Resource management method and device of data center, electronic equipment and storage medium

Abstract

The invention discloses a resource management method and device of a data center, electronic equipment and a storage medium, and relates to the technical field of financial science and technology and operation and maintenance, and the method comprises the steps: obtaining a business log of a target business system in a business center through a log center, and obtaining a matched business index according to the business log; aggregating the service indexes according to the current time granularity to obtain service evaluation indexes, and comparing the service evaluation indexes with each rule in a rule set; and if it is determined that the business evaluation index hits a target rule in the rule set, adjusting resources occupied by a target business system in the data center according to the target rule. According to the technical scheme provided by the embodiment of the invention, the dynamic allocation of the resources of the data center is realized, the rationalization and saving of the resources required by the service system are ensured, meanwhile, the phenomena of misestimation and missed estimation existing in a manual estimation mode are avoided, and the labor cost and time cost consumed by resource management are reduced.

Description

Technical Field

[0001] This invention relates to the fields of financial technology and operation and maintenance technology, and in particular to a data center resource management method, apparatus, electronic device and storage medium. Background Technology

[0002] With the development of cloud computing and container technology, more and more financial institutions are choosing to build data centers and centralize the operation and deployment of their business systems in the data center. Since the data center manages a large number of servers of various types, how to effectively manage server resources has become a key focus for every financial institution.

[0003] Because business systems experience peak and trough periods in their workload, the performance requirements for system configuration vary significantly at different times. Configuring hardware resources based on peak workloads would result in substantial waste during off-peak periods, while configuring them based on off-peak workloads would lead to insufficient performance. Therefore, current technologies typically rely on manual estimation and container technology to scale up the system during peak periods to improve performance. After peak periods, excess containers are manually decommissioned, maintaining a higher system hardware configuration than during off-peak periods.

[0004] However, manual estimation not only relies on the experience of maintenance personnel, requiring a large amount of manpower and time, but also suffers from miscalculations and omissions. The results of manual estimation have a large error and cannot guarantee the rational utilization of data center server resources. Summary of the Invention

[0005] This invention provides a resource management method, apparatus, electronic device, and storage medium for data centers to solve the problem of unreasonable server resource allocation in data centers.

[0006] According to another aspect of the present invention, a resource management method for a data center is provided, comprising:

[0007] Obtain business logs from the target business system in the business center through the log center, and obtain matching business metrics based on the business logs;

[0008] The business metrics are aggregated according to the current time granularity to obtain business evaluation metrics, and the business evaluation metrics are compared with each rule in the rule set.

[0009] If it is determined that the business evaluation indicator matches the target rule in the rule set, the resources occupied by the target business system in the data center are adjusted according to the target rule.

[0010] The step of aggregating the business metrics according to the current time granularity to obtain business evaluation metrics includes: aggregating the business metrics according to the current time granularity to obtain average business data and fluctuation parameters, and using the average business data and fluctuation parameters as business evaluation metrics.

[0011] After comparing the business evaluation metric with each rule in the rule set, the method further includes: if it is determined that the business evaluation metric does not match any rule in the rule set, obtaining a predicted business curve based on the business metric using a time series model; obtaining a predicted business metric based on the predicted business curve, and comparing the predicted business metric with each rule in the rule set; if it is determined that the predicted business metric matches a first rule in the rule set, adjusting the resources occupied by the target business system in the data center according to the first rule.

[0012] After comparing the predicted business metrics with each rule in the rule set, the method further includes: if it is determined that the predicted business metrics do not match any rule in the rule set, inputting the business evaluation metrics and the predicted business curve into the rule prediction model to obtain the resource adjustment rules of the target business system through the rule prediction model.

[0013] The step of aggregating the business metrics according to the current time granularity to obtain business evaluation metrics specifically includes: obtaining the resource load rate of the target business system and the energy consumption metrics of the server where the resources are located, and aggregating the business metrics, the resource load rate and the energy consumption metrics according to the time granularity to obtain business evaluation metrics.

[0014] After comparing the business evaluation index with each rule in the rule set, the method further includes: if it is determined that the business evaluation index does not match any rule in the rule set, updating the existing rules of the target business system through reinforcement learning, or displaying a related domain rule template to the user to guide the user to configure new rules for the target business system.

