Adaptive tuning method and system for cloud distributed cache database

Causal AI technology analyzes causal relationships and automatically adjusts cache database configuration items, solving the problem of optimizing cache database business indicators in cross-multi-subject cloud environments, and improving tuning efficiency and quality.

CN119066095AActive Publication Date: 2024-12-03BEIJING VOLCANO ENGINE TECH CO LTD

Patent Information

Application Number
CN202411047939.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-31
Publication Date
2024-12-03
Estimated Expiration
2044-07-31

AI Technical Summary

Technical Problem

In cache databases distributedly deployed across multi-subject cloud environments, it is difficult for existing technology to quickly optimize business metrics such as Apdex, P99 Latency, QPS and Availability, resulting in inefficient system tuning.

Method used

Causal AI technology is used to analyze the causal relationship between configuration items and business indicators in the distributed cache database through the network causal model, generate an adaptive tuning strategy, and automatically adjust the configuration parameters to optimize business indicators.

Benefits of technology

It improves the optimization efficiency and quality of business indicators, reduces the time and cost of manual intervention, and realizes adaptive tuning of cached databases in the cloud environment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119066095A_ABST
    Figure CN119066095A_ABST
Patent Text Reader

Abstract

The embodiment of the invention provides an adaptive tuning method and system for a cloud distributed cache database, and the method comprises the steps: obtaining an expected value of at least one business index carried in a tuning request in response to the tuning request for the cloud distributed cache database; according to the expected value of the at least one business index and a network causal model corresponding to the cloud distributed cache database, determining an adjustment and optimization strategy corresponding to the cloud distributed cache database, the adjustment and optimization strategy comprising at least one target configuration item and an adjustment parameter of each target configuration item, the network causal model is used for analyzing a causal relationship from a configuration item to a business index variable based on multiple types of data, and predicting and analyzing a configuration parameter of the configuration item according to an expected value of a business index; and according to the tuning strategy, adjusting the configuration parameters of each target configuration item of the cloud distributed cache database into respective corresponding adjustment parameters. According to the method, the optimization efficiency of the business indexes can be improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The disclosed embodiments relate to the field of cloud computing technology, and in particular to a method and system for adaptively tuning a cloud-based distributed cache database. Background Art

[0002] The distributed cache database deployed in a multi-subject cloud environment needs to ensure high reliability, high availability, high throughput and low latency. However, since the usage scenarios often have diverse business requests, high concurrent access, real-time data return requirements and high deployment architecture complexity, it is difficult to optimize the system around business goals. With the increase of cluster scale and business system retrieval concurrency, the difficulty of manual parameter optimization around core business indicators increases rapidly.

[0003] Currently, the main means to solve this problem is to require professional and experienced database administrators to manually configure some configuration items of the cluster system according to business load requirements and operating characteristics, and then continuously try and error to optimize the business indicators on the core business side.

[0004] However, since there are a large number of configuration items in the cluster system, it takes a long time to manually determine which configuration item to adjust and its adjustment parameters, so the efficiency of optimizing business indicators using existing technologies is low. Summary of the invention

[0005] The disclosed embodiments provide a method and system for adaptively tuning a cloud-based distributed cache database, which can improve the optimization efficiency and quality of business indicators.

[0006] In a first aspect, an embodiment of the present disclosure provides a cloud-based distributed cache database adaptive tuning method, including:

[0007] In response to a tuning request for a cloud-based distributed cache database, obtaining an expected value of at least one business indicator carried in the tuning request;

[0008] Determine a tuning strategy corresponding to the Yunji distributed cache database according to the expected value of the at least one business indicator and the network causal model corresponding to the Yunji distributed cache database, wherein the tuning strategy includes at least one target configuration item and adjustment parameters of each target configuration item, wherein the network causal model is used to predict and analyze the configuration parameters of the configuration items based on the causal relationship between the configuration items and the business indicator variables constructed based on multiple types of data analysis;

[0009] According to the tuning strategy corresponding to the cloud-based distributed cache database, the configuration parameters of each target configuration item of the cloud-based distributed cache database are adjusted to their corresponding adjustment parameters.

[0010] In a second aspect, the present disclosure provides a cloud-based distributed cache database adaptive tuning system, including:

[0011] An acquisition unit, configured to respond to a tuning request for the cloud-based distributed cache database and acquire an expected value of at least one business indicator carried in the tuning request;

[0012] A determination unit, configured to determine a tuning strategy corresponding to the cloud-based distributed cache database according to the expected value of the at least one business indicator and the network causal model corresponding to the cloud-based distributed cache database, wherein the tuning strategy includes at least one target configuration item and an adjustment parameter for each target configuration item, wherein the network causal model is used to predict and analyze the configuration parameters of the configuration item based on the causal relationship from the configuration item to the business indicator variable constructed based on multiple types of data analysis;

[0013] The adjustment unit is used to adjust the configuration parameters of each target configuration item of the cloud-based distributed cache database to their corresponding adjustment parameters according to the tuning strategy corresponding to the cloud-based distributed cache database.

