System resource intelligent allocation and optimization service system based on flow prediction

Abstract

The invention provides a system resource intelligent allocation and optimization service system based on flow prediction, which belongs to the field of data processing, and comprises the following steps of: acquiring information of all webpage requests processed by a computer to be allocated in a time length from operation start to current time, the access starting time and the access ending time of each access user during each access, the data sequence of the CPU utilization rate of each to-be-allocated computer and the optimal CPU utilization rate of each to-be-allocated computer are processed; obtaining a comparison time period of each to-be-distributed computer and the efficiency of each to-be-distributed computer for processing information at a plurality of times in the comparison time period, and further obtaining a selection weight of each to-be-distributed computer; the invention aims to solve the problem that resource utilization and task completion time cannot be well balanced when tasks are allocated to the to-be-allocated computers at present due to different performances of different to-be-allocated computers.

Description

Technical Field

[0001] This invention belongs to the field of data processing, specifically relating to a system resource intelligent allocation and optimization service system based on traffic prediction. Background Technology

[0002] Traffic prediction is the foundation of intelligent resource allocation, and Deep Reinforcement Learning (DRL) is currently the core algorithm for intelligent resource allocation. The system treats network environment factors such as link status, server load, and predicted traffic demand as "states," and resource allocation actions such as adjusting bandwidth and allocating computing power as "actions." Through continuous trial and error and reward feedback, it learns the optimal strategy. Resource allocation typically requires balancing multiple competing objectives such as bandwidth utilization, latency, cost, and energy consumption. Multi-objective optimization algorithms such as NSGA-II are used to solve these problems. For example, in the case of a cloud service provider, this technology reduced bandwidth costs by 22% while maintaining QoS. The intelligent resource allocation and optimization service system represents the development direction of Automation for Operations and Maintenance (AIOps). It deeply integrates artificial intelligence with operations and maintenance, shifting from "passive response" to "proactive planning," with the ultimate goal of building a highly available, high-performance, and cost-optimal elastic system. Achieving this system requires close collaboration among multiple teams, including data science, software development, and security operations.

[0003] For a website or other computer platform, the amount of system resources processed is enormous, and a large number of people perform many complex operations. Therefore, it is unrealistic to expect a single machine to process all the information. Thus, it is often not a single machine that processes all the information, but rather many machines that process it simultaneously.

[0004] Traditional methods, when allocating tasks to the backend computers of a website, ignore the differences in hardware performance among different machines, such as CPU aging and transmission latency, leading to unbalanced load distribution. Furthermore, CPU utilization and processing efficiency have a non-linear relationship; static threshold allocation strategies cannot adapt to dynamic loads, easily causing high latency or low resource utilization efficiency. In other words, existing algorithms cannot achieve a good balance between resource utilization and rapid task completion. Summary of the Invention

[0005] To address the issue that current methods for allocating tasks to the backend computers of websites ignore the differences in hardware performance between different machines, leading to unbalanced load distribution and preventing existing algorithms from effectively balancing resource utilization and task completion speed, this invention proposes a system resource intelligent allocation and optimization service based on traffic prediction.

[0006] To achieve the above objectives, the present invention provides the following technical solution: Data acquisition module: The backend computers of the website to be optimized are denoted as computers to be assigned; the request time, start time and end time of all web page requests processed by each computer to be assigned during the time from the start of operation to the current time, the start time and end time of each user's access for each access, and the data sequence of CPU utilization of each computer to be assigned and the best CPU utilization of each computer to be assigned. Data processing module: Based on the CPU utilization data sequence of each computer to be assigned and the optimal CPU utilization of each computer to be assigned, the comparison time period of each computer to be assigned is obtained; combined with the request time, start processing time and end processing time of all web page requests processed by each computer to be assigned during the time length from the start of operation to the current time, and the start access time and end access time of each user during each access, the efficiency of each computer to be assigned in processing information at certain times within its comparison time period, the optimal efficiency of each computer to be assigned in processing information, and the number of web page requests processed by each computer to be assigned in each time interval within its comparison time period are obtained, and thus the selection weight of each computer to be assigned is obtained. Traffic allocation module: Distributes traffic based on the selection weight of each computer to be allocated.

[0007] Furthermore, the specific steps for obtaining the comparison time period for each computer to be assigned are as follows: The first The acquisition time corresponding to the last data point with a value of 0 in the data sequence of CPU utilization of the computers to be allocated is denoted as the [number]. The start time for data collection on each computer awaiting allocation; The first Within the data sequence of CPU utilization of the computers awaiting allocation, the acquisition time is in the [number]th [period]. The first value after the start time of data collection for the computers to be assigned is the number of... The data collection time corresponding to the optimal CPU utilization of the computers to be allocated is denoted as the [number]. The data collection termination time for each computer awaiting allocation; The first The time interval between the start and end times of data collection for each computer awaiting allocation is denoted as the [number]. The time period for comparing the number of computers waiting to be assigned.