[0015] According to another aspect of the present invention, a resource management device for a data center is provided, comprising:

[0016] The business metrics acquisition module is used to obtain business logs of the target business system in the business center through the log center, and to obtain matching business metrics based on the business logs;

[0017] The rule comparison and execution module is used to aggregate the business metrics according to the current time granularity to obtain business evaluation metrics, and compare the business evaluation metrics with each rule in the rule set;

[0018] The target rule acquisition module is used to adjust the resources occupied by the target business system in the data center according to the target rule if it is determined that the business evaluation index matches the target rule in the rule set.

[0019] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the data center resource management method according to any embodiment of the present invention.

[0020] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the data center resource management method described in any embodiment of the present invention.

[0021] According to another aspect of the present invention, a computer program product is provided, comprising a computer program that, when executed by a processor, implements the data center resource management method described in any embodiment of the present invention.

[0022] The technical solution of this invention obtains business logs of the target business system in the business center through a log center, and obtains matching business indicators based on the business logs; aggregates the business indicators according to the current time granularity to obtain business evaluation indicators, and compares the business evaluation indicators with each rule in the rule set; if it is determined that the business evaluation indicator matches the target rule in the rule set, the resources occupied by the target business system in the data center are adjusted according to the target rule. This achieves dynamic allocation of data center resources, ensuring the rationalization and conservation of resources required by the business system, while avoiding the miscalculations and omissions that exist in manual estimation methods, and reducing the manpower and time costs of resource management.

[0023] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a flowchart of a data center resource management method according to Embodiment 1 of the present invention;

[0026] Figure 2 This is a schematic diagram illustrating an application scenario of a data center resource management method according to Embodiment 1 of the present invention;

[0027] Figure 3 This is a flowchart of another data center resource management method provided according to Embodiment 2 of the present invention;

[0028] Figure 4 This is a flowchart of another data center resource management method provided according to Embodiment 3 of the present invention;

[0029] Figure 5 This is a schematic diagram of the structure of a data center resource management device according to Embodiment 4 of the present invention;

[0030] Figure 6 This is a schematic diagram of the structure of an electronic device that implements the data center resource management method of this invention. Detailed Implementation

[0031] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0032] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0033] Example 1

[0034] Figure 1This is a flowchart of a data center resource management method provided in Embodiment 1 of the present invention. This embodiment is applicable to adjusting the data center resources occupied by the business system based on business indicators and rule sets in the business logs. This method can be executed by a data center resource management device, which can be implemented in hardware and / or software. The data center resource management device can be configured in a resource management system, which is configured in an electronic device. Figure 1 As shown, the method includes:

[0035] S101. Obtain the business logs of the target business system in the business center through the log center, and obtain the matching business indicators based on the business logs.

[0036] like Figure 2 As shown, a data center refers to a server room managed using containerization technology. In this embodiment of the invention, the resource management system can abstract the hardware resources of each server in the containerized data center, such as CPU (Central Processing Unit), GPU (Graphics Processing Unit), memory, and network bandwidth, into resources in a resource pool, and monitor the load of each resource (e.g., utilization, throughput, latency, etc.) to form a visual dashboard. Based on this, the system utilizes the lightweight characteristics of containers to achieve on-demand allocation of resources and reclamation of idle resources through API (Application Programming Interface).

[0037] For example, server 1 has 10 CPUs. These 10 CPUs can be distributed equally among 10 business systems, or they can be distributed among 5 business systems, with each business system occupying 2 CPUs. Other physical resources (such as network bandwidth, memory, disk, etc.) can be allocated in the same way. At the same time, the allocated physical resources can be reclaimed and redistributed. For example, business system 2 occupies 2 CPUs, 200MB (megabyte) of memory, and 1GB (gigabyte) of disk space on server 1. When business system 2 is no longer running, the physical resources it occupies can be redistributed to other business systems.

[0038] A business center refers to a business platform composed of multiple business systems, which facilitates the realization of closed-loop business processes and data interoperability across systems through integration and collaboration. Different business systems carry different business functions. After a business request enters the business center, it enters the matching business system for processing. In this embodiment of the invention, the resource management system can package each business system into different container images. The container image contains the complete data of the business system, such as code, operating system dependency libraries, and configuration files, thereby forming a single lightweight executable file.

[0039] Container images enable business systems to run consistently on any server node, avoiding performance fluctuations caused by environmental differences. During initial deployment, the number of container instances can be configured based on baseline load (e.g., 2 containers, each containing 5 CPU cores and 500MB of memory), and cross-server node scheduling can be achieved through existing container orchestration platforms (e.g., business system A can be distributed across server 1 and server 2), thereby achieving elastic deployment of multiple instances.

