Information processing resource configuration method, prediction model training method and device

Abstract

The embodiment of the invention provides an information processing resource configuration method and device and a prediction model training method and device. The method comprises the steps of extracting target resource features of a time dimension for historical use data of information processing resources by a service before a target time period; inputting the target resource characteristics into a prediction model of the information processing resources, and obtaining a predicted demand quantity of the business for the information processing resources in the target time period; wherein the prediction model is obtained by training historical use data of information processing resources based on business; generating resource configuration information based on the predicted demand quantity and information of existing information processing resources; wherein the resource configuration information is used for indicating recommendation expansion and contraction operation for the existing information processing resources. According to the method, normal service processing in the target time period can be ensured, idle and waste of resources are avoided, and the service processing cost is saved.

Description

Technical Field

[0001] The embodiments of this specification relate to the field of computer technology, and in particular to a method for configuring information processing resources; they also relate to a device for configuring information processing resources, a method for training a prediction model for information processing resources, a device for training a prediction model for information processing resources, a computing device, a computer-readable storage medium, and a computer program product. Background Art

[0002] With the development of computer technology, various businesses can be processed by computers. The total capacity of the computer's processing resources and the amount of processing resources allocated to each business will affect the processing effect of the business.

[0003] In related technologies, in order to ensure the continuity of business processing and avoid sudden interruption of business processing due to insufficient available processing resources, the computer will expand the capacity of the processing resources after receiving processing requests for certain businesses.

[0004] However, many businesses do not need to be processed continuously. Once they are processed within a period of time, they no longer need to occupy the corresponding processing resources. This will cause waste and idleness of computer processing resources, resulting in higher overall costs for business processing. Summary of the Invention

[0005] The embodiments of this specification provide a method for configuring information processing resources, which can ensure normal business processing within a target time period while avoiding idle and wasted resources, thereby saving business processing costs. One or more embodiments of this specification also relate to a method for training a prediction model for information processing resources, an apparatus for configuring information processing resources, an apparatus for training a prediction model for information processing resources, a computing device, a computer-readable storage medium, and a computer program product.

[0006] According to one aspect of an embodiment of this specification, a method for configuring information processing resources is provided, the method comprising:

[0007] Extract the target resource characteristics in the time dimension based on the historical usage data of information processing resources by businesses before the target period;

[0008] Inputting the target resource characteristics into a prediction model for information processing resources to obtain a predicted demand for information processing resources by the business during the target period; wherein the prediction model is trained based on historical usage data of information processing resources by the business;

[0009] Based on the predicted demand and information about existing information processing resources, resource configuration information is generated; wherein the resource configuration information is used to indicate recommended scaling operations for the existing information processing resources.

[0010] According to another aspect of an embodiment of this specification, a method for training a prediction model of information processing resources is provided, the method comprising:

[0011] Extracting auxiliary resource features in a time dimension from historical usage data of information processing resources by businesses before an auxiliary period, wherein the auxiliary period is earlier than the target period;

[0012] Inputting the auxiliary resource characteristics into an initial model to obtain an auxiliary predicted amount of information processing resources required for the business during the auxiliary period;

[0013] The model parameters of the initial model are adjusted based on the auxiliary prediction quantity to obtain a prediction model of information processing resources.

[0014] According to another aspect of the embodiments of this specification, a device for configuring information processing resources is provided, the device for configuring information processing resources comprising:

[0015] The first extraction module is used to extract the target resource characteristics in the time dimension based on the historical usage data of information processing resources by the business before the target period;

[0016] An input module, configured to input the target resource characteristics into a prediction model for information processing resources to obtain a predicted demand for information processing resources by the business during the target period; wherein the prediction model is trained based on historical usage data of information processing resources by the business;

[0017] An information generation module is used to generate resource configuration information based on the predicted demand and information about existing information processing resources; wherein the resource configuration information is used to indicate recommended scaling operations for the existing information processing resources.

[0018] According to another aspect of the embodiments of this specification, a training device for a prediction model of information processing resources is provided, the training device comprising:

[0019] a feature extraction module configured to extract auxiliary resource features in a time dimension based on historical usage data of information processing resources by businesses before an auxiliary period, wherein the auxiliary period is earlier than the target period;

[0020] A prediction module, configured to input the auxiliary resource characteristics into an initial model to obtain an auxiliary prediction amount of information processing resources required for the business during the auxiliary period;

[0021] An adjustment module is used to adjust the model parameters of the initial model based on the auxiliary prediction quantity to obtain a prediction model of information processing resources.

[0022] According to another aspect of the embodiments of this specification, there is provided a computing device, including: a memory and a processor;

[0023] The memory is used to store computer programs / instructions, and the processor is used to execute the computer programs / instructions. When the programs / instructions are executed by the processor, the steps of the above method are implemented.

[0024] According to another aspect of the embodiments of this specification, a computer-readable storage medium is provided, which stores a computer program / instruction, and the computer program / instruction implements the steps of the above method when executed by a processor.

[0025] According to another aspect of the embodiments of this specification, a computer program product is provided, comprising a computer program / instruction, which implements the steps of the above method when the computer program / instruction in the computer program product is executed by a processor.

