network users, characterized by: The steps include: S1: Conduct demand analysis. First, analyze the characteristics of users' short-term demands, such as computing volume, data transmission volume, and time sensitivity, through user surveys or historical data. Then classify user demands by type, scale, and urgency to facilitate the formulation of subsequent scheduling strategies. S2: Perform resource monitoring and evaluation. Real-time monitoring refers to using the monitoring tools provided by the cloud computing platform to monitor resource usage in real time, including CPU usage, memory usage, and network bandwidth. Resource evaluation refers to evaluating the status of currently available resources based on monitoring data and predicting resource load conditions in the future. S3: Design the scheduling strategy. First, design the goal and clarify the goal of task scheduling, such as minimizing task response time and maximizing resource utilization. S4: Perform task preprocessing. First, perform task decomposition, breaking down large tasks into small tasks to improve scheduling flexibility and efficiency. Then, perform preprocessing on the data required for the task, such as compression and encryption, to reduce data transmission time and improve security. S5: Perform task scheduling and execution. First, assign tasks to appropriate resources according to the scheduling strategy through task allocation. Then, monitor the running status of tasks during task execution to ensure that tasks can be completed on time through execution monitoring. Dynamically adjust task allocation according to resource status and task execution status to improve scheduling efficiency. S6: Provide feedback and optimization, collect task execution results and user feedback to evaluate the effectiveness of the scheduling strategy, and optimize the scheduling strategy based on the feedback results to improve the effectiveness of future task scheduling; S7: Perform security and privacy protection to ensure the security and privacy of user data during task scheduling; and implement strict access control policies to prevent unauthorized access and resource abuse; S8: Conduct continuous monitoring and maintenance. Continuously monitor the cloud computing environment and changes in user needs to identify potential problems; and regularly maintain and upgrade the scheduling system to keep it running efficiently and stably.
2. A method for optimizing task scheduling based on short-term demand of cloud computing network users as claimed in claim 1, characterized in that: The step S1 specifically includes the following steps: S1. Data collection: Conduct user surveys to collect information about users’ computing task requirements; analyze historical data to extract task features, such as task execution time and resource consumption; S2. Extraction of demand characteristics: Based on the collected data, determine the main characteristic parameters of the demand; characteristic parameters may include computational effort (such as the number of CPU cycles), data transfer volume (such as data size), and time sensitivity (such as task deadline); S3. Demand classification: Classify the demands into different categories based on characteristic parameters; categories can be defined based on task type (e.g., batch processing, real-time processing), task size (e.g., small, medium, large), and urgency (e.g., low, medium, high); S4. Demand quantification: Quantify each demand category to facilitate mathematical modeling of scheduling strategies; for example, the amount of computation can be quantified as CPU hours, the amount of data transferred can be quantified as megabytes (MB), and the time sensitivity can be quantified as relative deadlines (such as the maximum delay that a task can tolerate); S5. Prioritize requirements: Prioritize requirements based on their importance and urgency; a weighted approach can be used to determine priorities, for example, giving higher weights to time-sensitive tasks.
3. The method for optimizing task scheduling based on short-term demand of cloud computing network users according to claim 1, characterized in that: The step S2 specifically includes: Configure monitoring parameters: define the resource parameters that need to be monitored, such as CPU usage, memory usage, disk space, and network bandwidth; set the monitoring frequency and determine the time interval for collecting data; Implement monitoring: Deploy monitoring tools on the cloud computing platform and start collecting resource usage data; ensure the accuracy and real-time nature of monitoring data; Data storage and analysis: Store the collected monitoring data in a database or data warehouse; use data analysis tools to visualize the data to understand resource usage trends; Resource assessment: Evaluate the current resource status based on monitoring data, including resource idle rate, saturation, and performance bottlenecks; predict resource load conditions in the future to provide a basis for scheduling strategies.
4. The method for optimizing task scheduling based on short-term demand of cloud computing network users according to claim 1, characterized in that: In step S3, it is also necessary to select or design a suitable scheduling algorithm according to the design goal, such as polling scheduling, minimum connection scheduling, and priority-based scheduling; and then design a dynamic adjustment mechanism to automatically adjust the scheduling strategy when the resource status changes.
