Hardware resource allocation method and device, equipment, storage medium and program product
By acquiring and predicting multi-dimensional characteristic data of the business, and dynamically adjusting the allocation of hardware resources, the shortcomings of the traditional data platform resource allocation mechanism are solved, and the stable operation of the business system and timely adjustment of resources are achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- INDUSTRIAL AND COMMERCIAL BANK OF CHINA
- Filing Date
- 2026-02-04
- Publication Date
- 2026-06-19
AI Technical Summary
Traditional data middleware hardware resource allocation mechanisms are difficult to meet the hardware requirements of different scenarios, resulting in business systems experiencing response delays or crashes during peak periods.
By acquiring multi-dimensional feature data of the target business, the scenario to which it belongs can be determined, and hardware resource allocation can be dynamically adjusted based on feature data of predicted future time, so as to realize scenario perception and trend perception, and dynamically adjust hardware resource allocation.
It improves the reliability and smoothness of business system operation, avoids business system lag or crashes caused by insufficient hardware resources, and improves the timeliness of hardware resource allocation.
Smart Images
Figure CN122240293A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of big data, and in particular to a hardware resource allocation method, apparatus, device, storage medium, and program product. Background Technology
[0002] With the digital transformation of various industries, their business models are rapidly evolving towards digitalization. To ensure the stable operation of these businesses, data platforms are often deployed within them. These digital platforms centrally manage hardware resources and allocate them to various business units, providing them with basic computing, storage, and other capabilities.
[0003] Currently, traditional data platforms typically employ a pre-configured resource allocation mechanism, which pre-allocates hardware resources to run each business based on its hardware requirements. However, pre-allocated hardware resources are insufficient to meet the hardware demands of different scenarios. For example, during peak business periods, pre-allocated hardware resources may fail to handle sudden surges in traffic, leading to system delays or crashes. Summary of the Invention
[0004] This application provides a hardware resource allocation method, apparatus, device, storage medium, and program product. By dynamically adjusting the hardware resources allocated to each business according to the business scenario, an adaptive hardware resource allocation mechanism is realized, which avoids business system response delays or crashes caused by insufficient hardware resources and improves the smoothness of business operation.
[0005] Firstly, this application provides a hardware resource allocation method, including:
[0006] Acquire multi-dimensional feature data of the target business;
[0007] Based on the multi-dimensional feature data of the target business, determine the target scenario to which the target business belongs;
[0008] Based on the multi-dimensional feature data of the target business and the target scenario, predict the multi-dimensional feature data of the target business in the future time.
[0009] If the multi-dimensional feature data or the multi-dimensional feature data of future time meet the preset adjustment conditions corresponding to the target scenario, then the hardware resources allocated to the target business will be adjusted.
[0010] Secondly, this application provides a hardware resource allocation device, comprising:
[0011] The acquisition module is used to acquire multi-dimensional feature data of the target business.
[0012] The scenario determination module is used to determine the target scenario to which the target business belongs based on the multi-dimensional feature data of the target business;
[0013] The prediction module is used to predict the multi-dimensional feature data of the target business in the future based on the multi-dimensional feature data of the target business and the target scenario;
[0014] The resource adjustment module is used to adjust the hardware resources allocated to the target business if the multi-dimensional feature data or multi-dimensional feature data of future time meet the preset adjustment conditions corresponding to the target scenario.
[0015] Thirdly, this application provides an electronic device, including: a processor, and a memory communicatively connected to the processor;
[0016] The memory stores the instructions that the computer executes;
[0017] The processor executes computer-executable instructions stored in memory to implement the method provided in the first aspect above.
[0018] Fourthly, this application provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the method provided in the first aspect above.
[0019] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the method provided in the first aspect above.