[0040] The log center is a platform for centralized management and storage of log data generated by business systems, recording all log data generated by these systems. In this invention, the resource management system obtains business logs from various business systems in the business center through the log center, and acquires business metrics matching the current business system based on these logs. Since different business systems handle different types of business, the business metrics reflecting their workload also differ; therefore, different business metrics can be calculated and acquired separately for each business system.

[0041] Taking the target business system to be monitored as a transaction system as an example, since the number of transaction requests reflects the busyness of the business system, the transaction concurrency per second (TPS) can be used as the business indicator of the transaction system. The transaction concurrency per second can be calculated by obtaining transaction data (i.e. business data) from the business logs, or the transaction concurrency per second can be extracted directly from the business logs; where transaction concurrency per second is the number of transaction requests that the business system processes simultaneously per second.

[0042] In addition, the query concurrency per second (QPS) can also be used as a business indicator for the transaction system, which is the number of query requests processed simultaneously by the business system per second. In particular, multiple types of data can be used as business indicators, for example, both the transaction concurrency per second and the query concurrency per second mentioned above can be used as business indicators. Optionally, in the embodiments of the present invention, the number and type of business indicators of the business system are not specifically limited.

[0043] S102. Aggregate the business metrics according to the current time granularity to obtain business evaluation metrics, and compare the business evaluation metrics with each rule in the rule set.

[0044] Since instantaneous data from business systems (such as the aforementioned transaction concurrency per second) is often random and sudden, business metrics at a single moment cannot truly reflect the actual business status of the system. Therefore, it is necessary to calculate business metrics over a period of time based on time granularity, and use these metrics as business evaluation indicators. For example, by summing up the transaction concurrency per second over the past day and then dividing the sum by the total number of seconds (i.e., 3600 seconds × 24 hours), the average transaction concurrency per second for the day can be obtained.

[0045] Specifically, the current time granularity refers to the time granularity matched with the business system. Since different business systems carry different types of business, the frequency of adjustment of the data center resources they occupy may also be different. Therefore, different business systems may correspond to different time granularities. For example, for business system A, it is necessary to determine whether its data center resources need to be adjusted on a per-minute basis; for business system B, it is necessary to determine whether its data center resources need to be adjusted on an hourly basis.

[0046] The rule set predefines various expansion or contraction rules. Each expansion or contraction rule is formulated for one or more business evaluation indicators. By setting trigger conditions (e.g., upper and lower limit trigger thresholds) and execution actions (e.g., expansion or contraction multiples), specific expansion or contraction rules are defined. For example, rule A triggers expansion when the average daily transaction concurrency is greater than or equal to 50, and the specific execution actions (i.e., the types and quantities of resources to be added) have also been pre-configured. Rule B triggers contraction when the average daily transaction concurrency is less than 30, and the specific execution actions (i.e., the types and quantities of resources to be reduced) have also been pre-configured.

[0047] The business evaluation metrics are compared with the rules in the rule set. For example, if the business evaluation metric is an average daily transaction concurrency of 60, it obviously matches rule A; if the business evaluation metric is an average daily transaction concurrency of 20, it obviously matches rule B; if the business evaluation metric is an average daily transaction concurrency of 40, it does not match any rule. In particular, as described in the above technical solution, the number of business metrics in the business system can be multiple, and the number of corresponding business evaluation metrics can also be multiple. Each rule in the rule set can also be constructed based on one or more business evaluation metrics.

[0048] Optionally, in this embodiment of the invention, the step of aggregating the business indicators according to the current time granularity to obtain business evaluation indicators includes: aggregating the business indicators according to the current time granularity to obtain average business data and fluctuation parameters, and using the average business data and the fluctuation parameters as business evaluation indicators.

[0049] Specifically, fluctuation parameters can include various types such as variance, standard deviation, range, mean absolute deviation, coefficient of variation, interquartile range, and mean squared error. After calculating and obtaining average business data based on business indicators within the current time granularity, fluctuation parameters are calculated based on the average business data. The average business data and fluctuation parameters are then used together as business evaluation indicators. For example, after calculating and obtaining the average number of transactions per second, the overall variance is calculated based on the number of transactions per second and the average number of transactions per second. The average number of transactions per second and the variance are then used together as business evaluation indicators.