[0026] In one embodiment of this specification, the target resource characteristics along the time dimension are extracted from the historical usage data of information processing resources by the business. Based on these target resource characteristics, a prediction model is used to predict the information processing resources required by the business within the target period, thereby obtaining a predicted demand. Furthermore, based on this predicted demand and information about existing information processing resources, recommended scaling operations for existing information processing resources are generated. This allows users to obtain an advance understanding of the approximate information processing resources required by the business within the target period and, based on these recommended scaling operations, to scale existing information processing resources accordingly, ensuring that the business can be processed normally within the target period, avoiding idle resources and waste, and thus saving business processing costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 This is a flow chart of a method for configuring information processing resources provided in one embodiment of this specification;

[0028] Figure 2 This is a diagram showing resource configuration information provided in an embodiment of this specification;

[0029] Figure 3 This is another display diagram of resource configuration information provided in an embodiment of this specification;

[0030] Figure 4 This is another diagram showing resource configuration information provided in an embodiment of this specification;

[0031] Figure 5 This is a flowchart of a method for training a prediction model of information processing resources provided in one embodiment of this specification;

[0032] Figure 6 This is a simplified flowchart of a model training method provided in one embodiment of this specification;

[0033] Figure 7 This is a schematic diagram of the structure of an information processing resource configuration device provided in one embodiment of this specification;

[0034] Figure 8 This is a schematic diagram of the structure of a training device for a prediction model of information processing resources provided in one embodiment of this specification;

[0035] Figure 9 This is a structural block diagram of a computing device provided in one embodiment of this specification. DETAILED DESCRIPTION

[0036] The following description sets forth many specific details to facilitate a thorough understanding of this specification. However, this specification can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the scope of this specification. Therefore, this specification is not limited to the specific implementations disclosed below.

[0037] The terms used in one or more embodiments of this specification are for the purpose of describing specific embodiments only and are not intended to limit one or more embodiments of this specification. The singular forms "a", "said" and "the" used in one or more embodiments of this specification and the appended claims are also intended to include plural forms, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in one or more embodiments of this specification refers to and includes any or all possible combinations of one or more associated listed items. The term "at least one" in one or more embodiments of this specification refers to "one or more" and "a plurality" refers to "two or more". The term "including" is an open description and should be understood as "including but not limited to", and may include other content on the basis of what has been described.

[0038] It should be understood that although the terms "first," "second," and the like may be used to describe various information in one or more embodiments of this specification, such information should not be limited to these terms. These terms are merely used to distinguish information of the same type from one another. For example, without departing from the scope of one or more embodiments of this specification, "first" may also be referred to as "second," and similarly, "second" may also be referred to as "first." Depending on the context, the word "if" as used herein may be interpreted as "at the time of," "when," or "in response to determining."

[0039] In addition, the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in one or more embodiments of this specification are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant standards and requirements, and corresponding operation entrances must be provided for users to choose to authorize or refuse.

[0040] With the advancement of computer technology, applications for various functions have become widely used. The operation of these applications often requires servers to provide corresponding services. For example, a client can send data to the server for processing. The server then processes the data sent by the client or obtained from other channels and feeds back the results to the client. A business may require the use of many applications in its normal operations (such as those for employee communication, production planning, and data statistical analysis). Servers need to provide the corresponding information processing resources to ensure the normal operation of these applications.

[0041] Currently, the use of various applications is complex. For example, some applications are only used during set time periods, some require continuous operation, and some are activated only when certain conditions are triggered. This makes it difficult for server operators and maintenance personnel to accurately determine the information processing resources required for each application, making it difficult to rationally plan server sizing. Furthermore, as a company's business constantly changes, the demand for applications also fluctuates accordingly, requiring constant improvement in server configuration to cope with uncertain peak demand for information processing resources. This inevitably leads to unnecessary resource waste. If the immutability of servers is relied upon to respond to changing demand, insufficient information processing resources may prevent applications from functioning properly when needed, resulting in business discontinuity.

[0042] Related technologies have introduced containerization and autoscaling capabilities for server resources. Autoscaling automatically adjusts server resource capacity based on actual application requirements to ensure application performance and availability. However, due to the inability to accurately determine business changes, in practice, server resources are only automatically expanded, not scaled back, to ensure business continuity. This leads to persistent resource waste and idleness.

[0043] The embodiments of this specification provide a method for configuring information processing resources, which can predict the information processing resources required for the business of an enterprise, help the enterprise better understand the current business volume and future development trends, and help the enterprise plan and adjust resources in advance to avoid waste and idle resources. This can effectively save costs, improve resource utilization, and reduce business interruptions and service quality degradation caused by insufficient resources. The embodiments of this specification also involve a configuration device for information processing resources, a training method for a prediction model of information processing resources, a training device for a prediction model of information processing resources, a computing device, a computer-readable storage medium, and a computer program product. The configuration device, training device, and computing device can all be a server or a terminal device.

[0044] Figure 1 This is a flow chart of a method for configuring information processing resources provided in one embodiment of this specification. This method can be applied to a configuration device for information processing resources. Figure 1 As shown, the configuration method of the information processing resource includes the following steps:

[0045] Step 102: Extract target resource features in the time dimension based on historical usage data of information processing resources by businesses before the target period.

[0046] During the normal operation of an enterprise, it is necessary to process a variety of businesses, and the processing of various businesses requires a certain amount of information processing resources. When it is necessary to predict the information processing resources required for a business, the configuration device of the information processing resources can obtain the historical usage data of the business on the information processing resources. For example, if the information processing resources for a target period are predicted, the historical usage data of the business on the information processing resources before the target period can be obtained. The historical usage data can be the usage data for a period of time before the target period, or it can be all usage data before the target period. The historical usage data may include data such as the carrier location, usage time, downtime, processing efficiency of the information processing resources used by the business, and the information processing resource capacity of the server.

[0047] In the embodiments of this specification, the business targeted by the forecast may include all businesses required by the enterprise, or may only include one or several specific businesses required by the enterprise. The historical usage data of information processing resources by the business may include the amount of information processing resources required by the business at multiple times. The information processing resources may include the number of virtual machines, the number of hosts, the number of hard disks, the number of central processing unit (CPU) cores, the memory size, the disk (or hard disk) size, etc.