[0020] The hardware resource allocation method, apparatus, device, storage medium, and program product provided in this application accurately identify the scenario to which the target business belongs based on acquired multi-dimensional feature data of the target business, and predict the multi-dimensional feature data in the future based on the scenario. This achieves scenario perception and trend perception of the target business, and can uncover the hardware resource requirements of the target business through the perceived content. Then, driven by the scenario, when the hardware resources allocated to the target business do not meet the current or future hardware resource requirements, that is, when the multi-dimensional feature data meets the adjustment conditions, the hardware resources allocated to the target business are dynamically adjusted. This application realizes scenario-based elastic allocation of hardware resources, avoiding the problem of business operation lag or crash due to insufficient hardware resources, thereby improving the reliability of business system operation. At the same time, by predicting and analyzing the multi-dimensional feature data in the future, and then actively adjusting hardware resources based on the prediction results, it overcomes the passive adjustment after an anomaly occurs and improves the timeliness of hardware resource allocation. Attached Figure Description
[0021] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.
[0022] Figure 1A flowchart illustrating a hardware resource allocation method provided in an embodiment of this application;
[0023] Figure 2 A flowchart illustrating another hardware resource allocation method provided in an embodiment of this application;
[0024] Figure 3 This is a schematic diagram of the structure of a hardware resource allocation device provided in an embodiment of this application;
[0025] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.
[0026] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation
[0027] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.
[0028] It should be noted that 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, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, storage, use, processing, transmission, provision, disclosure, and application of the relevant data all comply with the relevant laws, regulations, and standards of the relevant countries and regions, have taken necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation access points for users to choose to authorize or refuse.
[0029] Furthermore, the technical solution involved in this application, which involves big data analysis of user information (including but not limited to personal biometrics, identity data, consumption data, asset data, electronic terminal operation data, etc.) and the use of artificial intelligence technology for automated decision-making, and makes decisions that have a significant impact on personal rights based on the results of automated decision-making, provides users with corresponding operation entry points for users to choose to agree to or reject the results of automated decision-making; if the user chooses to reject, the process will proceed to the expert decision-making process.
[0030] It should be noted that the hardware resource allocation method, apparatus, device, storage medium and program product provided in this application can be used in the field of big data, or in any field other than big data. The application field of the hardware resource allocation method, apparatus, device, storage medium and program product in this application is not limited.
[0031] This application is applied to scenarios where hardware resources are allocated to various business processes within a business system. Specifically, it can be applied to a data platform within a business system. For enterprise-level business systems that support multiple business processes, each relying on hardware resources for operation, the system needs to allocate hardware resources to each process. This includes providing CPU, memory, and storage space to support basic functions such as data computation and storage. The data platform is a module within the business system used to manage hardware resources. It centrally manages hardware resources and allocates them to various business processes to support their operation.
[0032] Currently, data platforms typically allocate fixed hardware resources to each business unit before the business system runs, based on a pre-set resource allocation mechanism. However, the hardware resources required by a business unit change dynamically according to the volume of business activity. Fixed hardware resources often cannot meet the hardware requirements of different scenarios, leading to delays or crashes in the business system. For example, in financial business systems, user service businesses need to process massive amounts of data during peak sales periods, such as processing a large number of orders and frequently updating data in the database. Using fixed hardware resources may not be able to support the processing of such massive amounts of data, resulting in delays or crashes in the business system and impacting business operations.
[0033] The hardware resource allocation method provided in this application aims to solve the aforementioned technical problems of the prior art. By determining the business scenario and dynamically adjusting the hardware resources allocated to each business according to the hardware resource requirements of each scenario, dynamic hardware resource adjustment is achieved. This avoids the problem that fixed hardware resources cannot meet the needs of different scenarios, which could lead to business system response delays or crashes, thereby improving the reliability of business system operation and ensuring the smoothness of business operation.
[0034] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.
[0035] Figure 1This is a flowchart illustrating a hardware resource allocation method provided in an embodiment of this application. The method provided in this application can be executed by an electronic device with corresponding processing capabilities, such as a data platform in a business system, or an electronic device communicatively connected to the data platform. Figure 1 As shown, the method provided in this embodiment includes:
[0036] Step S101: Obtain multi-dimensional feature data of the target business.
[0037] The target business refers to the business covered by the business system, including but not limited to operational support business, user business, and supply chain interaction business. Taking a business system in the financial sector as an example, the target business includes financial service business for various users, business that provides integrated financial services between upstream and downstream enterprises in the industry chain, and back-office support business that supports the compliant, undefined, and efficient operation of the aforementioned businesses.