[0008] Furthermore, the specific steps for obtaining the efficiency of each computer to be assigned in processing information at certain times within its comparison time period, the optimal efficiency of each computer to be assigned in processing information, and the number of web page requests processed by each computer to be assigned in each time interval within its comparison time period are as follows: by Calculate the frequency of one division, and then calculate the first division. The efficiency of the computers awaiting allocation in processing information at certain times within a comparative time period is used to obtain the first... Several time periods within the comparison time frame for the computers awaiting allocation; among them, This is the preset data collection time interval; The first The comparison period of the computers to be allocated a time and its predecessor Hour as the first The comparison period of the computers to be allocated The two time endpoints of the time interval are used to obtain the first time interval. The comparison period of the computers to be allocated A time period; among which, This is the preset second data acquisition time interval; According to the The computer awaiting allocation processes the start and end times of each access by every user during the time period from the start of operation to the current time, and obtains the first... The number of computers to be assigned within their comparison time period The number of users accessing the service within a given time period; According to the The request time, start processing time, and end processing time of all web page requests processed by the computer to be assigned within the time length from the start of operation to the current time are used to obtain the [number]th [page]. The number of computers to be assigned within their comparison time period Several final running time periods within a time period; The request time is in the The comparison period of the computers to be allocated The first time period in the time interval The number of web page requests to be processed by the assigned computer is denoted as the first request. The number of computers to be assigned within their comparison time period The number of web page requests processed within a specific time period; According to the The number of computers to be assigned within their comparison time period The number of users processed in the first time period, the number of users accessed in the second time period, the number of users accessed in the third ... The comparison period of the computers to be allocated The length of the time interval of the first time period, the first The number of computers to be assigned within their comparison time period Several final running time periods within a time period and the first time period The number of computers to be assigned within their comparison time period The number of web page requests processed within the time period of the nth time interval is obtained. The number of computers to be assigned within their comparison time period Efficiency in processing information over a given time period; The request time is in the The comparison period of the computers to be allocated indivual Assigned to indivual Within the time interval between minutes, the first The number of web page requests to be processed by the assigned computer is denoted as the first request. The number of computers to be assigned within their comparison time period The number of web page requests processed in each time interval; According to the The number of computers to be assigned within their comparison time period The method for obtaining the efficiency of information processing in a given time, through the first... The request time, start processing time, and end processing time of all web page requests processed by the computer to be assigned within the time length from the start of operation to the current time, and the start access time and end access time of each user for each access, are used to obtain the [number]th [page / request]. The number of computers to be assigned is in The efficiency of the computer waiting to be assigned in processing information during the data acquisition termination time is denoted as the [number]. The optimal efficiency for a computer to process information when it is assigned to a platform.

[0009] Furthermore, the obtained first The number of computers to be assigned within their comparison time period The specific steps for processing the number of accessing users within a given time period are as follows: If the first The first computer to be assigned for processing The user in the first The end time of the second visit is in The comparison period of the computers to be allocated Within the time period of the nth time, and the nth The first computer to be assigned for processing The user in the first The start time of the second visit is not in the [number]th visit. The comparison period of the computers to be allocated Within the time period of the first time, the first time... The first computer to be assigned for processing The user in the first The end time of the first visit is denoted as the [number]th visit. The number of computers to be assigned within their comparison time period The first end time of access within a time period; The first The comparison period of the computers to be allocated Within the time period, the first... The number of all users' start access times for all users awaiting computer processing, compared with the number of... The number of computers to be assigned within their comparison time period The sum of the number of first end access times within a time period is denoted as the i-th time period. The number of computers to be assigned within their comparison time period The number of users accessed during a given time period.

[0010] Furthermore, the obtained first The number of computers to be assigned within their comparison time period The specific steps for several final running time periods within a given time period are as follows: The first The first computer to be assigned for processing The start and end times of a webpage request are taken as two endpoints of a runtime period to obtain the first... A time period during which a computer awaiting allocation will be used to determine the first... The time included in a runtime period of a computer waiting to be assigned is denoted as the [number]. The running time of the computers awaiting allocation; The first The comparison period of the computers to be allocated The first time period in the time interval The runtime of all computers awaiting allocation is combined with the runtime of their adjacent computers to obtain the first... The number of computers to be assigned within their comparison time period Several final running time periods within a time period.

[0011] Furthermore, the obtained first The number of computers to be assigned within their comparison time period The specific formula for calculating the efficiency of information processing in a given time is as follows: In the formula, Indicates the first The number of computers to be assigned within their comparison time period Efficiency in processing information over time Indicates the first The number of computers to be assigned within their comparison time period The sum of the durations of all final running time segments within a given time period. Indicates the first The number of computers to be assigned within their comparison time period The number of users accessed during a given time period. Indicates the first The number of computers to be assigned within their comparison time period The number of web page requests processed within a given time period. Indicates the first The comparison period of the computers to be allocated The length of a time period.