[0050] When comparing business evaluation metrics with rules in the rule set, the average transaction concurrency per second (DPS) and variance are compared with the average DPS threshold and variance threshold of a specific rule. Only if both are met is the business evaluation metric considered to match that rule in the rule set. For example, if a business evaluation metric includes an average DPS of 60 and a variance of 8, while rule B requires an average DPS greater than or equal to 50 and a variance less than 10 to trigger scaling up, then the current business evaluation metric matches rule B. This fluctuation parameter reflects the dispersion of quantitative data from the average level, ensuring the stability of the business indicator data distribution and further improving the effectiveness of scaling up / down rules.

[0051] S103. If it is determined that the business evaluation index matches the target rule in the rule set, the resources occupied by the target business system in the data center are adjusted according to the target rule.

[0052] As described in the above technical solution, if the business evaluation indicator matches a target rule in the rule set, the resources occupied by the target business system in the data center are expanded, reduced, or the server hosting the occupied resources is changed according to the target rule. If the business evaluation indicator does not match any rule in the rule set, the resources occupied by the target business system in the data center are not adjusted, and the existing resource configuration is maintained. Furthermore, historical data can be used to create charts to more intuitively display the operating status of the target business system and provide early warnings for the time points when rules are triggered.

[0053] Optionally, in this embodiment of the invention, the step of aggregating the business metrics according to the current time granularity to obtain business evaluation metrics specifically includes: obtaining the resource load rate of the target business system and the energy consumption metrics of the server where the resources are located, and aggregating the business metrics, the resource load rate and the energy consumption metrics according to the time granularity to obtain business evaluation metrics.

[0054] Specifically, resource load rate refers to the actual utilization rate of resources occupied by the target business system. For example, the current CPU utilization rate is 90%. Although business metrics may reflect that the business system is carrying a large amount of business, there is only a need for expansion when the resource load rate is also high; otherwise, resources will remain idle. Energy consumption metrics refer to the energy consumption values ​​of servers, such as Power Usage Effectiveness (PUE). When it is necessary to reduce the resources occupied by the target business system, the target business system should be prioritized to be scheduled to low-energy-consumption server nodes, and high-load servers should be prioritized to be shut down to reduce the overall energy consumption of the data center. The business evaluation metrics obtained in this way include business metrics, resource load rate, and energy consumption metrics. The rules in the rule set can also be configured with trigger conditions and execution actions from the perspective of business metrics, resource load rate, and energy consumption metrics. This avoids misjudgment based on a single evaluation metric and reduces the overall energy consumption of the data center.

[0055] The technical solution of this invention obtains business logs of the target business system in the business center through a log center, and obtains matching business indicators based on the business logs; aggregates the business indicators according to the current time granularity to obtain business evaluation indicators, and compares the business evaluation indicators with each rule in the rule set; if it is determined that the business evaluation indicator matches the target rule in the rule set, the resources occupied by the target business system in the data center are adjusted according to the target rule. This achieves dynamic allocation of data center resources, ensuring the rationalization and conservation of resources required by the business system, while avoiding the miscalculations and omissions that exist in manual estimation methods, and reducing the manpower and time costs of resource management.

[0056] It should be noted that all information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for display, data used for analysis, etc.) involved in this disclosure are information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data comply with the relevant laws, regulations and standards of the relevant regions.

[0057] Example 2

[0058] Figure 3This is a flowchart of a data center resource management method provided in Embodiment 2 of the present invention. The relationship between this embodiment and the above embodiments is that the resources occupied by the business system are adjusted based on the predicted business curve of the target business system, such as... Figure 3 As shown, the method specifically includes:

[0059] S201. Obtain the business logs of the target business system in the business center through the log center, and obtain the matching business indicators based on the business logs.

[0060] S202. Aggregate the business metrics according to the current time granularity to obtain business evaluation metrics, and compare the business evaluation metrics with each rule in the rule set.

[0061] S203. If it is determined that the business evaluation index matches the target rule in the rule set, the resources occupied by the target business system in the data center are adjusted according to the target rule.

[0062] S204. If it is determined that the business evaluation indicator does not match any rule in the rule set, obtain the predicted business curve based on the business indicator through a time series model.