[0038] The multi-dimensional characteristic data of the target business refers to multi-dimensional data reflecting the dynamic characteristics of the business, collected from the business system, including but not limited to business type and hardware resource usage data. Hardware resource usage data includes at least one of the following: CPU utilization, memory utilization, storage space utilization, network utilization, and system load of the target business. Hardware resource usage data can indirectly reflect whether the currently allocated hardware resources can support the operation of the target business process. The business type can determine the possible scenarios in which the target business may be located.
[0039] In this embodiment, the business system can collect multi-dimensional feature data of each business by embedding data collection code or components in the business system.
[0040] In this step, multi-dimensional feature data within a preset time period prior to the current moment is obtained from the business system.
[0041] In some embodiments, the multi-dimensional characteristic data of the target business also includes business processing data, such as request volume. Request volume can reflect the activity level of the target business.
[0042] Optional, multi-dimensional feature data includes hardware resource usage data and user access data.
[0043] User access data refers to data generated when users interact with target business functions. This includes user clicks, personalized recommendation frequency, etc.
[0044] In this embodiment, data logs from the business system can be obtained, and user access data can be obtained based on the data logs.
[0045] By acquiring user access data, the characteristics of the target business are more comprehensively covered, enabling more accurate analysis of the target business's operation and thus making the subsequent scenario determination more accurate.
[0046] Optional, multi-dimensional feature data includes hardware resource usage data and historical resource adjustment data.
[0047] Historical resource adjustment data refers to data generated when the target business makes resource adjustments within a historical period.
[0048] For example, historical resource adjustment data includes the frequency of resource adjustments for the target business within a historical time period, the amount of resources adjusted, etc.
[0049] In one example, for the target business in the operations support category, the multi-dimensional feature data also includes the frequency of operations and maintenance updates. The frequency of operations and maintenance updates refers to the frequency at which the functions or parameters of the target business are maintained and updated.
[0050] Step S102: Based on the multi-dimensional feature data of the target business, determine the target scenario to which the target business belongs.
[0051] A scenario is information used to represent the operational environment during business execution. Multiple scenarios can be pre-defined, such as operational support scenarios, user scenarios, and supply chain interaction scenarios. Each scenario can be further categorized into multiple levels based on the importance and / or frequency of the business; for example, operational support scenarios can be further divided into high-frequency, medium-frequency, and low-frequency operational support scenarios.
[0052] In this step, based on the multi-dimensional feature data of the target business, a target scenario that matches the multi-dimensional feature data of the target business is determined from a variety of pre-set scenarios.
[0053] For example, based on the data range in which the multi-dimensional feature data of the target business falls, the target scenario to which the target business belongs is determined. The data range is a predefined range in which the multi-dimensional feature data of the business may fall under each scenario.
[0054] For example, multi-dimensional feature data of the target business can be input into a classification model. The classification model extracts features from the multi-article feature data and calculates the similarity with the features of each pre-learned scenario. The scenario corresponding to the feature with the highest similarity is determined as the target scenario.
[0055] The features of each scenario learned in advance are input into the classification model during the training phase of the classification model by inputting multi-dimensional feature data samples of the business under each scenario, so that the classification model learns the typical features of each scenario. These typical features are the features of each scenario.
[0056] Step S103: Based on the multi-dimensional feature data of the target business and the target scenario, predict the multi-dimensional feature data of the target business in the future time.
[0057] In this step, based on the changing patterns of multi-dimensional feature data indicated by the target scenario and the multi-dimensional data of the target business, multi-dimensional feature data for future time is predicted.
[0058] For example, if the target scenario is a high-frequency operation support scenario and the target scenario indicates an increasing frequency change trend, then multi-dimensional feature data of the future time is calculated based on the time interval with the future time and the change trend.
[0059] For example, if the target scenario is a user scenario that is about to enter a major promotion period, historical changes can be obtained and combined with multi-dimensional feature data of the target business to predict multi-dimensional feature data for future time.