[0012] Furthermore, the specific calculation formula for obtaining the selection weight of each computer to be assigned is as follows: In the formula, Indicates the first Resource consumption rate of computers awaiting allocation. Indicates the first The quantity of efficiency of the computers awaiting allocation in processing information acquired during their comparative time period. Indicates the first The optimal efficiency for a computer to process information when it is assigned to a platform. Indicates the first The number of computers to be assigned within their comparison time period Efficiency in processing information over time Indicates the first The number of computers to be assigned within their comparison time period The number of web page requests processed in each time interval; The selection weight of each computer to be assigned is obtained based on the resource consumption rate of each computer to be assigned and the number of web page requests processed by each computer to be assigned in each time interval during its comparison period.

[0013] Furthermore, the specific calculation formula for obtaining the selection weight of each computer to be allocated based on the resource consumption rate of each computer to be allocated and the number of web page requests processed by each computer to be allocated in each time interval during its comparison period is as follows: In the formula, Indicates the first The selection weight of the computers to be assigned. Indicates the first Resource consumption rate of computers awaiting allocation. Indicates the first Resource consumption rate of computers awaiting allocation. This indicates the number of computers to be allocated. Indicates the first The quantity of efficiency of the computers awaiting allocation in processing information acquired during their comparative time period. Indicates the first The quantity of efficiency of the computers awaiting allocation in processing information acquired during their comparative time period. Indicates the first The number of computers to be assigned within their comparison time period The number of web page requests processed in each time interval. Indicates the first The number of computers to be assigned within their comparison time period The number of web page requests processed in each time interval.

[0014] Furthermore, the specific steps for traffic allocation are as follows: When assigning web page request tasks to the computer to be assigned, the CPU utilization of the computer with the highest selection weight is first brought to its optimal CPU utilization. Then, the subsequent web page request tasks are assigned to the next computer with the highest selection weight, until the CPU utilization of that computer reaches its optimal CPU utilization. Subsequent webpage request tasks are then assigned to the next computer with the highest selection weight, until the CPU utilization of that computer reaches its optimal CPU utilization. This process is repeated continuously to complete the traffic allocation.

[0015] The intelligent resource allocation and optimization service system based on traffic prediction provided by this invention has the following beneficial effects: When allocating traffic to a website to be optimized, due to the performance differences of different computers to be allocated, the time required for different computers to process web page requests varies under the same CPU utilization. Therefore, this invention obtains the efficiency of each computer in processing information at each time based on the number of web page requests and users processed by each computer in the period before each time, solving the problem that the performance differences of different computers cannot be quantified by the CPU utilization of each computer. Furthermore, this invention quantifies the resource waste of each computer by measuring the change in the efficiency of information processing when the CPU utilization of each computer increases from 0 to the optimal CPU utilization, solving the problem that the resource waste of different computers cannot be directly quantified by the CPU utilization of different computers. Finally, based on the resource waste of each computer and the amount of information processed by each computer, the priority of each computer is quantified, solving the problem that current methods cannot effectively balance resource utilization and information processing time when allocating tasks to computers. Attached Figure Description

[0016] To more clearly illustrate the embodiments and design schemes of the present invention, the accompanying drawings required for this embodiment will be briefly described below. The drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1This is a flowchart of a system resource intelligent allocation and optimization service system based on traffic prediction, according to an embodiment of the present invention. Detailed Implementation

[0018] To enable those skilled in the art to better understand and implement the technical solutions of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention and should not be construed as limiting the scope of protection of the present invention.

[0019] Example 1: This invention provides a system for intelligent allocation and optimization of system resources based on traffic prediction, specifically as follows: Figure 1 As shown, it includes: Step S001: The backend computers of the website to be optimized are denoted as computers to be assigned; obtain the request time, start processing time and end processing time of all web page requests processed by each computer to be assigned during the time length from the start of operation to the current time, the start access time and end access time of each user during each access, and the data sequence of CPU utilization of each computer to be assigned and the optimal CPU utilization of each computer to be assigned.

[0020] Specifically, the backend computers of the website to be optimized are denoted as "computers to be assigned". For each computer to be assigned, the request time, start time, and end time of all web page requests processed from the start of operation to the current time are obtained, along with the start and end times of each user's access for each visit. The CPU utilization of each computer to be allocated is acquired at second intervals over a period of time from the start of operation to the current time, resulting in a data sequence of CPU utilization for each computer to be allocated. The optimal CPU utilization for each computer to be allocated is then determined. In this embodiment, a first acquisition interval is preset. This example is used for illustration; other values ​​can be set in other implementations. Furthermore, each data point within the CPU utilization data sequence of each computer to be allocated corresponds to a collection time.

[0021] At this point, the data to be analyzed is obtained.