[0063] A time series model is a mathematical model used to analyze and predict data arranged in chronological order. The input time series data has characteristics such as noise, seasonality, and trend. Among them, trend indicates that the data shows a long-term upward or downward trend; noise indicates random fluctuations in the data under the interference of external factors; and seasonality indicates that the data has periodic fluctuations.

[0064] After inputting the business metrics within the current time granularity into the time series model, the time series model first plots the data change curve for that time period. Then, based on this data change curve, the time series model predicts the data change curve for subsequent time periods, i.e., obtains the predicted business curve for subsequent time periods. The subsequent time periods can be downstream time periods adjacent to the current time granularity or downstream time periods not adjacent to the current time granularity.

[0065] S205. Obtain the predicted business indicators based on the predicted business curve, and compare the predicted business indicators with each rule in the rule set.

[0066] After obtaining the predicted business indicators for subsequent time periods based on the predicted business curve, the predicted business indicators are compared with each rule in the rule set.

[0067] S206. If it is determined that the predicted business indicator matches the first rule in the rule set, then the resources occupied by the target business system in the data center are adjusted according to the first rule.

[0068] The purpose of resource adjustment is to ensure that resources meet the business processing needs of the system in future business operations while avoiding resource redundancy. Compared to using business evaluation indicators at the current time granularity (i.e., historical time period) as the basis for rule acquisition, the predicted business indicators obtained through predicted business curves reflect the future business volume trend of the system, more accurately reflecting the potential workload of the system, and further ensuring the accuracy of resource adjustment results. Furthermore, if it is determined that the predicted business indicators do not match any rule in the rule set, the current resource utilization of the target business system can be maintained.

[0069] Optionally, in this embodiment of the invention, after comparing the predicted business indicator with each rule in the rule set, the method further includes: if it is determined that the predicted business indicator does not match any rule in the rule set, inputting the business evaluation indicator and the predicted business curve into the rule prediction model to obtain the resource adjustment rules of the target business system through the rule prediction model.

[0070] Specifically, due to the limitations of the resource adjustment rule construction process, the rule set may not cover all reasonable resource adjustment rules. If the predicted business indicators do not match any rule in the rule set, it does not mean that the resources occupied by the target business system do not need to be adjusted. It may also be that there are missing rules in the rule set. In this case, the above-mentioned business evaluation indicators and predicted business curves can be input into the rule prediction model.

[0071] Rule prediction models are based on large language models (LLM) in deep learning. Essentially, they are a type of processing result prediction model. They abstract the effect of resource adjustment rules (i.e., processing effect) into a dynamic network propagation problem. By evaluating node vulnerability and edge propagation, they calculate the effect of each possible resource adjustment rule after execution and give the corresponding adjustment rule (i.e., predict the adjustment rule) based on the effect.

[0072] For example, the rule prediction model can provide resource adjustment methods for the target business system under high load based on the business relationship between transaction concurrency per second and query concurrency per second. It can also provide the predicted adjustment effect and use the predicted adjustment method with the best effect as the resource adjustment rule for the target business system. This enables the automatic generation of resource adjustment rules. When there is no corresponding resource adjustment rule in the rule set, the rule prediction model can generate a resource adjustment rule with a better effect.

[0073] The technical solution of this invention involves determining that a business evaluation indicator does not match any rule in the rule set, obtaining a predicted business curve based on the business indicator using a time series model, then obtaining a predicted business indicator based on the predicted business curve, and comparing the predicted business indicator with each rule in the rule set. This ensures that when the predicted business indicator does not match any rule in the rule set, the resources occupied by the target business system are not adjusted. Therefore, the predicted business indicator obtained through the predicted business curve reflects the future business volume trend of the business system, more accurately reflects the possible load level of the business system, and further ensures the accuracy of resource adjustment results.

[0074] Example 3

[0075] Figure 4 This is a flowchart of a data center resource management method provided in Embodiment 3 of the present invention. The relationship between this embodiment and the above embodiments is that when the business evaluation indicator does not match any rule in the rule set, the existing rules are updated or the user is guided to configure new rules, such as... Figure 4 As shown, the method specifically includes:

[0076] S301. Obtain the business logs of the target business system in the business center through the log center, and obtain the matching business indicators based on the business logs.

[0077] S302. Aggregate the business metrics according to the current time granularity to obtain business evaluation metrics, and compare the business evaluation metrics with each rule in the rule set.