[0048] The configuration device may process the acquired historical usage data, such as extracting target resource features related to the time dimension from the historical usage data. For example, the target resource features may include time series-related features, seasonal features, and historical development trend features. The target resource features may include the amount of information processing resources occupied by business processing.

[0049] Optionally, step 102 may include the following steps:

[0050] Step s1: Collect historical usage data of information processing resources by the business.

[0051] For example, during each use of information processing resources by a business, relevant usage data can be recorded in a server or other device. When the configuration device needs to make a prediction of information processing resources, it can obtain the historical usage data corresponding to the required period.

[0052] Step s2: perform data cleaning and denoising on the historical usage data to obtain backup data.

[0053] The configuration device may pre-process the acquired historical usage data to obtain backup data. For example, the pre-processing may include data cleaning and denoising, such as deleting duplicate data in the historical usage data and correcting erroneous data to ensure the accuracy and reliability of the data.

[0054] Step s3: extract features from the backup data in the time dimension to obtain target resource features.

[0055] The configuration device can obtain the conditions that the characteristics in the time dimension need to meet, and then extract the characteristics of the backup data based on the conditions to obtain the target resource characteristics. For example, the conditions can be the resource usage at a certain specific time, the resource usage at a certain business processing progress, etc.

[0056] Step 104: Input the target resource characteristics into the prediction model of the information processing resource to obtain the predicted demand for information processing resources by the business during the target period; wherein the prediction model is trained based on the historical usage data of the business for information processing resources.

[0057] The configuration device can analyze the target resource characteristics to determine the predicted demand for information processing resources by the business during the target period, that is, the predicted amount of information processing resources required by the business during the target period. In the embodiments of this specification, the configuration device can use a prediction model for information processing resources to analyze the target resource characteristics to determine the predicted demand.

[0058] The prediction model can be trained based on historical data on information processing resource usage by businesses. The businesses can be historical businesses of a reference enterprise, which can be the same enterprise as the business for which information processing resources are being predicted (e.g., the target enterprise). Alternatively, the reference enterprise can be different from the target enterprise; it only needs to have similarities between the reference enterprise's business and the target enterprise's business to serve as a reference for business prediction for the target enterprise.

[0059] For example, feature extraction can be performed on historical business usage data of information processing resources to extract resource features related to the time dimension. A portion of these resource features can then be used as training data, and another portion as validation data, to train the model until a training stop condition is met, resulting in a predictive model. For example, after training the model using the training data, validation data can be used to verify the model's predictive accuracy. When the model's predictive accuracy satisfies the conditions, the training stop condition is determined to have been met, resulting in a predictive model.

[0060] After obtaining the prediction model, the prediction model can be used to predict the demand for information processing resources by the business within a specified time period (e.g., a target time period) (e.g., to obtain a predicted demand). For example, the prediction model can be directly obtained by directly obtaining the predicted demand output by the prediction model by simply inputting the target resource characteristics obtained by extracting historical usage data of information processing resources by the business before the target time period into the prediction model.

[0061] In the embodiments of this specification, a prediction model is trained using historical data on actual information processing resource usage by enterprise businesses. This resulting prediction model can more accurately predict the business's demand for information processing resources during the target period, ensuring that the predicted demand for information processing resources by the business is relatively accurate and minimally deviates from the actual demand. Furthermore, model training utilizes time-dimensional features extracted from this historical usage data, which better characterize business development trends. Therefore, the resulting prediction model can more accurately predict business demand for information processing resources.

[0062] Step 106: Generate resource configuration information based on the predicted demand and information about existing information processing resources; wherein the resource configuration information is used to indicate recommended scaling operations for existing information processing resources.

[0063] After obtaining the predicted demand for information processing resources by the business during the target period, the configuration device can adjust the information processing resources for the servers providing services for each business based on the predicted demand. For example, the configuration device can compare the predicted demand with the server's existing information processing resources to determine whether to expand or reduce the existing information processing resources. This can effectively save costs, improve resource utilization, and reduce production interruptions and service quality degradation caused by insufficient resources.

[0064] If the predicted demand exceeds the available information processing resources, resource configuration information can be generated to recommend expansion of existing information processing resources. Based on this resource configuration information, staff can then expand server capacity, such as by activating previously unused information processing resources or purchasing new ones. Alternatively, the server can automatically activate new information processing resources based on this resource configuration information.

[0065] If the predicted demand is less than the existing information processing resources, resource configuration information can be generated to recommend scaling down the existing information processing resources. Accordingly, the staff can scale down the server based on the resource configuration information, or the server can automatically disable new information processing resources based on the resource configuration information.

[0066] Optionally, the configuration information may be presented to the user in the form of a chart. Figure 2 This is a diagram showing resource configuration information provided by an embodiment of this specification. The configuration device can be used Figure 2 The display diagram shown in the figure shows the original resource volume, predicted demand volume and the difference between the various information processing resources through a table. The difference can represent the resource configuration information. If the difference is greater than 0, it means that the existing information processing resources are scaled down, and the capacity to be reduced is the difference. Figure 2 As shown in the figure, it can be concluded that the number of virtual machines and hard disks does not need to be changed; the number of hosts needs to be reduced by 31, or 93.9%; the number of CPU cores needs to be reduced by 2232, or 97.9%; and the memory capacity needs to be reduced by 46.5TB (terabytes), or 96.9%. Figure 2 It also displays information on host density and hard disk density, allowing users to gain a more comprehensive understanding of information processing resources. Figure 2 The direction of the small black triangle in the chart indicates whether the value is rising or falling. If it is facing upward, the value is rising, and if it is facing downward, the value is falling.