[0022] Step S002: Based on the CPU utilization data sequence of each computer to be assigned and the optimal CPU utilization of each computer to be assigned, obtain the comparison time period for each computer to be assigned; combine the request time, start processing time, and end processing time of all web page requests processed by each computer to be assigned during the time length from the start of operation to the current time, and the start access time and end access time of each user during each access, to obtain the efficiency of each computer to be assigned in processing information at certain times within its comparison time period, the optimal efficiency of each computer to be assigned in processing information, and the number of web page requests processed by each computer to be assigned in each time interval within its comparison time period, and thus obtain the selection weight of each computer to be assigned.

[0023] It's important to note that CPU efficiency isn't a simple linear curve; it's more like an inverted U-shaped curve. When CPU utilization is low, there's ample idle time to handle sudden computational tasks, resulting in very rapid response. Threads can obtain CPU time slices with almost no waiting time, leading to very low latency. However, the CPU automatically enters a power-saving state by reducing clock frequency and voltage, resulting in lower efficiency. But at a certain utilization level, such as 70% to 90%, the CPU is almost constantly busy processing useful tasks, without wasting power on idling. At this point, the CPU typically runs at a higher frequency, maximizing computational throughput. This CPU utilization range is ideal for tasks requiring batch processing of large amounts of data. Conversely, when CPU utilization is very high, all CPU cores are fully occupied. New threads must wait in the run queue, drastically increasing the response time of newly added web page tasks, leading to a lag-like user experience. Therefore, when allocating web page tasks, priority should be given to computers with lower CPU utilization.

[0024] It should be further explained that when assigning webpage response tasks to computers, the CPU utilization of multiple computers may not reach their optimal value. Therefore, a priority is calculated for each computer. Due to inherent differences between individual computers, their optimal CPU utilization may differ, and the time spent processing webpage requests at the same CPU utilization may also vary, making it impossible to directly calculate the priority of each computer to be assigned based on CPU utilization. Therefore, this invention first calculates the efficiency of each computer in processing information when obtaining the priority of each computer to be assigned. Then, this invention quantifies the resource waste of each computer as its CPU utilization rises from 0 to its optimal value by comparing its efficiency in processing information at its optimal CPU utilization with its efficiency in processing information at each data collection time, thus obtaining the priority of each computer to be assigned. To ensure real-time performance when calculating the priority of each computer to be assigned, data from the last CPU time period of each computer from 0 to its optimal value is selected for priority calculation.

[0025] It should be further explained that in order to allocate computer system resources in real time, when obtaining the efficiency of the computer to be allocated in processing information, in order to ensure real-time performance, it is necessary to use relevant data such as the number of web page requests and the number of visitors collected in the period before the calculation time to obtain the efficiency of the computer to be allocated in processing information at each time.

[0026] It's important to further explain that the more web page requests a computer processes, the more efficient it is at processing information. Therefore, dividing the total time each computer takes to process web page requests within a given time period by the total number of users accessing the page on that computer during that time period gives the time each computer takes to process each user's request. Then, dividing the total number of web page requests within each time period by the length of that time period gives the total number of web page requests per unit time. This data reflects the volume of web page requests at any given moment, but it is collectively created by all visitors assigned to that computer. Therefore, dividing the total number of web page requests per unit time by the number of visitors within that time period gives the number of web page requests created by each person per unit time. Finally, multiplying the time each computer takes to process each user's request by the number of web page requests created by each person per unit time gives the number of web page requests processed by each assigned computer for each individual user, i.e., the efficiency of each assigned computer in processing information.

[0027] It should be further noted that the more powerful a computer is at processing web page requests, the greater the likelihood that it will be prioritized for web page request allocation. Therefore, the number of web page requests a computer processes over a period of time is used to quantify its web page request processing capability, thus obtaining the selection weight for each computer to be allocated a request.

[0028] Specifically, the first The acquisition time corresponding to the last data point with a value of 0 in the data sequence of CPU utilization of the computers to be allocated is denoted as the [number]. The start time for data collection on the computers awaiting allocation. (The number will be...) Within the data sequence of CPU utilization of the computers awaiting allocation, the acquisition time is in the [number]th [period]. The first value after the start time of data collection for the computers to be assigned is the number of... The data collection time corresponding to the optimal CPU utilization of the computers to be allocated is denoted as the [number]. The data collection termination time for the computers awaiting allocation. (The number will be...) The time interval between the start and end times of data collection for each computer awaiting allocation is denoted as the [number]. The time period for comparing the number of computers awaiting allocation. Among them, the first... The time interval between the start and end times of data collection for each computer awaiting allocation is defined as the period starting from the first data collection time. The start and end times of data collection for the computers to be assigned are used as the left and right endpoints to obtain a time period. Obtaining a data segment or a time period based on these two endpoints is a well-known technique and will not be elaborated upon in this embodiment.