[0060] Step S104: If the multi-dimensional feature data or the multi-dimensional feature data of future time meet the preset adjustment conditions corresponding to the target scenario, then adjust the hardware resources allocated to the target business.
[0061] The preset adjustment conditions are the pre-defined correspondence between the data range of multi-dimensional feature data and the resource adjustment strategy for each scenario. For example, the preset adjustment conditions include, for the first scenario, if the multi-dimensional feature data is within the first data range, such as if the request volume is greater than or equal to the preset request volume, then the hardware resources allocated to the target service will be increased.
[0062] In this step, for the target scenario, preset adjustment conditions corresponding to the target scenario are determined from the preset adjustment conditions. Multi-dimensional feature data and future time-based multi-dimensional feature data are matched against the preset adjustment conditions corresponding to the target scenario to determine if any preset adjustment conditions are met. If so, the hardware resources for the child's target service are adjusted and allocated according to the resource adjustment strategy corresponding to the preset adjustment conditions.
[0063] For example, in a user-based scenario, the preset adjustment conditions may include a request volume greater than or equal to a request volume threshold, and / or, user access data greater than or equal to a user threshold. The resource adjustment strategy corresponding to these preset adjustment conditions could be to increase the number of CPU nodes allocated to the target service.
[0064] For operations support scenarios, the preset adjustment conditions can include an operations update frequency greater than or equal to an update frequency threshold. The resource adjustment strategy corresponding to this preset adjustment condition can be to increase the memory allocated to the target service.
[0065] In some embodiments, the preset adjustment conditions are flexibly adjusted according to the scenario. For example, for user-related scenarios, the preset adjustment conditions include request volume being greater than or equal to a request volume threshold. The request volume threshold varies for specific user-related scenarios. For instance, the request volume threshold for high-frequency user-related scenarios is lower than that for medium-frequency user-related scenarios, which is lower than that for low-frequency user-related scenarios. This approach provides more sufficient hardware resources for higher-frequency target services, ensuring the smooth operation of these services.
[0066] The hardware resource allocation method provided in this embodiment accurately identifies the scenario to which the target service belongs based on the acquired multi-dimensional feature data of the target service, and predicts the multi-dimensional feature data for future time based on the scenario. This achieves scenario perception and trend perception of the target service, and the hardware resource requirements of the target service can be mined through the perceived content. Then, driven by the scenario, when the hardware resources allocated to the target service do not meet the current or future hardware resource requirements, that is, when the multi-dimensional feature data meets the adjustment conditions, the hardware resources allocated to the target service are dynamically adjusted. This application realizes scenario-based elastic allocation of hardware resources, avoiding the problem of business operation lag or crash due to insufficient hardware resources, thereby improving the reliability of business system operation. At the same time, by predicting and analyzing the multi-dimensional feature data for future time, and then actively adjusting hardware resources based on the prediction results, it overcomes the passive adjustment after an anomaly occurs and improves the timeliness of hardware resource allocation.
[0067] Figure 2 This is a flowchart illustrating another hardware resource allocation method provided in an embodiment of this application. The method provided in this embodiment is... Figure 1 Based on the illustrated embodiment, steps S102-S104 are further refined. For example... Figure 2 As shown, the method provided in this embodiment includes:
[0068] Step S201: Obtain multi-dimensional feature data of the target business.
[0069] Step S202: Input the multi-dimensional feature data of the target business into the scene classification model based on the long short-term memory network model to obtain the target scene to which the target business belongs.
[0070] The scene classification model is used to input multi-dimensional feature data into the long short-term memory network layer, extract feature vectors from the multi-dimensional feature data, and input the feature vectors into the fully connected layer to map the feature vectors to the classification space.
[0071] Long Short-Term Memory (LSTM) network models are a type of deep learning model that can effectively capture long-range dependencies in data.
[0072] In this embodiment, a scene classification model is constructed using a Long Short-Term Memory (LSTM) network framework and then trained. Specifically, multi-dimensional feature data of historical business data is obtained from the business system as training samples, and corresponding target scene labels are added to them. The training samples are input into the constructed scene classification model to obtain classification results. Based on the classification results and target scene labels, the loss value of the scene classification model is calculated. If the loss value meets the convergence condition, or the number of iterations is greater than or equal to the preset number, the scene classification model training is complete. If the loss value does not meet the convergence condition, and the number of iterations is less than the preset number, the parameters of the scene classification model are adjusted based on a preset optimization algorithm, such as gradient descent. After adjustment, the steps of inputting the training samples into the constructed scene classification model to obtain classification results are repeated.