[0067] Figure 3 This is another display diagram of resource configuration information provided in an embodiment of this specification. Figure 4 This is another display diagram of resource configuration information provided in an embodiment of this specification. Figure 3The figure shows the changing trends of the CPU request ratio and utilization. The dotted line represents the predicted demand for the CPU, which shows that the request for the CPU will increase in the future. Figure 4 The changing trend of the utilization of the virtual memory is shown, wherein the dotted line portion represents the predicted amount of the utilization of the virtual memory, from which it can be seen that the utilization of the virtual memory will decrease in the future.

[0068] In summary, in the configuration method of information processing resources provided in the embodiments of this specification, the target resource characteristics in the time dimension are extracted based on the historical usage data of information processing resources by the business. Based on the target resource characteristics, the information processing resources required by the business in the target period are predicted using a prediction model to obtain the predicted demand. Then, based on the predicted demand and the information of the existing information processing resources, a recommended scaling operation for the existing information processing resources is generated. In this way, the user can know in advance the information processing resources roughly required for the business in the target period, and can accordingly expand or shrink the existing information processing resources based on the recommended scaling operation to ensure that the business can be processed normally in the target period and avoid idleness and waste of resources, thereby saving the processing cost of the business.

[0069] In the embodiments of this specification, before predicting information processing resources (such as before step 102 or step 104), it is necessary to first perform model training to obtain a prediction model for information processing resources. The following describes a method for training the prediction model. This method can be used for a training device for a prediction model. The training device can be the same device as the above-mentioned configuration device, or it can be another device different from the above-mentioned configuration device. If the training device and the above-mentioned configuration device are the same device, the following training method can be part of the process that needs to be executed in the above-mentioned configuration method for information processing resources, and is executed before step 102 or step 104 of the above-mentioned configuration method. The flowchart under this method is not further illustrated in this specification.

[0070] In embodiments of this specification, the training device can extract auxiliary resource characteristics in the time dimension from historical information processing resource usage data for services prior to an auxiliary period, where the auxiliary period precedes the target period; input the auxiliary resource characteristics into an initial model to obtain an auxiliary prediction of the information processing resources required by services during the auxiliary period; and then adjust the model parameters of the initial model based on the auxiliary prediction to obtain a prediction model for information processing resources. The following details the training of this prediction model, with reference to the accompanying figures.

[0071] Figure 5 This is a flowchart of a method for training a prediction model of information processing resources provided by an embodiment of this specification, which can be used in a training device for a prediction model of information processing resources, such as Figure 5 As shown, the method may include the following steps:

[0072] Step 502: Acquire historical usage data of information processing resources by services before the auxiliary period, and usage data of information processing resources by services during the auxiliary period.

[0073] In the embodiments of this specification, the historical usage data of information processing resources by the business before the auxiliary period is referred to as the historical usage data corresponding to the auxiliary period, and the usage data of information processing resources by the business during the auxiliary period is referred to as actual usage data. The training device of the prediction model can first collect training data and obtain labels for the training data before performing model training. Each piece of training data can correspond to an auxiliary period, and the training data can be obtained based on the historical usage data of information processing resources by the business before the auxiliary period. The training device can also obtain labels for the corresponding training data based on the actual usage data of information processing resources by the business during the auxiliary period.

[0074] Assume that the auxiliary period is a single day, the business is the enterprise's internal communications business, and information processing resources include memory. In this case, the historical data on information processing resource usage by the business before the auxiliary period can refer to the memory usage of the communications business during the period before that day (e.g., a week); the data on information processing resource usage by the business during the auxiliary period refers to the memory usage of the communications business during that day.

[0075] In the embodiments of this specification, the training device can directly determine the acquired historical usage data and actual usage data as training data and corresponding labels, respectively. After acquiring the raw usage data, the training device can also perform data preprocessing on the usage data to obtain backup data, and then determine the required training data and labels based on the backup data. For details on the data preprocessing method, please refer to the relevant description in the aforementioned step 102.

[0076] The training device can first obtain backup historical usage data, such as all or part of the historical usage data of information processing resources used by the business before the training time. This historical usage data can be divided into multiple auxiliary time periods to obtain historical usage data, and actual usage data for each auxiliary time period can also be obtained.

[0077] Step 504: extract resource features in the time dimension from historical usage data of information processing resources by businesses before the auxiliary period to obtain auxiliary resource features corresponding to the auxiliary period.

[0078] The extraction of the auxiliary resource features is the same as the aforementioned method for extracting the target resource features. For details, please refer to the relevant introduction of extracting the target resource features based on the historical usage data in step 102, which will not be repeated here.

[0079] The auxiliary period may be earlier than the target period for which information processing resources prediction is required, and the auxiliary period is the period between target periods. In one embodiment, the length of the auxiliary period may be equal to the length of the target period. For example, if the server needs to be adjusted for resource configuration once every set period, the length of the auxiliary period and the target period may both be equal to the period. Optionally, the length of the auxiliary period may not be equal to the length of the target period, such as the length of the auxiliary period is equal to the length of multiple target periods, or the length of the target period is equal to the length of multiple auxiliary periods. When predicting the information processing resources for the target period in this way, the overall predicted amount of information processing resources within the target period can be obtained by combining the predicted amounts of information processing resources for multiple sub-periods.