[0078] S303. If it is determined that the business evaluation index matches the target rule in the rule set, the resources occupied by the target business system in the data center are adjusted according to the target rule.

[0079] S304. If it is determined that the business evaluation indicator does not match any rule in the rule set, the existing rules of the target business system are updated through reinforcement learning, or the relevant domain rule template is displayed to the user to guide the user to configure new rules for the target business system.

[0080] Related domains refer to other domains that have business similarities with the current business center. For example, for the business system of a financial institution, the e-commerce business in the e-commerce domain has a great similarity to its financial business. In this case, the existing resource adjustment rule templates in the e-commerce domain can be displayed to the user. Each related domain rule template includes structured fields such as trigger conditions (e.g., upper and lower limit trigger thresholds), execution actions (expansion multiple), and constraint parameters (e.g., available resource limit). By providing users with a visual configuration interface, users are guided to modify the various parameters in the related domain rule template to complete the configuration and generation of new rules.

[0081] Furthermore, if it is determined that the business evaluation metric does not match any rule in the rule set, it may be because the existing rules in the rule set are no longer suitable for the business needs of the current business system due to changes in the business type. In this case, the existing rules can be modified through reinforcement learning (RL) based on the triggering conditions, execution actions, and historical execution effects of the existing rules in the target business system. For example, the expansion threshold for the number of transactions per second can be increased from 50 to 60 to ensure the iterative update of each rule in the rule set.

[0082] The technical solution of this invention, when determining that any rule in the rule set is not matched by a business evaluation indicator, updates the existing rules of the target business system through reinforcement learning, or displays a relevant domain rule template to the user to guide the user in configuring new rules for the target business system. Thus, by modifying existing rules through reinforcement learning, it ensures that each rule in the rule set adapts to the iterative updates of the business system. Simultaneously, by guiding the user to modify various parameters in the relevant domain rule template, it ensures the timely generation of new rules and the timely adjustment of resource usage by the business system.

[0083] Example 4

[0084] Figure 5 This is a structural block diagram of a data center resource management device provided in Embodiment 4 of the present invention. The device specifically includes:

[0085] The business metric acquisition module 401 is used to acquire business logs of the target business system in the business center through the log center, and to acquire matching business metrics based on the business logs.

[0086] The rule comparison execution module 402 is used to aggregate the business indicators according to the current time granularity to obtain business evaluation indicators, and compare the business evaluation indicators with each rule in the rule set;

[0087] The target rule acquisition module 403 is used to adjust the resources occupied by the target business system in the data center according to the target rule if it is determined that the business evaluation index matches the target rule in the rule set.

[0088] The technical solution of this invention obtains business logs of the target business system in the business center through a log center, and obtains matching business indicators based on the business logs; aggregates the business indicators according to the current time granularity to obtain business evaluation indicators, and compares the business evaluation indicators with each rule in the rule set; if it is determined that the business evaluation indicator matches the target rule in the rule set, the resources occupied by the target business system in the data center are adjusted according to the target rule. This achieves dynamic allocation of data center resources, ensuring the rationalization and conservation of resources required by the business system, while avoiding the miscalculations and omissions that exist in manual estimation methods, and reducing the manpower and time costs of resource management.

[0089] Optionally, the rule comparison execution module 402 is specifically used to aggregate the business indicators according to the current time granularity to obtain average business data and fluctuation parameters, and use the average business data and fluctuation parameters as business evaluation indicators.

[0090] Optionally, the data center resource management device is further configured to, if it is determined that the business evaluation indicator does not match any rule in the rule set, obtain a predicted business curve based on the business indicator through a time series model; obtain a predicted business indicator based on the predicted business curve, and compare the predicted business indicator with each rule in the rule set; if it is determined that the predicted business indicator matches a first rule in the rule set, adjust the resources occupied by the target business system in the data center according to the first rule.

[0091] Optionally, the data center resource management device is further configured to, if it is determined that the predicted business indicator does not match any rule in the rule set, input the business evaluation indicator and the predicted business curve into the rule prediction model, so as to obtain the resource adjustment rules of the target business system through the rule prediction model.

[0092] Optionally, the rule comparison execution module 402 is further configured to obtain the resource load rate of the target business system and the energy consumption index of the server where the resource is located, and aggregate the business index, the resource load rate and the energy consumption index according to the time granularity to obtain the business evaluation index.