[0073] In this step, multi-dimensional feature data of the target business is input into a pre-trained scene classification model. The scene classification model inputs the multi-dimensional feature data into a Long Short-Term Memory (LSTM) network layer built based on the trained parameters, extracting feature vectors from the multi-dimensional feature data. The extracted feature vectors are then input into a fully connected layer built based on the trained parameters, mapping the feature vectors to a classification space. This classification space includes various scenes; mapping the feature vectors to the classification space determines the target scene corresponding to the feature vectors.
[0074] Step S203: Obtain the historical feature data change trend corresponding to the target scene.
[0075] In this step, historical feature data corresponding to the target scenario for a specific time period can be obtained from the business system. For example, extracting historical feature data for the same period of the previous year, N days before and after the current moment. Based on the historical feature data for the historical time period, statistical analysis of the changing trends of the historical feature data is performed.
[0076] Step S204: Based on the changing trends of multi-dimensional feature data and historical feature data, predict the multi-dimensional feature data of the target business in the future.
[0077] In this step, guided by the changing trends of historical feature data, the changes in multi-dimensional feature data over future time are calculated to determine the multi-dimensional feature data of the target business over future time.
[0078] Step S205: If the multi-dimensional feature data or the multi-dimensional feature data of future time meets the resource expansion conditions, then the expanded resources are fully allocated to the hardware resources of the target business.
[0079] The preset adjustment conditions include resource expansion conditions and resource reduction conditions.
[0080] Resource expansion conditions are used to indicate the expansion of allocated hardware resources, i.e., adding hardware resources. Resource reduction conditions are used to indicate the reduction of allocated hardware resources, i.e., releasing some hardware resources.
[0081] In this step, if the multi-dimensional feature data or the multi-dimensional feature data of future time meets the resource expansion conditions, such as the multi-dimensional feature data being greater than or equal to the first threshold, then the hardware resources of the target business are fully allocated.
[0082] For example, if the multi-dimensional feature data is greater than or equal to the first threshold, then some hardware resources are obtained from the idle resource pool and allocated to the target business.
[0083] The idle resource pool is a component deployed in the business system to manage reserved hardware resources.
[0084] In some embodiments, if multi-dimensional feature data or multi-dimensional feature data for future time meet the resource expansion conditions, the amount of expanded hardware resources can be determined according to the target scenario, and then the hardware resources of that amount in the idle resource pool can be allocated to the target service.
[0085] Based on the target scenario, the amount of expanded hardware resources is determined, which can be based on a pre-set correspondence between scenarios and resource amounts. For example, the resource amount for operation support scenarios can be pre-set to 10%, and the resource amount for user scenarios to 20%; the resource amount for high-frequency user scenarios can be pre-set to 30%, and the resource amount for medium-frequency user scenarios to 20%. This approach enables elastic resource scheduling, and the allocation of hardware resources is more closely aligned with the actual scenario.
[0086] Optionally, expanding the hardware resources allocated to the target business includes: if there are multiple target businesses with multi-dimensional feature data or multi-dimensional feature data for future time that meet the resource expansion conditions, then determining the target expansion quantity corresponding to the multiple target businesses according to their priorities; and allocating the hardware resources of the target expansion quantity in the idle resource pool to the corresponding target businesses.
[0087] The priority of target business is used to indicate the importance of each business to the business system. For example, since operation support businesses are used to maintain the operation of the business system, the priority of operation support businesses can be set to be greater than that of user businesses, which in turn are greater than that of supply chain interaction businesses.
[0088] In this embodiment, if multiple target services have multi-dimensional feature data or future time-based multi-dimensional feature data that meet the resource expansion conditions, it indicates that hardware resources need to be expanded for multiple target services to avoid competition for hardware resources. This embodiment expands the hardware resources for multiple target services based on their priority.