[0080] The training device can extract features in the time dimension for the historical usage data corresponding to each auxiliary period to obtain corresponding auxiliary resource features (that is, the auxiliary resource features corresponding to the auxiliary period). The training device can also extract features in the time dimension for the actual usage data of each auxiliary period to obtain resource features corresponding to the actual usage data. The auxiliary resource features corresponding to the auxiliary period can be used as training data for the model, and the resource features of the actual usage data of the auxiliary period can be used as labels for the training data. The training device can also use the auxiliary resource features corresponding to some auxiliary periods and the resource features of the actual usage data as verification data to verify the accuracy of the model.

[0081] Optionally, the training device may first extract features in the time dimension for the spare historical usage data, and then divide the extracted features to obtain auxiliary resource features corresponding to multiple auxiliary time periods; in this method, the auxiliary resource features are also obtained by extracting the historical usage information corresponding to the auxiliary time periods.

[0082] Step 506: Based on the business characteristics of the business, select an initial model that matches the business characteristics from multiple candidate models.

[0083] In the embodiments of this specification, the prediction model is used to predict the demand for information processing resources by the business, so the prediction accuracy of the model is closely related to the characteristics of the business. The training device can select a suitable prediction algorithm based on the business characteristics of the business to perform model training, and different prediction algorithms can correspond to different candidate models. For example, the training device selects an initial model that matches the business characteristics from multiple candidate models based on the business characteristics of the business. The business characteristics may include the production needs of the enterprise and the data characteristics of the production data.

[0084] For example, the training device can select a corresponding initial model based on the periodic characteristics of the production data. In the case of periodic execution of the business, a larger initial model can be selected, and historical data over a longer period of time can be obtained for model training. In the case of non-periodic execution of the business, a smaller initial model can be selected, and historical data over a shorter period of time can be obtained for model training. The training device can perform data analysis based on the obtained production data to determine whether the data has periodicity.

[0085] In the embodiments of this specification, the training device can also obtain the enterprise's business execution plan to determine the model parameters of the initial model based on the business execution plan. For example, the business execution plan may include the enterprise's production plan, which may include the enterprise's worker schedules, production progress, etc. Using this production plan as a strong intervention measure in model training can ensure that the trained model produces more reasonable prediction results.

[0086] Step 508: Input the auxiliary resource characteristics into the initial model to obtain the auxiliary predicted amount of information processing resources required for the business during the auxiliary period.

[0087] Each piece of training data obtained by the training device may include auxiliary resource features corresponding to an auxiliary period. The training device may input each piece of training data obtained into the initial model to obtain the auxiliary predicted amount of information processing resources required for the business in the auxiliary period output by the initial model.

[0088] After step 508, the training model can adjust the model parameters of the initial model based on the auxiliary prediction quantity to obtain the prediction model of the information processing resources. For example, the training model can adjust the model parameters of the initial model through the following steps 510 to 514 to obtain the prediction model of the information processing resources.

[0089] Step 510: Compare the auxiliary prediction data with the actual usage data of the information processing resources during the auxiliary period to obtain a comparison result.

[0090] After obtaining a prediction result for each training data point, the training device can compare the prediction result with the label of the training data to determine whether the prediction result is accurate. If the difference between the prediction result and the label is greater than a threshold, it indicates that the prediction result is inaccurate and further training of the prediction model is required.

[0091] In the embodiments of this specification, since labels can be derived based on actual usage data of information processing resources during the auxiliary period, the auxiliary predictions can also be compared with this actual usage data, and the prediction model can be optimized based on the comparison results. For example, the training device can compare the auxiliary predictions with the resource characteristics of the actual usage data during the auxiliary period to fine-tune the initial model based on the comparison results. This allows the prediction results to be compared with actual business data, allowing timely identification of issues and adjustments to ensure the accuracy and practicality of the predictions.

[0092] Step 512: Display the auxiliary prediction information and receive analysis information for the auxiliary prediction.

[0093] When the training device obtains auxiliary predictions, it can display the auxiliary prediction information on a front-end display page to inform the user of the prediction results. The user can analyze the prediction results to determine whether they are reasonable and meet actual business needs, and then enter the corresponding analysis information. The training device can then receive the analysis information for the auxiliary predictions. This analysis information is also the user's evaluation of the prediction results.

[0094] Step 514: Adjust the model parameters of the initial model based on the comparison result and the analysis information to obtain a prediction model of information processing resources.

[0095] The above-mentioned comparison results and analysis information can be used as the basis for adjustment of model tuning. The embodiment of this specification does not limit the order of step 510 and step 512. Step 510 and step 512 can be executed simultaneously or any one of them can be executed first. Optionally, only step 510 or step 512 can be executed. Correspondingly, if only step 510 is executed, the model parameters of the initial model are adjusted only based on the comparison results in step 514 to obtain a prediction model of information processing resources. If only step 512 is executed, the model parameters of the initial model are adjusted only based on the analysis information in step 514 to obtain a prediction model of information processing resources. Based on each adjustment of the model parameters of the initial model, the training device can again execute the steps of adjusting the model parameters of the initial model in steps 508 to 514 until the training stop condition is met, and the model at this time is determined as the prediction model of information processing resources.

[0096] After retraining the prediction model based on the analysis information, the training device can resend the prediction results output by the retrained prediction model to the user. If at least two consecutive analysis messages indicate that the prediction results meet the requirements, the model's prediction accuracy can be considered to meet the requirements and model training is complete.

[0097] In the embodiment of this specification, the training device can also compare the obtained prediction results with the business execution plan to determine whether the prediction results match the business execution plan. If they do not match, the prediction model is retrained.

[0098] Step 516: When the business execution plan is updated, adjust the model parameters of the prediction model and retrain the prediction model.