[0029] Among them, if the first Within the data sequence of CPU utilization of the computers awaiting allocation, the acquisition time is in the [number]th [period]. All data values ​​after the start time of the data collection for the computers to be assigned are not the first. The optimal CPU utilization of the computers awaiting allocation. Then, the first... The acquisition time corresponding to the last data point with a value of 0 in the data sequence of CPU utilization of the computers to be allocated is denoted as the th . The start time of data collection for the computers to be assigned; and the first The acquisition time corresponding to the second-to-last data point with a value of 0 in the data sequence of CPU utilization of the computers to be allocated is denoted as the [number]. The start time for data collection on the computers awaiting allocation. (The number will be...) Within the data sequence of CPU utilization of the computers awaiting allocation, the acquisition time is in the [number]th [period]. The first value after the start time of data collection for the computers to be assigned is the number of... The data collection time corresponding to the optimal CPU utilization of the computers to be allocated is denoted as the [number]. The data collection termination time for the computers awaiting allocation. If the... Within the data sequence of CPU utilization of the computers awaiting allocation, the acquisition time is in the [number]th [period]. All data values ​​after the start time of the data collection for the computers to be assigned are not the first. The optimal CPU utilization of the computers awaiting allocation. Then, the first... The acquisition time corresponding to the last two 0 values ​​in the data sequence of CPU utilization of the computers to be allocated is denoted as the th acquisition time. The start time of data collection for the computers to be assigned, and the first... The acquisition time corresponding to the third-to-last data point with a value of 0 in the data sequence of CPU utilization of the computers to be allocated is denoted as the [number]. The above operation is repeated continuously, starting from the data collection start time of the computers to be assigned, until the first computer is found. The data collection termination time for the computers to be assigned.

[0030] Furthermore, with Calculate the frequency of one division, and then calculate the first division. The efficiency of the computers awaiting allocation in processing information at certain times within a comparative time period is used to obtain the first... The comparison period for the computers awaiting allocation is several time points within that timeframe. Among them, the first... The number of computers to be assigned within their comparison time period The time and the first The time interval is The sampling time interval is preset in this embodiment. This example is used to illustrate the concept; other values ​​can be set in other implementations.

[0031] Furthermore, the first The comparison period of the computers to be allocated a time and its predecessor Hour as the first The comparison period of the computers to be allocated The two time endpoints of the time interval are used to obtain the first time interval. The comparison period of the computers to be allocated The time period is defined as a time interval. Specifically, the second data acquisition time interval is preset in this embodiment. This example is used for illustration; other values ​​can be set in other implementations. Furthermore, obtaining the time period based on two time endpoints is a well-known technique and will not be elaborated upon in this embodiment.

[0032] Furthermore, if the first The first computer to be assigned for processing The user in the first The end time of the second visit is in The comparison period of the computers to be allocated Within the time period of the nth time, and the nth The first computer to be assigned for processing The user in the first The start time of the second visit is not in the [number]th visit. The comparison period of the computers to be allocated In the time period of the nth time, then the nth time... The first computer to be assigned for processing The user in the first The end time of the first visit is denoted as the [number]th visit. The number of computers to be assigned within their comparison time period The first end time of access within a time period.

[0033] Furthermore, the first The comparison period of the computers to be allocated Within the time period, the first... The number of all users' start access times for all users awaiting computer processing, compared with the number of... The number of computers to be assigned within their comparison time period The sum of the number of first end access times within a time period is denoted as the i-th time period. The number of computers to be assigned within their comparison time period The number of users accessed during a given time period.

[0034] Furthermore, the request time will be in the [number]th [year]. The comparison period of the computers to be allocated The first time period in the time interval The number of web page requests to be processed by the assigned computer is denoted as the first request. The number of computers to be assigned within their comparison time period The number of web page requests processed within a given time period.

[0035] Furthermore, the first The first computer to be assigned for processing The start and end times of a webpage request are taken as two endpoints of a runtime period to obtain the first... A time period during which a computer awaiting allocation will be used to determine the first... The time included in a runtime period of a computer waiting to be assigned is denoted as the [number]. The running time of the computer to be allocated. The process of obtaining a data segment based on data from two endpoints is a well-known technique and will not be described in detail in this embodiment.

[0036] Furthermore, the first The comparison period of the computers to be allocated The first time period in the time interval The runtime of all computers awaiting allocation is combined with the runtime of their adjacent computers to obtain the first... The number of computers to be assigned within their comparison time period The implementation involves several final running time periods within a given time period. The process of merging each running time period with its adjacent running times to obtain a single time period is a well-known technique and will not be elaborated upon here.

[0037] Furthermore, obtain the first The number of computers to be assigned within their comparison time period The specific formula for calculating the efficiency of information processing in a given time is as follows: In the formula, Indicates the first The number of computers to be assigned within their comparison time period Efficiency in processing information over time Indicates the first The number of computers to be assigned within their comparison time period The sum of the durations of all final running time segments within a given time period. Indicates the first The number of computers to be assigned within their comparison time period The number of users accessed during a given time period. Indicates the first The number of computers to be assigned within their comparison time period The number of web page requests processed within a given time period. Indicates the first The comparison period of the computers to be allocated The length of a time period.