5. The method for optimizing task scheduling based on short-term demand of cloud computing network users according to claim 1, characterized in that: In step S4, the specific process includes: A. Task decomposition: Task analysis: Analyze the composition of large tasks and determine the independent parts that can be decomposed; evaluate the dependencies between tasks and determine which parts can be processed in parallel; Divide tasks: Divide large tasks into multiple small tasks based on analysis results; ensure that each small task is a manageable and executable unit; Define interfaces: Define clear input and output interfaces for each small task; ensure that tasks can exchange data and results through standardized interfaces; Task scheduling: Put the decomposed small tasks into the scheduling queue and prepare for scheduling; select the appropriate scheduling strategy according to the characteristics of the task and resource conditions; B. Data preprocessing: Data identification: Determine the type and amount of data required for the task; identify sensitive data, such as personally identifiable information and trade secrets; Data compression: compress non-sensitive data to reduce the size of data transmission; select appropriate compression algorithms, such as ZIP, GZIP, Snappy; Data encryption: Encrypt sensitive data to protect data security; select the appropriate AES encryption algorithm; Data transmission optimization: optimize data transmission protocols to reduce data transmission time; use data caching and pre-fetching technology to increase data access speed; Data verification: Data integrity verification is performed before and after data transmission to ensure that the data is not damaged during transmission; verification is performed using a calibrator and digital signature technology; Through task decomposition and data preprocessing, the flexibility and efficiency of task scheduling can be significantly improved; task decomposition allows a large number of tasks to be executed in parallel on different resources, thereby shortening the overall execution time; data preprocessing reduces the overhead of data transmission, improves the efficiency of data processing, and ensures data security; these steps lay the foundation for the subsequent task scheduling and execution stages.
6. The method for optimizing task scheduling based on short-term demand of cloud computing network users according to claim 1, characterized in that: In step S5, the specific process method includes: A. Task allocation: Scheduling strategy selection: Select an appropriate task scheduling strategy based on the results of demand analysis and resource evaluation; strategies may include shortest job first, highest response ratio first, priority scheduling, and round-robin scheduling; Resource matching: Match the most appropriate resources based on the task's resource requirements (such as CPU, memory, storage, and network bandwidth) and resource monitoring data; match based on resource usage, performance, and location factors; Task mapping: Map tasks to selected resources and create task instances; set the environment and necessary parameters for task execution; Task start: Start the task instance and start executing the task; ensure that the task is started correctly on the resources and record the start time of the task; B.Execution monitoring: Status tracking: real-time monitoring of the running status of tasks, including progress, resource usage, and performance indicators; use monitoring tools to collect data during task execution; Anomaly detection: detect anomalies in task execution, such as insufficient resources, execution errors, and deadlocks; set up an alarm mechanism to immediately notify the system administrator once an anomaly is detected; Performance analysis: Analyze the performance data of task execution to evaluate whether the task is performed as expected; compare actual performance with expected performance to determine whether resource allocation needs to be adjusted; C. Dynamic adjustment: Resource reallocation: Dynamically adjust task allocation based on real-time resource status and task execution status; migrate tasks from resources with higher loads to resources with lower loads to balance the load; Priority adjustment: adjust the priority of tasks based on their importance and urgency; may need to rearrange the order of tasks in the task queue; Scheduling strategy optimization: Optimize scheduling strategies based on historical data and feedback from task execution; regularly evaluate and adjust scheduling algorithms to improve scheduling efficiency; The task scheduling system ensures efficient execution of tasks on cloud computing resources through task allocation, execution monitoring, and dynamic adjustment; these steps help improve resource utilization, shorten task response time, and ensure stable system operation.