[0099] In the embodiments of this specification, the training device can monitor changes to the business execution plan and, if the business execution plan is updated, adjust the model parameters of the prediction model to align with the new business execution plan. Steps 508 to 514 can then be re-executed to retrain the model. This ensures that sufficient information processing resources are available to support business processing within the business execution time specified in the business execution plan, avoiding waste and idleness of information processing resources outside of business execution time.

[0100] Figure 6 This is a simplified flow chart of a model training method provided in one embodiment of this specification. Figure 6 As shown, the training device can first collect historical data on the use of information processing resources by the business (that is, the above-mentioned historical usage data), and pre-process the collected data, and perform feature extraction on the pre-processed data, such as feature extraction in the time dimension. Afterwards, the prediction model can be trained and tuned based on the extracted features. During the tuning process of the prediction model, the generated capacity prediction report can be sent to the user for viewing, such as showing the predicted business demand for information processing resources (such as the above-mentioned auxiliary prediction amount). The user can evaluate the report and use the user evaluation as strong intervention information for model tuning, so as to perform model tuning based on the user evaluation. The training device can also use the production plan (that is, the above-mentioned business execution plan) as strong intervention information for model tuning, so as to perform model tuning based on the production plan. After the model tuning is completed, the capacity prediction report generated subsequently can be used to expand or reduce the capacity of information processing resources to ensure that the capacity of information processing resources meets business needs, avoid resource waste, and ensure low business processing costs.

[0101] In summary, in the training method of the prediction model of information processing resources provided in the embodiment of this specification, the historical usage data of information processing resources for the business can be used to extract auxiliary resource features in the time dimension, and the prediction model can be trained based on the auxiliary resource features. And the analysis information and business execution plan can be used to strongly intervene in the tuning of the model to ensure that the obtained prediction model has a high degree of matching with the business. The prediction model obtained in this way can more accurately predict the business's demand for information processing resources, so that the user can know in advance the information processing resources roughly required for the business in the target period, and can expand or shrink the existing information processing resources accordingly to ensure that the business can be processed normally and avoid idleness and waste of resources, saving business processing costs.

[0102] Corresponding to the above method embodiments, this specification also provides an embodiment of a device for configuring information processing resources. Figure 7 This is a schematic diagram of the structure of a configuration device for information processing resources provided in one embodiment of this specification. Figure 7 As shown, the configuration device of the information processing resource includes:

[0103] The first extraction module 701 is configured to extract target resource features in a time dimension based on historical usage data of information processing resources by businesses before a target period;

[0104] Input module 702, configured to input target resource characteristics into a prediction model for information processing resources to obtain a predicted demand for information processing resources by a business within a target period; wherein the prediction model is trained based on historical usage data of information processing resources by the business;

[0105] The information generation module 703 is used to generate resource configuration information based on the predicted demand and information about existing information processing resources; wherein the resource configuration information is used to indicate recommended scaling operations for existing information processing resources.

[0106] Optionally, the information processing resource configuration device further includes:

[0107] A second extraction module is configured to extract auxiliary resource features in a time dimension from historical usage data of information processing resources by businesses before an auxiliary period before inputting the target resource features into the prediction model of the information processing resources; wherein the auxiliary period is earlier than the target period;

[0108] An acquisition module, used to input auxiliary resource characteristics into the initial model to obtain the auxiliary prediction amount of information processing resources required for the business during the auxiliary period;

[0109] The first adjustment module is used to adjust the model parameters of the initial model based on the auxiliary prediction quantity to obtain a prediction model of the information processing resources.

[0110] Optionally, the first adjustment module is configured to:

[0111] Comparing the auxiliary prediction amount with the actual usage data of the information processing resources during the auxiliary period to obtain a comparison result;

[0112] Based on the comparison results, the model parameters of the initial model are adjusted to obtain a prediction model of information processing resources.

[0113] Optionally, the first adjustment module is configured to:

[0114] Displaying information on auxiliary prediction and receiving analytical information on the auxiliary prediction;

[0115] The model parameters of the initial model are adjusted based on the analysis information to obtain a prediction model of information processing resources.

[0116] Optionally, the information processing resource configuration device further includes:

[0117] a determination module, configured to determine model parameters of the initial model based on the business execution plan before inputting the auxiliary resource features into the initial model;

[0118] The second adjustment module is used to adjust the model parameters of the initial model based on the auxiliary prediction quantity to obtain the prediction model of the information processing resources, and then adjust the model parameters of the prediction model and retrain the prediction model when the business execution plan is updated.

[0119] Optionally, the information processing resource configuration device further includes:

[0120] The selection module is used to select an initial model that matches the business characteristics from multiple candidate models based on the business characteristics of the business before inputting the auxiliary resource characteristics into the initial model.

[0121] Optionally, the first extraction module 701 is used to:

[0122] Collect historical data on the business's use of information processing resources;

[0123] Perform data cleaning and denoising on historical usage data to obtain backup data;

[0124] Feature extraction is performed on the backup data in the time dimension to obtain target resource features.

[0125] In summary, in the configuration device for information processing resources provided in the embodiments of this specification, the target resource characteristics in the time dimension are extracted based on the historical usage data of information processing resources by the business. Based on the target resource characteristics, a prediction model is used to predict the information processing resources required by the business within the target period to obtain a predicted demand. Then, based on the predicted demand and information about existing information processing resources, a recommended scaling operation for the existing information processing resources is generated. In this way, the user can know in advance the information processing resources roughly required for the business within the target period, and can accordingly expand or shrink the existing information processing resources based on the recommended scaling operation to ensure that the business can be processed normally within the target period and avoid idleness and waste of resources, thereby saving business processing costs.