[0089] Specifically, if there are multiple target business multi-dimensional feature data or future time multi-dimensional feature data that meet the resource expansion conditions, then according to the pre-set priorities of multiple target businesses, the target expansion quantity, i.e. resource quantity, of the target business with higher priority is determined first, and the hardware resources of the target expansion data quantity in the idle resource pool are allocated to the corresponding target business to realize the expansion of hardware resources of the target business with higher priority.
[0090] In some embodiments, the required expansion quantity can be determined based on each target service. If the total expansion quantity is greater than or equal to the amount of hardware resources in the idle resource pool, then expansion is prioritized for target services with higher priority, i.e., the target expansion quantity for target services with higher priority is determined in a limited manner.
[0091] In some embodiments, if the total expansion quantity is greater than or equal to the amount of hardware resources in the idle resource pool, the target expansion quantity for the target service can be determined proportionally based on priority.
[0092] In some embodiments, if the total expansion quantity is greater than or equal to the amount of hardware resources in the idle resource pool, a portion of the hardware resources can be obtained from the hardware resources corresponding to the services that meet the resource reduction conditions based on multi-dimensional feature data or multi-dimensional feature data of future time, and allocated to the target services.
[0093] When multiple target businesses have resource expansion needs at the same time, resource expansion can be carried out in an orderly manner for each target business according to priority. This can achieve precise matching and efficient allocation of resources, ensure that the resource needs of high-priority businesses are met first, and at the same time take into account the overall resource supply order of the business, thereby improving resource utilization efficiency and overall business operation stability.
[0094] Step S206: If the multi-dimensional feature data or the multi-dimensional feature data of future time meets the resource reduction conditions, then release at least a portion of the hardware resources allocated to the target business.
[0095] In this step, if the multi-dimensional feature data or the multi-dimensional feature data of future time meets the resource reduction conditions, such as the multi-dimensional feature data being less than the second threshold, then at least a portion of the hardware resources allocated to the target business will be released.
[0096] For example, if multi-dimensional feature data or multi-dimensional feature data for future time meets the resource reduction conditions, then the amount of hardware resources to be released is determined according to the target business or target scenario, and that amount of hardware resources allocated to the target business is released.
[0097] In some embodiments, the released hardware resources can be returned to the idle resource pool.
[0098] In this embodiment, while expanding hardware resources, idle hardware resources can also be released. This not only replenishes the resources required by the business in a timely manner to ensure efficient business operation, but also improves the overall utilization rate of hardware resources and avoids resource waste, thus realizing dynamic optimization of hardware resource configuration. Furthermore, by utilizing the powerful temporal sequence processing and reasoning capabilities of the Long Short-Term Memory Network model, the target scenarios obtained are more accurate and effective. At the same time, predictions of multi-dimensional feature data based on the changing trends of historical feature data are made, and the prediction results conform to actual laws, making the prediction results more accurate and effective.
[0099] In one possible implementation, the hardware resource allocation method provided in this application further includes managing hardware resources.
[0100] Specifically, the data platform of the business system deploys intelligent management modules and data governance modules.
[0101] The data platform statistically analyzes the hardware resources of business systems and records the hardware resources allocated to each business unit. These statistically analyzed hardware resources and the recorded allocated hardware resources are then encapsulated as an API service, namely the intelligent management module. Business systems or the data platform can directly access hardware resources and their allocated business units through this intelligent management module. This reduces the development workload of business systems, avoids repeatedly retrieving hardware resources and their assigned business units each time the hardware resource allocation method is executed, and improves data utilization.
[0102] The data governance module is used to identify data problems in the intelligent management module, such as checking whether the data recorded in the intelligent management module is updated in a timely manner or whether there are any errors, thus ensuring the accuracy of the data in the intelligent management module.
[0103] Figure 3 This is a schematic diagram of a hardware resource allocation device provided in an embodiment of this application. Figure 3 As shown, the hardware resource allocation device provided in this embodiment includes an acquisition module 301, a scene determination module 302, a prediction module 303, and a resource adjustment module 304.