[0093] Optionally, the data center resource management device is also used to update the existing rules of the target business system through reinforcement learning if it is determined that the business evaluation index does not match any rule in the rule set, or to display the relevant domain rule template to the user to guide the user to configure new rules for the target business system.

[0094] The above-described apparatus can execute the data center resource management method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects for executing the method. Technical details not described in detail in this embodiment can be found in the data center resource management method provided in any embodiment of the present invention.

[0095] Example 5

[0096] Figure 6 A schematic diagram of an electronic device 10, which can be used to implement embodiments of the present invention, is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, electronic devices, blade electronic devices, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0097] like Figure 6 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0098] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0099] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as resource management methods in a data center.

[0100] In some embodiments, the data center resource management method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as a storage unit. In some embodiments, part or all of the computer program may be loaded and / or installed on a heterogeneous hardware accelerator via ROM and / or a communication unit. When the computer program is loaded into RAM and executed by a processor, one or more steps of the data center resource management method described above may be performed. Alternatively, in other embodiments, the processor may be configured to perform the data center resource management method by any other suitable means (e.g., by means of firmware).

[0101] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0102] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0103] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0104] To provide user interaction, the systems and techniques described herein can be implemented on a heterogeneous hardware accelerator, which includes: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the heterogeneous hardware accelerator. Other types of devices can also be used to provide user interaction; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or haptic feedback); and input from the user can be received in any form (including sound input, voice input, or haptic input).

[0105] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0106] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0107] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0108] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A resource management method for a data center, characterized in that, include: Obtain business logs from the target business system in the business center through the log center, and obtain matching business metrics based on the business logs; The business metrics are aggregated according to the current time granularity to obtain business evaluation metrics, and the business evaluation metrics are compared with each rule in the rule set. If it is determined that the business evaluation indicator matches the target rule in the rule set, the resources occupied by the target business system in the data center are adjusted according to the target rule.

2. The data center resource management method according to claim 1, characterized in that, The process of aggregating the business metrics according to the current time granularity to obtain business evaluation metrics includes: The business metrics are aggregated according to the current time granularity to obtain average business data and fluctuation parameters, and the average business data and fluctuation parameters are used as business evaluation metrics.

3. The data center resource management method according to claim 1, characterized in that, After comparing the business evaluation metrics with the rules in the rule set, the process also includes: If it is determined that the business evaluation indicator does not match any rule in the rule set, a predicted business curve is obtained based on the business indicator through a time series model. The predicted business indicators are obtained based on the predicted business curve, and the predicted business indicators are compared with each rule in the rule set. If it is determined that the predicted business metric matches the first rule in the rule set, then the resources occupied by the target business system in the data center are adjusted according to the first rule.

4. The data center resource management method according to claim 3, characterized in that, After comparing the predicted business metrics with the various rules in the rule set, the process also includes: If it is determined that the predicted business indicator does not match any rule in the rule set, the business evaluation indicator and the predicted business curve are input into the rule prediction model to obtain the resource adjustment rules of the target business system through the rule prediction model.

5. The data center resource management method according to claim 1, characterized in that, The aggregation of the business metrics based on the current time granularity to obtain business evaluation metrics specifically includes: Obtain the resource load rate of the target business system and the energy consumption index of the server where the resource is located, and aggregate the business index, the resource load rate and the energy consumption index according to the time granularity to obtain the business evaluation index.

6. The data center resource management method according to claim 1, characterized in that, After comparing the business evaluation metrics with the rules in the rule set, the process also includes: If it is determined that the business evaluation indicator does not match any rule in the rule set, the existing rules of the target business system are updated through reinforcement learning, or a related domain rule template is displayed to the user to guide the user in configuring new rules for the target business system.

7. A resource management device for a data center, characterized in that, include: The business metrics acquisition module is used to obtain business logs of the target business system in the business center through the log center, and to obtain matching business metrics based on the business logs; The rule comparison and execution module is used to aggregate the business metrics according to the current time granularity to obtain business evaluation metrics, and compare the business evaluation metrics with each rule in the rule set; The target rule acquisition module is used to adjust the resources occupied by the target business system in the data center according to the target rule if it is determined that the business evaluation index matches the target rule in the rule set.

8. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the data center resource management method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed by a processor, implement the data center resource management method according to any one of claims 1-6.

10. A computer program product comprising a computer program that, when executed by a processor, implements the data center resource management method of any one of claims 1-6.

Images