[0126] Corresponding to the above method embodiments, this specification also provides an embodiment of a training device for a prediction model of information processing resources. Figure 8 This is a structural diagram of a training device for a prediction model of information processing resources provided in one embodiment of this specification. Figure 8 As shown, the training device includes:

[0127] Feature extraction module 801 is used to extract auxiliary resource features in the time dimension based on historical usage data of information processing resources by businesses before the auxiliary period, where the auxiliary period is earlier than the target period;

[0128] Prediction module 802, used to input auxiliary resource characteristics into the initial model to obtain auxiliary prediction amount of information processing resources required by the business in the auxiliary period;

[0129] The first adjustment module 803 is used to adjust the model parameters of the initial model based on the auxiliary prediction quantity to obtain a prediction model of the information processing resources.

[0130] Optionally, the first adjustment module 803 is configured to:

[0131] Comparing the auxiliary prediction amount with the actual usage data of the information processing resources during the auxiliary period to obtain a comparison result;

[0132] Based on the comparison results, the model parameters of the initial model are adjusted to obtain a prediction model of information processing resources.

[0133] Optionally, the first adjustment module 803 is configured to:

[0134] Displaying information on auxiliary prediction and receiving analytical information on the auxiliary prediction;

[0135] The model parameters of the initial model are adjusted based on the analysis information to obtain a prediction model of information processing resources.

[0136] Optionally, the training device further comprises:

[0137] a determination module, configured to determine model parameters of the initial model based on the business execution plan before inputting the auxiliary resource features into the initial model;

[0138] The second adjustment module is used to adjust the model parameters of the initial model based on the auxiliary prediction quantity to obtain the prediction model of the information processing resources, and then adjust the model parameters of the prediction model and retrain the prediction model when the business execution plan is updated.

[0139] Optionally, the training device further comprises:

[0140] The selection module is used to select an initial model that matches the business characteristics from multiple candidate models based on the business characteristics of the business before inputting the auxiliary resource characteristics into the initial model.

[0141] Optionally, the feature extraction module 801 is used to:

[0142] Collect historical data on the use of information processing resources by businesses before the auxiliary period;

[0143] Perform data cleaning and denoising on the historical usage data to obtain backup data;

[0144] Feature extraction is performed on the backup data in the time dimension to obtain auxiliary resource features.

[0145] In summary, in the training method of the prediction model of information processing resources provided in the embodiment of this specification, the historical usage data of information processing resources for the business can be used to extract auxiliary resource features in the time dimension, and the prediction model can be trained based on the auxiliary resource features. And the analysis information and business execution plan can be used to strongly intervene in the tuning of the model to ensure that the obtained prediction model has a high degree of matching with the business. The prediction model obtained in this way can more accurately predict the business's demand for information processing resources, so that the user can know in advance the information processing resources roughly required for the business in the target period, and can expand or shrink the existing information processing resources accordingly to ensure that the business can be processed normally and avoid idleness and waste of resources, saving business processing costs.

[0146] The various embodiments in this specification are described in a progressive manner. Similar portions between the various embodiments can be referenced to each other. Each embodiment focuses on the differences from the other embodiments. In particular, the description of the information processing resource configuration device is relatively simple, as it is fundamentally similar to the information processing resource configuration method embodiment. For relevant portions, refer to the description of the information processing resource configuration method embodiment.

[0147] Figure 99 is a block diagram of a computing device according to an embodiment of the present disclosure. Components of the computing device 900 include, but are not limited to, a memory 910 and a processor 920. The processor 920 is connected to the memory 910 via a bus 930, and a database 950 is used to store data.

[0148] The computing device 900 also includes an access device 940 that enables the computing device 900 to communicate via one or more networks 960. Examples of such networks include a public switched telephone network (PSTN), a local area network (LAN), a wide area network (WAN), a personal area network (PAN), or a combination of communication networks such as the Internet. The access device 940 may include one or more of any type of network interface (e.g., a network interface card (NIC)) whether wired or wireless, such as an IEEE 802.11 wireless local area network (WLAN) wireless interface, a Worldwide Interoperability for Microwave Access (Wi-MAX) interface, an Ethernet interface, a universal serial bus (USB) interface, a cellular network interface, a Bluetooth interface, or a near field communication (NFC) interface.

[0149] In one embodiment of the present specification, the above components of the computing device 900 and Figure 9 Other components not shown in the figure may also be connected to each other, for example, via a bus. Figure 9 The computing device structure block diagram shown is for illustrative purposes only and is not intended to limit the scope of this specification. Those skilled in the art may add or replace other components as needed.

[0150] The computing device 900 may be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., a tablet computer, a personal digital assistant, a laptop computer, a notebook computer, a netbook computer, etc.), a mobile phone (e.g., a smartphone), a wearable computing device (e.g., a smartwatch, smart glasses, etc.), or other types of mobile devices, or a stationary computing device such as a desktop computer or a personal computer (PC). The computing device 900 may also be a mobile or stationary server.

[0151] The processor 920 is used to execute the following computer executable instructions, which, when executed by the processor, implement the above Figure 1 Or the method shown in 2.

[0152] As for the computing device embodiment, since it is basically similar to the above method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the above method embodiment.

[0153] One embodiment of this specification also provides a computer-readable storage medium, which stores computer instructions, which, when executed by a processor, implement the steps of the configuration method of the above-mentioned information processing resources. The computer instructions include computer program code, which can be in source code form, object code form, executable file or some intermediate form. The computer-readable storage medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable storage medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable storage media do not include electric carrier signals and telecommunication signals.

[0154] One embodiment of the present specification further provides a computer program product, comprising a computer program / instruction, which, when executed in a computer, causes the computer to execute the steps of the above-mentioned method for configuring information processing resources.