[0014] In a third aspect, an embodiment of the present disclosure provides an electronic device, including: a processor and a memory;

[0015] The memory stores computer-executable instructions;

[0016] The processor executes the computer-executable instructions stored in the memory, so that the at least one processor executes the cloud-based distributed cache database adaptive tuning method as described in the first aspect and various possible designs of the first aspect.

[0017] In a fourth aspect, an embodiment of the present disclosure provides a computer-readable storage medium, in which computer execution instructions are stored. When a processor executes the computer execution instructions, the cloud-based distributed cache database adaptive tuning method as described in the first aspect and various possible designs of the first aspect is implemented.

[0018] In a fifth aspect, an embodiment of the present disclosure provides a computer program product, including a computer program, which, when executed by a processor, implements the cloud-based distributed cache database adaptive tuning method as described in the first aspect and various possible designs of the first aspect.

[0019] The present embodiment provides a method and system for adaptive tuning of a cloud-based distributed cache database, the method comprising: in response to a tuning request for a cloud-based distributed cache database, obtaining an expected value of at least one business indicator carried in the tuning request; determining a tuning strategy corresponding to the cloud-based distributed cache database according to the expected value of the at least one business indicator and a network causal model corresponding to the cloud-based distributed cache database, the tuning strategy comprising at least one target configuration item and adjustment parameters for each target configuration item, wherein the network causal model is used to predict and analyze configuration parameters of configuration items based on causal relationships from configuration items to business indicator variables constructed based on multiple types of data analysis; and according to the tuning strategy corresponding to the cloud-based distributed cache database, adjusting the configuration parameters of each target configuration item of the cloud-based distributed cache database to their respective corresponding adjustment parameters. In this technical solution, since the network causal model is used to construct the causal relationship from configuration items to business indicator variables based on various types of data analysis, and predicts the configuration parameters of the configuration items based on the expected values ​​of the business indicators, the network causal model can determine the adjustment parameters of at least one target configuration item corresponding to the expected value of the business indicator, implement counterfactual reasoning to generate the tuning strategy of the Yunji distributed cache database, and realize the adaptive tuning of the Yunji distributed cache database, thereby improving the optimization efficiency of the business indicators. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0021] Figure 1 A schematic diagram of an application scenario of the cloud-based distributed cache database adaptive tuning method provided in an embodiment of the present disclosure;

[0022] Figure 2 The process of the cloud distributed cache database adaptive tuning method provided by the embodiment of the present disclosure Figure 1 ;

[0023] Figure 3 Schematic diagram of the cloud distributed cache database adaptive tuning method provided by the embodiment of the present disclosure Figure 1 ;

[0024] Figure 4 The process of the method for generating a network causal model provided in the embodiment of the present disclosure is as follows Figure 1 ;

[0025] Figure 5 Schematic diagram of updating the network causal model provided by the embodiment of the present disclosure Figure 1 ;

[0026] Figure 6 A schematic diagram of the structure of the cloud-based distributed cache database adaptive tuning system provided in an embodiment of the present disclosure;

[0027] Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solution and advantages of the embodiments of the present disclosure clearer, the technical solution in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.

[0029] First, the technical terms and technical background involved in the embodiments of the present disclosure are explained:

[0030] Apdex (Application Performance Index): An indicator for measuring application performance that evaluates the performance of an application or service based on user satisfaction.

[0031] P99 Latency (99 Percentile Latency): A business metric that measures the time it takes for 99% of all requests to complete.

[0032] QPS (Queries Per Second): A business metric used to measure the number of requests a system can handle per unit time, usually used to measure the performance of databases, servers, or network services.

[0033] Availability: A measure of the ability of a system, service or component to operate normally. It describes the degree to which a system can provide services when needed. High availability means that the system rarely or almost never fails and can continuously provide services to users.

[0034] Configuration item: usually used to store various parameters and settings that a program or system needs to use during operation. In the disclosed embodiment, the configuration item of the cache database can be any adjustable system parameter. For example, the configuration item can be active defrag cycle max (maximum number of cycles for memory defragmentation), or it can be Append-only (append only), or it can be Slave Read Only (read only from the server). Among them, Append-only means storing and processing data in append-only mode. Among them, Slave Read Only is a setting in the database replication architecture, which is used to indicate that the slave server (slave) is only used for read operations, not for write operations.

[0035] Business indicator: In an optimization problem, a business indicator is an indicator that the optimization algorithm attempts to maximize or minimize. In the embodiment of the present disclosure, the business indicator of the cache database may include one of the following: Apdex, P99 Latency, QPS, Availability.