[0104] The acquisition module 301 is used to acquire multi-dimensional feature data of the target business; the scenario determination module 302 is used to determine the target scenario to which the target business belongs based on the multi-dimensional feature data of the target business; the prediction module 303 is used to predict the multi-dimensional feature data of the target business in the future time based on the multi-dimensional feature data of the target business and the target scenario; the resource adjustment module 304 is used to adjust the hardware resources allocated to the target business if the multi-dimensional feature data or the multi-dimensional feature data in the future time meets the preset adjustment conditions corresponding to the target scenario.
[0105] Optionally, the preset adjustment conditions include resource expansion conditions and resource reduction conditions; the resource adjustment module 304 is specifically used for:
[0106] If the multi-dimensional feature data or the multi-dimensional feature data of future time meets the resource expansion conditions, then the hardware resources allocated to the target business will be expanded; if the multi-dimensional feature data or the multi-dimensional feature data of future time meets the resource reduction conditions, then at least a portion of the hardware resources allocated to the target business will be released.
[0107] Optional, resource adjustment module 304, specifically used for:
[0108] If there are multi-dimensional feature data or future multi-dimensional feature data for multiple target services that meet the resource expansion conditions, then the target expansion quantity corresponding to the multiple target services is determined according to the priority of the multiple target services; the hardware resources of the target expansion quantity in the idle resource pool are allocated to the corresponding target services; if the multi-dimensional feature data or future multi-dimensional feature data meet the resource reduction conditions, then at least part of the hardware resources allocated to the target services are released.
[0109] Optionally, the scene determination module 302 is specifically used for:
[0110] The multi-dimensional feature data of the target business is input into a scene classification model based on a long short-term memory network model to obtain the target scene to which the target business belongs. The scene classification model is used to input the multi-dimensional feature data into the long short-term memory network layer, extract the feature vectors of the multi-dimensional feature data, and input the feature vectors into the fully connected layer to map the feature vectors to the classification space.
[0111] Optional, multi-dimensional feature data includes hardware resource usage data and user access data.
[0112] Optionally, the prediction module 303 is specifically used for:
[0113] Obtain the historical feature data change trends corresponding to the target scenario; based on the multi-dimensional feature data and the historical feature data change trends, predict the multi-dimensional feature data of the target business in the future.
[0114] The hardware resource allocation device provided in this application embodiment can be used to execute the technical solution of the hardware resource allocation method provided in any of the above embodiments of this application. Its implementation principle and technical effect are similar, and will not be repeated here.
[0115] Figure 4This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. As shown in the figure, the electronic device of this embodiment may include: at least one processor 401; and a memory 402 communicatively connected to at least one processor 401; wherein, the memory 402 stores instructions executable by at least one processor 401, and the instructions are executed by at least one processor 401 to cause the electronic device to perform the method as described in any of the above embodiments.
[0116] Optionally, the memory 402 can be either standalone or integrated with the processor 401. When the memory 402 is set up independently, the device also includes a bus for connecting the memory 402 and the processor 401.
[0117] The implementation principle and technical effects of the electronic device provided in this embodiment can be found in the foregoing embodiments, and will not be repeated here.
[0118] This application also provides a computer-readable storage medium storing computer-executable instructions. When the computer-executable instructions are executed by a processor, the methods provided in any of the foregoing embodiments can be implemented.
[0119] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the method provided in any of the foregoing embodiments.
[0120] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily essential to this application.
[0121] It should be further noted that although the steps in the flowchart are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowchart may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.
[0122] It should be understood that the above-described device embodiments are merely illustrative, and the device of this application can also be implemented in other ways. For example, the division of units / modules in the above embodiments is only a logical functional division, and there may be other division methods in actual implementation. For example, multiple units, modules, or components may be combined, or integrated into another system, or some features may be ignored or not executed.
[0123] Furthermore, unless otherwise specified, the functional units / modules in the various embodiments of this application can be integrated into one unit / module, or each unit / module can exist physically separately, or two or more units / modules can be integrated together. The integrated units / modules described above can be implemented in hardware or as software program modules.