[0155] As for the computer-readable storage medium embodiment and the computer program product embodiment, since they are basically similar to the above-mentioned method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the above-mentioned method embodiment.

[0156] The foregoing description of this specification describes specific embodiments. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that described in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order shown or the sequential order to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0157] It should be noted that the above description is of a specific embodiment of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-tasking and parallel processing are also possible or may be advantageous. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the embodiments of this specification.

[0158] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0159] The preferred embodiments disclosed above are intended only to help illustrate this specification. The optional embodiments do not exhaustively describe all details, nor do they limit the invention to the specific embodiments described. Obviously, many modifications and variations are possible based on the content of the embodiments described herein. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the embodiments, thereby enabling those skilled in the art to better understand and utilize this specification.

Claims

1. A method for configuring information processing resources, characterized in that: The method comprises: Extract the target resource characteristics in the time dimension based on the historical usage data of information processing resources by businesses before the target period; Inputting the target resource characteristics into a prediction model for information processing resources to obtain a predicted demand for information processing resources by the business during the target period; wherein the prediction model is trained based on historical usage data of information processing resources by the business; Based on the predicted demand and information about existing information processing resources, resource configuration information is generated; wherein the resource configuration information is used to indicate recommended scaling operations for the existing information processing resources.

2. The method according to claim 1, characterized in that Before inputting the target resource characteristics into the prediction model of the information processing resource, the method further includes: Extracting auxiliary resource features in a time dimension from historical usage data of information processing resources by businesses before an auxiliary period, wherein the auxiliary period is earlier than the target period; Inputting the auxiliary resource characteristics into an initial model to obtain an auxiliary predicted amount of information processing resources required for the business during the auxiliary period; The model parameters of the initial model are adjusted based on the auxiliary prediction quantity to obtain a prediction model of information processing resources.

3. The method according to claim 2, characterized in that The adjusting the model parameters of the initial model based on the auxiliary prediction quantity to obtain a prediction model of information processing resources includes: Comparing the auxiliary predicted amount with actual usage data of information processing resources during the auxiliary period to obtain a comparison result; The model parameters of the initial model are adjusted based on the comparison results to obtain a prediction model of information processing resources.

4. The method according to claim 2, characterized in that The adjusting the model parameters of the initial model based on the auxiliary prediction quantity to obtain a prediction model of information processing resources includes: displaying information of the auxiliary prediction measurement and receiving analysis information of the auxiliary prediction measurement; The model parameters of the initial model are adjusted based on the analysis information to obtain a prediction model of information processing resources.

5. The method according to any one of claims 2 to 4, characterized in that: Before inputting the auxiliary resource features into the initial model, the method further includes: Determining model parameters of the initial model based on the business execution plan; After adjusting the model parameters of the initial model based on the auxiliary prediction quantity to obtain the prediction model of the information processing resources, the method further includes: When the business execution plan is updated, the model parameters of the prediction model are adjusted and the prediction model is retrained.

6. The method according to any one of claims 2 to 4, characterized in that: Before inputting the auxiliary resource features into the initial model, the method further includes: Based on the business characteristics of the business, an initial model matching the business characteristics is selected from multiple candidate models.

7. The method according to claim 1, characterized in that The extraction of target resource characteristics in the time dimension based on historical usage data of information processing resources by businesses before the target period includes: Collect historical data on the business's use of information processing resources; Performing data cleaning and denoising on the historical usage data to obtain backup data; Feature extraction is performed on the backup data in a time dimension to obtain target resource features.

8. A method for training a prediction model for information processing resources, characterized in that: The method comprises: Extracting auxiliary resource features in a time dimension from historical usage data of information processing resources by businesses before an auxiliary period, wherein the auxiliary period is earlier than the target period; Inputting the auxiliary resource characteristics into an initial model to obtain an auxiliary predicted amount of information processing resources required for the business during the auxiliary period; The model parameters of the initial model are adjusted based on the auxiliary prediction quantity to obtain a prediction model of information processing resources.

9. A device for configuring information processing resources, characterized in that: The information processing resource configuration device includes: The first extraction module is used to extract the target resource characteristics in the time dimension based on the historical usage data of information processing resources by the business before the target period; An input module, configured to input the target resource characteristics into a prediction model for information processing resources to obtain a predicted demand for information processing resources by the business during the target period; wherein the prediction model is trained based on historical usage data of information processing resources by the business; An information generation module is used to generate resource configuration information based on the predicted demand and information about existing information processing resources; wherein the resource configuration information is used to indicate recommended scaling operations for the existing information processing resources.

10. A training device for a prediction model of information processing resources, characterized in that: The training device comprises: a feature extraction module configured to extract auxiliary resource features in a time dimension based on historical usage data of information processing resources by businesses before an auxiliary period, wherein the auxiliary period is earlier than the target period; A prediction module, configured to input the auxiliary resource characteristics into an initial model to obtain an auxiliary prediction amount of information processing resources required for the business during the auxiliary period; An adjustment module is used to adjust the model parameters of the initial model based on the auxiliary prediction quantity to obtain a prediction model of information processing resources.

11. A computing device, characterized in that include: memory and processor; The memory is used to store computer programs / instructions, and the processor is used to execute the computer programs / instructions. When the programs / instructions are executed by the processor, the method according to any one of claims 1 to 8 is implemented.

12. A computer-readable storage medium, characterized in that A computer program / instruction is stored, and when the computer program / instruction is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

13. A computer program product, characterized in that The computer program product comprises a computer program / instruction, and when the computer program / instruction in the computer program product is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images