[0038] It should be noted that, The value is used to reflect the first The number of computers to be assigned within their comparison time period How many web page requests need to be processed per unit of time within a given time period? The value is used to reflect the first The number of computers to be assigned within their comparison time period How many web page requests will a single user generate per unit of time within a given time period? It is used to reflect the first The number of computers to be assigned within their comparison time period The time required to process all web page requests for a user within a given time period; It is used to reflect the first The number of web page requests a user can handle by the computer to be assigned; the larger the value, the more web page requests the computer can handle, that is, the higher the efficiency of information processing.

[0039] Furthermore, according to The method of obtaining the first The number of computers to be assigned is in The efficiency of the computer waiting to be assigned in processing information during the data acquisition termination time is denoted as the [number]. The optimal efficiency for a computer to process information when it is assigned to a platform.

[0040] Furthermore, the request time will be in the [number]th [year]. The comparison period of the computers to be allocated indivual Assigned to indivual Within the time interval between minutes, the first The number of web page requests to be processed by the assigned computer is denoted as the first request. The number of computers to be assigned within their comparison time period The number of web page requests processed in each time interval. Among them, when At that time, the first The 0th computer in the comparison period for the computers to be allocated Divided into the first The time corresponding to the left endpoint of the comparison time period for the computers to be assigned.

[0041] Thus, we obtain the efficiency of each computer to be assigned in processing information over a certain period of time within its comparison time frame.

[0042] Furthermore, the specific formula for calculating the resource consumption rate of each computer to be allocated is as follows: In the formula, Indicates the first Resource consumption rate of computers awaiting allocation. Indicates the first The quantity of efficiency of the computers awaiting allocation in processing information acquired during their comparative time period. Indicates the first The optimal efficiency for a computer to process information when it is assigned to a platform. Indicates the first The number of computers to be assigned within their comparison time period Efficiency in processing information over time Indicates the first The number of computers to be assigned within their comparison time period The number of web page requests processed in the time interval. Where, if the ... The number of computers awaiting allocation during the comparison period was [number missing]. The efficiency of processing information in a given time, then at this time .

[0043] It should be noted that, , is used to explain the first The waste of CPU resources on the computers waiting to be allocated; the larger the value, the more wasted the CPU resources. There is significant resource waste when the CPU utilization of the computers awaiting allocation is below optimal. To quantify the first Resource waste when the CPU utilization of a pending computer increases from 0 to its optimal value; The resource waste ratio over a period of time is obtained, and then... Divide by To reflect the average waste of resources; The larger the value, the more likely it is to be the first. The more resources are wasted when the CPU utilization of a computer awaiting allocation does not reach its optimal value.

[0044] At this point, the resource consumption rate of each computer to be allocated is obtained.

[0045] Further, calculate the first The specific formula for calculating the selection weight of each computer to be assigned is as follows: In the formula, Indicates the first The selection weight of the computers to be assigned. Indicates the first Resource consumption rate of computers awaiting allocation. Indicates the first Resource consumption rate of computers awaiting allocation. This indicates the number of computers to be allocated. Indicates the first The quantity of efficiency of the computers awaiting allocation in processing information acquired during their comparative time period. Indicates the first The quantity of efficiency of the computers awaiting allocation in processing information acquired during their comparative time period. Indicates the first The number of computers to be assigned within their comparison time period The number of web page requests processed in each time interval. Indicates the first The number of computers to be assigned within their comparison time period The number of web page requests processed in each time interval.

[0046] It should be noted that, The larger the value, the more significant the [value]. When the CPU utilization of the computers awaiting allocation increases from 0 to the optimal value, a significant amount of resources are wasted. Therefore, priority should be given to the first... The more likely a computer is to be assigned a webpage, the greater the likelihood of that computer being assigned a webpage. The larger the value, the more significant the [value]. The stronger the computer's ability to process web page requests, the more priority it will give to the first request when there is a web page request. The request to assign a webpage to a computer awaiting allocation is being sent.

[0047] At this point, the selection weights for each computer to be assigned are obtained.

[0048] Step S003: Distribute traffic according to the selection weight of each computer to be assigned.

[0049] Specifically, through the time before the current time The request times of all web page requests awaiting computer processing within a day, to obtain future... The trend of the number of web page requests to the website to be optimized within one hour. The preset number of days in this embodiment is [not specified]. Hour length This example is used for illustration; other values ​​can be set in other implementations. Obtaining the trend of the number of web page requests processed by the website to be optimized through historical data is a well-known technique and will not be elaborated upon in this embodiment.

[0050] Furthermore, when assigning web page request tasks to the computers to be assigned, the CPU utilization of the computer with the highest selection weight is first brought to its optimal CPU utilization. Subsequent web page request tasks are then assigned to the next computer with the highest selection weight, until that computer's CPU utilization reaches its optimal CPU utilization. This process is repeated, continuing until the next computer with the highest selection weight is reached. This allows all web page request tasks to be processed faster and with higher resource utilization while ensuring all information is processed, thus enabling intelligent allocation of system resources based on traffic prediction.