[0036] The distributed cache database deployed in a multi-agent cloud environment needs to ensure high reliability, high availability, high throughput and low latency. However, since the usage scenarios often have diverse business requests, high concurrent access, real-time data return requirements and high deployment architecture complexity, it is difficult to optimize the system around business goals. With the increase in cluster scale and business system retrieval concurrency, the difficulty of manual parameter optimization around core business indicators such as Apdex, P99 Latency, QPS, and Availability increases rapidly.

[0037] Currently, the main means to solve this problem is to require professional and experienced database administrators to manually configure some configuration items of the cluster system according to business load requirements and operating characteristics, and then continuously try and error to optimize the business indicators on the core business side.

[0038] However, since there are a large number of configuration items in the cluster system, it takes a long time to manually determine which configuration item to adjust and its adjustment parameters, so the efficiency of optimizing business indicators using existing technologies is low.

[0039] It can be seen that how to quickly determine which configuration item of the cache database to adjust and how much to adjust in order to improve the efficiency of optimizing business indicators is a technical problem that needs to be solved urgently.

[0040] In response to the technical problems in the prior art, the inventor's technical conception is as follows: using Causal AI (Causal Artificial Intelligence) technology to discover the causal relationship between configuration items and business indicators from distributed cache data, and to achieve adaptive tuning of the distributed cache database in a cloud environment.

[0041] Accordingly, the specific steps may include: first, in response to a tuning request for a Yunji distributed cache database, obtaining an expected value of at least one business indicator carried in the tuning request. Then, based on the expected value of at least one business indicator and the network causal model corresponding to the Yunji distributed cache database, determining a tuning strategy corresponding to the Yunji distributed cache database, the tuning strategy including at least one target configuration item and adjustment parameters for each target configuration item, wherein the network causal model is used to predict and analyze the configuration parameters of the configuration items based on the causal relationship from the configuration items to the business indicator variables constructed based on multiple types of data analysis. Finally, according to the tuning strategy corresponding to the Yunji distributed cache database, the configuration parameters of each target configuration item of the Yunji distributed cache database are adjusted to their corresponding adjustment parameters.

[0042] In this technical solution, since the network causal model is used to construct the causal relationship from configuration items to business indicator variables based on various types of data analysis, and predicts the configuration parameters of the configuration items based on the expected values ​​of the business indicators, the network causal model can determine the adjustment parameters of at least one target configuration item corresponding to the expected value of the business indicator, implement counterfactual reasoning to generate the tuning strategy of the Yunji distributed cache database, and realize the adaptive tuning of the Yunji distributed cache database, thereby improving the optimization efficiency of the business indicators.

[0043] The application scenarios of the embodiments of the present disclosure are explained below:

[0044] The cloud-based distributed cache database adaptive tuning method provided by the embodiments of the present disclosure can be applied to various scenarios for tuning the cloud-based distributed cache database. Figure 1 A schematic diagram of an application scenario of a cloud-based distributed cache database adaptive tuning method provided by an embodiment of the present disclosure. Figure 1 As shown, the terminal 101 transmits the tuning request for the cloud-interface distributed cache database to the server 102 via the wireless network. The server 102 receives the tuning request, and through the cloud-interface distributed cache database adaptive tuning method provided by the embodiment of the present disclosure, determines a tuning strategy including at least one target configuration item and an adjustment parameter of each target configuration item, and adjusts each target configuration item of the cloud-interface distributed cache database to its corresponding adjustment parameter according to the tuning strategy.

[0045] The following is an example of a specific scenario: During the operation of the Yunji distributed cache database, the database administrator discovered from the business side that the Apdex index of the cache database was lower than 0.5. If we want to know which configuration variables need to be changed and to what values ​​the configuration variables should be adjusted in order to achieve the Apdex index of 0.8 under load conditions within a recent period of time. At this time, the target variable value of Apdex = 0.8 can be entered into the causal reasoning engine through the configuration change manager. The causal reasoning engine can find the observed indicator covariates that are linearly related to it through the SEM causal model and calculate its value range. When the prior knowledge of the values ​​of some observed indicator variables is known, the CBN causal network model is used to infer the maximum a posteriori probability of other indicators, and then the maximum probability value of each configuration variable is obtained, and the tuning strategy of the distributed cache database is obtained.

[0046] Figure 2 The process of adaptive tuning method of cloud distributed cache database provided by the embodiment of the present disclosure Figure 1 In the embodiment of the present disclosure, the execution subject of the adaptive tuning method may be a terminal or a server. Figure 2 As shown, the adaptive tuning method may include:

[0047] S201. In response to a tuning request for a cloud-based distributed cache database, obtain an expected value of at least one business indicator carried in the tuning request.