[0124] When integrated units / modules are implemented in hardware, the hardware can be digital circuits, analog circuits, etc. The physical implementation of the hardware structure includes, but is not limited to, transistors, memristors, etc. Unless otherwise specified, the processor can be any suitable hardware processor, such as a CPU, GPU, FPGA, DSP, and ASIC, etc. Unless otherwise specified, the storage unit can be any suitable magnetic or magneto-optical storage medium, such as Resistive Random Access Memory (RRAM), Dynamic Random Access Memory (DRAM), Static Random Access Memory (SRAM), Enhanced Dynamic Random Access Memory (EDRAM), High-Bandwidth Memory (HBM), Hybrid Memory Cube (HMC), etc.
[0125] If the integrated unit / module is implemented as a software program module and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.
[0126] In the above embodiments, the descriptions of each embodiment have their own emphasis. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments. The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification.
[0127] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the following claims.
[0128] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.
Claims
1. A hardware resource allocation method, characterized in that, include: Acquire multi-dimensional feature data of the target business; Based on the multi-dimensional feature data of the target business, the target scenario to which the target business belongs is determined; Based on the multi-dimensional feature data of the target business and the target scenario, predict the multi-dimensional feature data of the target business in the future time. If the multi-dimensional feature data or the multi-dimensional feature data of future time meets the preset adjustment conditions corresponding to the target scenario, then the hardware resources allocated to the target service will be adjusted.
2. The method according to claim 1, characterized in that, The preset adjustment conditions include resource expansion conditions and resource reduction conditions; if the multi-dimensional feature data or the multi-dimensional feature data of future time meets the preset adjustment conditions corresponding to the target scenario, then the hardware resources allocated to the target service are adjusted, including: If the multi-dimensional feature data or the multi-dimensional feature data of future time meets the resource expansion conditions, then the expanded resources are fully allocated to the hardware resources of the target service. If the multi-dimensional feature data or the multi-dimensional feature data of future time satisfies the resource reduction condition, then at least a portion of the hardware resources allocated to the target service will be released.
3. The method according to claim 2, characterized in that, If the multi-dimensional feature data or the multi-dimensional feature data of future time satisfies the resource expansion conditions, then the hardware resources allocated to the target service are expanded, including: If there are multiple target services with multi-dimensional feature data or the multi-dimensional feature data of the future time that satisfy the resource expansion conditions, then the target expansion quantity corresponding to the multiple target services is determined according to the priority of the multiple target services. Allocate the target number of hardware resources in the idle resource pool to the corresponding target services.
4. The method according to claim 1, characterized in that, The determination of the target scenario to which the target service belongs based on the multi-dimensional feature data of the target service includes: The multi-dimensional feature data of the target service is input into a scene classification model based on a long short-term memory network model to obtain the target scene to which the target service belongs; The scene classification model is used to input the multi-dimensional feature data into a long short-term memory network layer, extract the feature vector of the multi-dimensional feature data, and input the feature vector into a fully connected layer to map the feature vector to the classification space.
5. The method according to any one of claims 1-4, characterized in that, The multi-dimensional feature data includes hardware resource usage data and user access data.
6. The method according to any one of claims 1-4, characterized in that, The multi-dimensional feature data for predicting the future time of the target business based on the multi-dimensional feature data of the target business and the target scenario includes: Obtain the historical feature data change trend corresponding to the target scene; Based on the multi-dimensional feature data and the changing trends of the historical feature data, predict the multi-dimensional feature data of the target business in the future.
7. A hardware resource allocation device, characterized in that, include: The acquisition module is used to acquire multi-dimensional feature data of the target business. The scenario determination module is used to determine the target scenario to which the target service belongs based on the multi-dimensional feature data of the target service; The prediction module is used to predict the multi-dimensional feature data of the target business in the future time based on the multi-dimensional feature data of the target business and the target scenario; The resource adjustment module is used to adjust the hardware resources allocated to the target service if the multi-dimensional feature data or the multi-dimensional feature data of the future time meets the preset adjustment conditions corresponding to the target scenario.
8. An electronic device, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method as described in any one of claims 1 to 6.
9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1 to 6.
10. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method of any one of claims 1 to 6.