7. The method for optimizing task scheduling based on short-term demand of cloud computing network users according to claim 1, characterized in that: In step S6, the specific process method is: A. Result Collection: Execution result recording: Record the results of each task execution, including task completion time, resource consumption, and output data; collect system performance indicators such as response time, throughput, and resource utilization; User feedback acquisition: Collect user satisfaction with task execution through questionnaires, user interviews, and online feedback; Get direct feedback and suggestions from users on task scheduling strategies; B. Analysis and Evaluation: Performance analysis: Analyze task execution results and system performance indicators to evaluate the effectiveness of scheduling strategies; Identify bottlenecks and problems in the scheduling process; User satisfaction evaluation: Evaluate user satisfaction with task execution based on user feedback; Analyze the problems and needs mentioned in user feedback; C. Strategy Optimization: Problem diagnosis: Diagnose problems in the scheduling strategy based on the analysis results; determine whether the problem is caused by improper resource allocation, inappropriate scheduling algorithm, or other reasons; Strategy adjustment: Adjust the scheduling strategy based on the diagnosis results, such as modifying algorithm parameters and adjusting resource allocation rules. It may be necessary to redesign the scheduling algorithm or introduce a new scheduling mechanism. Optimization implementation: deploy the optimized scheduling strategy into the task scheduling system; Monitor the effects of strategy adjustments to ensure that the improvements are positive; Continuous iteration: Treat result feedback and optimization as an ongoing process, regularly collect data and analyze results; continue to optimize scheduling strategies based on new data and feedback.
8. The method for optimizing task scheduling based on short-term demand of cloud computing network users according to claim 1, characterized in that: In step S7, the AES-CTR variant is selected to ensure the security and privacy of user data during the task scheduling process. The encryption process of the AES-CTR variant is: Choose an initial value IV (usually a random number); use AES to encrypt the IV to get the first ciphertext block; XOR the plaintext with the ciphertext block to get the ciphertext; increment the IV (can be a counter or other method) and repeat the above steps; the decryption process of the AES-CTR variant is: use the same IV and key to encrypt the ciphertext block with AES; XOR the ciphertext with the ciphertext block to get the plaintext; increment the IV and repeat the above steps.
9. The method for optimizing task scheduling based on short-term demand of cloud computing network users according to claim 1, characterized in that: In step S7, the encryption process of the AES-GCM variant is: S1. Select an initial value IV; S2. Use AES to encrypt the plaintext to obtain the ciphertext; S3. Calculate the GCM tag (GMAC) for data integrity verification; S4. Transmit the IV, ciphertext and GCM tag together; The decryption process of S5.AES-GCM variant is: S6. Separate the IV, ciphertext and GCM tag from the received data; S7. Use AES to decrypt the ciphertext to obtain the plaintext; S8. Recalculate the GCM tag and compare it with the received GCM tag to ensure data integrity; In step S4, the AES encryption formula is: C=AES(K,P) Among them, C is the ciphertext, K is the key, and P is the plaintext; The AES decryption formula is: P=AES^(-1)(K,C) Among them, P is plaintext, K is the key, and C is the ciphertext; The formula for the XOR operation in the AES-CTR variant is: C_i=P_iXORAES(K,IV_i) Where C_i is the i-th ciphertext block, P_i is the i-th plaintext block, and IV_i is the i-th IV value; The GMAC calculation formula for the AES-GCM variant is: GMAC=AES_GCM_MAC(K,A,C) Where GMAC is the GCM tag, K is the key, A is the Additional Authentication Data (AAD), and C is the ciphertext.
10. The method for optimizing task scheduling based on short-term demand of cloud computing network users according to claim 1, characterized in that: In step S8, the maintenance work includes regularly upgrading the task scheduling system, patching security vulnerabilities, improving the stability and performance of the system, and regularly backing up the configuration data and monitoring data of the task scheduling system, and formulating a disaster recovery plan; performance optimization is also an important part of continuous monitoring and maintenance. By regularly evaluating the performance of the system, identifying and resolving performance bottlenecks, optimizing database queries, caching strategies and system architecture, the system response speed can be improved.
Citation Information
Patent Citations
Resource scheduling system of AI intelligent computing center
CN117472587A
Resource dynamic scheduling technology of distributed cloud computing system
CN117762644A
Protection method and system for information security big data
CN117857018A
Task allocation system based on big data analysis
CN118193172A
Cited By
Power distribution terminal communication network system based on WAPI
CN119362722A
Multi-device collaborative management and optimization system based on soft bus
CN119766809A
Intelligent and safe distribution method and system for digital media
CN120342786A
A method and system for intelligent and secure distribution of digital media
CN120342786B