[0051] This concludes the embodiment.

Claims

1. A system for intelligent allocation and optimization of system resources based on traffic prediction, characterized in that: include: Data acquisition module: The backend computers of the website to be optimized are designated as computers to be allocated; The system acquires the request time, start time, and end time of all web page requests processed by each computer to be assigned within the time span from the start of operation to the current time, the start time and end time of each user's access for each visit, and the CPU utilization data sequence and optimal CPU utilization of each computer to be assigned. The data processing module obtains the comparison time period for each computer to be assigned based on the CPU utilization data sequence and optimal CPU utilization. Combining the request time, start time, and end time of all web page requests processed by each computer within the time span from the start of operation to the current time, and the start time and end time of each user's access for each visit, the system obtains the efficiency of each computer in processing information at certain times within its comparison time period, the optimal efficiency of each computer in processing information, and the number of web page requests processed by each computer in each time interval within its comparison time period, thus obtaining the selection weight of each computer to be assigned. The traffic allocation module allocates traffic based on the selection weight of each computer to be assigned.

2. The intelligent resource allocation and optimization service system based on traffic prediction according to claim 1, characterized in that, The specific steps for obtaining the comparison time period for each computer to be assigned are as follows: [The text abruptly ends here, likely due to an incomplete sentence or a formatting error.] The acquisition time corresponding to the last data point with a value of 0 in the data sequence of CPU utilization of the computers to be allocated is denoted as the [number]. The start time for data collection on each computer awaiting allocation; The first Within the data sequence of CPU utilization of the computers awaiting allocation, the acquisition time is in the [number]th [period]. The first value after the start time of data collection for the computers to be assigned is the number of... The data collection time corresponding to the optimal CPU utilization of the computers to be allocated is denoted as the [number]. The data collection termination time for the computers awaiting allocation; the first... The time interval between the start and end times of data collection for each computer awaiting allocation is denoted as the [number]. The time period for comparing the number of computers waiting to be assigned.

3. The intelligent resource allocation and optimization service system based on traffic prediction according to claim 1, characterized in that, The specific steps for obtaining the efficiency of each computer to be assigned in processing information at certain times within its comparison time period, the optimal efficiency of each computer to be assigned in processing information, and the number of web page requests processed by each computer in each time interval within its comparison time period are as follows: Calculate the frequency of one division, and then calculate the first division. The efficiency of the computers awaiting allocation in processing information at certain times within a comparative time period is used to obtain the first... Several time periods within the comparison time frame for the computers awaiting allocation; among them, The preset data collection time interval; the first... The comparison period of the computers to be allocated a time and its predecessor Hour as the first The comparison period of the computers to be allocated The two time endpoints of the time interval are used to obtain the first time interval. The comparison period of the computers to be allocated A time period; among which, The second preset acquisition time interval; according to the first The computer awaiting allocation processes the start and end times of each access by every user during the time period from the start of operation to the current time, and obtains the first... The number of computers to be assigned within their comparison time period The number of users processed within a time period; according to the number of... The request time, start processing time, and end processing time of all web page requests processed by the computer to be assigned within the time length from the start of operation to the current time are used to obtain the [number]th [page]. The number of computers to be assigned within their comparison time period Several final running time periods within a time period; the requested time will be in the [number]th [time period]. The comparison period of the computers to be allocated The first time period in the time interval The number of web page requests to be processed by the assigned computer is denoted as the first request. The number of computers to be assigned within their comparison time period The number of web page requests processed within a time period; based on the number of... The number of computers to be assigned within their comparison time period The number of users processed in the first time period, the number of users accessed in the second time period, the number of users accessed in the third ... The comparison period of the computers to be allocated The length of the time interval, the first The number of computers to be assigned within their comparison time period Several final running time periods within a time period and the first time period The number of computers to be assigned within their comparison time period The number of web page requests processed within the time period of the nth time interval is obtained. The number of computers to be assigned within their comparison time period Efficiency in processing information at a given time; request time at the 1st... The comparison period of the computers to be allocated indivual Assigned to indivual Within the time interval between minutes, the first The number of web page requests to be processed by the assigned computer is denoted as the first request. The number of computers to be assigned within their comparison time period The number of web page requests processed in each time interval; according to the number of... The number of computers to be assigned within their comparison time period The method for obtaining the efficiency of information processing in a given time, through the first... The request time, start processing time, and end processing time of all web page requests processed by the computer to be assigned within the time length from the start of operation to the current time, and the start access time and end access time of each user for each access, are used to obtain the [number]th [page / request]. The number of computers to be assigned is in The efficiency of the computer waiting to be assigned in processing information during the data acquisition termination time is denoted as the [number]. The optimal efficiency for a computer to process information when it is assigned to a platform.

4. The intelligent resource allocation and optimization service system based on traffic prediction according to claim 3, characterized in that, The obtained number The number of computers to be assigned within their comparison time period The specific steps for processing the number of accessing users within a given time period are as follows: If the first... The first computer to be assigned for processing The user in the first The end time of the second visit is in The comparison period of the computers to be allocated Within the time period of the nth time, and the nth The first computer to be assigned for processing The user in the first The start time of the second visit is not in the [number]th visit. The comparison period of the computers to be allocated Within the time period of the first time, the first time... The first computer to be assigned for processing The user in the first The end time of the first visit is denoted as the [number]th visit. The number of computers to be assigned within their comparison time period The first end time of the access within the time period; will the first The comparison period of the computers to be allocated Within the time period, the first... The number of all users' start access times for all users awaiting computer processing, compared with the number of... The number of computers to be assigned within their comparison time period The sum of the number of first end access times within a time period is denoted as the i-th time period. The number of computers to be assigned within their comparison time period The number of users accessed during a given time period.

5. The intelligent resource allocation and optimization service system based on traffic prediction according to claim 3, characterized in that, The obtained number The number of computers to be assigned within their comparison time period The specific steps for several final running time periods within a given time period are as follows: [The text abruptly ends here, likely due to an incomplete sentence or a formatting error.] The first computer to be assigned for processing The start and end times of a webpage request are taken as two endpoints of a runtime period to obtain the first... A time period during which a computer awaiting allocation will be used to determine the first... The time included in a runtime period of a computer waiting to be assigned is denoted as the [number]. The running time of the computers to be allocated; the first The comparison period of the computers to be allocated The first time period in the time interval The runtime of all computers awaiting allocation is combined with the runtime of their adjacent computers to obtain the first... The number of computers to be assigned within their comparison time period Several final running time periods within a time period.

6. The intelligent allocation and optimization service system for system resources based on traffic prediction according to claim 3, characterized in that, The obtained number The number of computers to be assigned within their comparison time period The specific formula for calculating the efficiency of information processing in a given time is as follows: In the formula, Indicates the first The number of computers to be assigned within their comparison time period Efficiency in processing information over time Indicates the first The number of computers to be assigned within their comparison time period The sum of the durations of all final running time segments within a given time period. Indicates the first The number of computers to be assigned within their comparison time period The number of users accessed during a given time period. Indicates the first The number of computers to be assigned within their comparison time period The number of web page requests processed within a given time period. Indicates the first The comparison period of the computers to be allocated The length of a time period.

7. The intelligent allocation and optimization service system for system resources based on traffic prediction according to claim 1, characterized in that, The specific formula for calculating the selection weight of each computer to be assigned is as follows: In the formula, Indicates the first Resource consumption rate of computers awaiting allocation. Indicates the first The quantity of efficiency of the computers awaiting allocation in processing information acquired during their comparative time period. Indicates the first The optimal efficiency for a computer to process information when it is assigned to a platform. Indicates the first The number of computers to be assigned within their comparison time period Efficiency in processing information over time Indicates the first The number of computers to be assigned within their comparison time period The number of web page requests processed in each time interval; the selection weight of each computer to be assigned is obtained based on the resource consumption rate of each computer to be assigned and the number of web page requests processed by each computer to be assigned in each time interval within its comparison time period.

8. The intelligent allocation and optimization service system for system resources based on traffic prediction according to claim 7, characterized in that, The specific calculation formula for obtaining the selection weight of each computer to be assigned, based on the resource consumption rate of each computer to be assigned and the number of web page requests processed by each computer to be assigned in each time interval during its comparison period, is as follows: In the formula, Indicates the first The selection weight of the computers to be assigned. Indicates the first Resource consumption rate of computers awaiting allocation. Indicates the first Resource consumption rate of computers awaiting allocation. This indicates the number of computers to be allocated. Indicates the first The quantity of efficiency of the computers awaiting allocation in processing information acquired during their comparative time period. Indicates the first The quantity of efficiency of the computers awaiting allocation in processing information acquired during their comparative time period. Indicates the first The number of computers to be assigned within their comparison time period The number of web page requests processed in each time interval. Indicates the first The number of computers to be assigned within their comparison time period The number of web page requests processed in each time interval.

9. The intelligent allocation and optimization service system for system resources based on traffic prediction according to claim 1, characterized in that, The specific steps for traffic allocation are as follows: When allocating web page request tasks to the computer to be allocated, first ensure that the CPU utilization of the computer with the highest selection weight reaches its optimal CPU utilization, then allocate subsequent web page request tasks to the next computer with the highest selection weight, until the CPU utilization of that computer reaches its optimal CPU utilization; then allocate subsequent web page request tasks to the next computer with the highest selection weight, until the CPU utilization of that computer reaches its optimal CPU utilization, and continue performing the above operations to complete the traffic allocation.

Images