[0048] In the embodiments of the present disclosure, the business indicator may be any indicator for the cloud-based distributed cache database. Optionally, the business indicator may be a business indicator, stability indicator, security indicator, etc. of the cloud-based distributed cache database. Exemplarily, the business indicators may include: Apdex indicator, P99 Latency indicator, QPS indicator, Availability indicator, RTT (Round Trip Time, request response time) indicator, RPS (Requests Per Second, request processing throughput per second) indicator.

[0049] For example, Figure 3 As shown, the system tuning controller can receive the tuning request, obtain the expected value of at least one business indicator carried in the tuning request, generate a vector set and send it to the configuration change strategy generator.

[0050] S202. Determine a tuning strategy corresponding to the Yunji distributed cache database based on the expected value of at least one business indicator and the network causal model corresponding to the Yunji distributed cache database. The tuning strategy includes at least one target configuration item and adjustment parameters for each target configuration item. The network causal model is used to analyze the causal relationship from the configuration item to the business indicator variable based on various types of data analysis, and predicts and analyzes the configuration parameters of the configuration item based on the expected value of the business indicator.

[0051] In some embodiments, the network causal model includes a multi-layer heterogeneous causal model constructed based on multiple types of data, wherein different causal models are constructed and generated based on different types of data; wherein the multiple types of data include verification data, cloud observation data, and empirical data.

[0052] In some embodiments, a tuning strategy corresponding to the cloud-based distributed cache database is determined based on the expected value of at least one business indicator and the network causal model corresponding to the cloud-based distributed cache database, including: calling each layer of the causal model in turn according to the calling order from the bottom layer to the top layer of the multi-layer heterogeneous causal model in the network causal model, using the output of the current layer of the causal model as the input of the causal model of the previous layer connected to it, obtaining at least one target configuration item output by the top-level causal model and the adjustment parameters of each target configuration item as the tuning strategy corresponding to the cloud-based distributed cache database.

[0053] Optionally, the network causal model includes a three-layer heterogeneous causal model constructed based on multiple types of data; wherein, the first layer of the causal model of the three-layer heterogeneous causal model is constructed based on verification data in multiple types of data, and is used to determine the causal relationship between configuration items and operating indicators, the second layer of the causal model of the three-layer heterogeneous causal model is constructed based on cloud observation data in multiple types of data, and is used to determine the causal relationship between operating indicators, and the third layer of the causal model of the three-layer heterogeneous causal model is constructed based on cloud observation data and empirical data in multiple types of data, and is used to determine the causal relationship between operating indicators and business indicators.

[0054] Correspondingly, according to the expected value of at least one business indicator and the network causal model corresponding to the Yunji distributed cache database, the tuning strategy corresponding to the Yunji distributed cache database is determined, including: based on the expected value of at least one business indicator and the third-layer causal model, searching for at least one first operating indicator that is causally related and the value range of each first operating indicator; determining at least one second operating indicator that has a causal relationship with at least one first operating indicator and the value range of each second operating indicator according to at least one first operating indicator, the value range of each first operating indicator and the second-layer causal model; determining at least one target configuration item and the adjustment parameters of each target configuration item according to at least one first operating indicator, the value range of each first operating indicator, at least one second operating indicator, the value range of each second operating indicator and the first-layer causal model.

[0055] Optionally, the network causal model includes a three-layer heterogeneous causal model that communicates layer by layer from the bottom layer to the top layer, wherein the first layer causal model includes a Rubin Causal Model (RCM), the second layer causal model includes a Bayesian network causal model (Causal Bayesian Network, CBN), and the third layer causal model includes a structural equation model (Structure Equation Model, SEM).

[0056] For example, Figure 3 As shown, a configuration change strategy generator can receive a vector set of expected values ​​of at least one business indicator sent by a system tuning controller, send it to a causal inference engine to obtain a vector set containing configuration items that need to be changed, and generate a configuration change plan (i.e., a tuning strategy) and send it to a configuration change manager.

[0057] S203. According to the tuning strategy corresponding to the Yunji distributed cache database, the configuration parameters of each target configuration item of the Yunji distributed cache database are adjusted to their corresponding adjustment parameters.

[0058] For example, Figure 3 As shown, the variable manager can be configured to perform configuration changes on the distributed cache database deployed on the cloud according to the received tuning strategy.

[0059] The disclosed embodiment provides a method for adaptive tuning of a cloud-based distributed cache database: in response to a tuning request for a cloud-based distributed cache database, obtaining an expected value of at least one business indicator carried in the tuning request; determining a tuning strategy corresponding to the cloud-based distributed cache database according to the expected value of the at least one business indicator and a network causal model corresponding to the cloud-based distributed cache database, wherein the tuning strategy includes at least one target configuration item and adjustment parameters for each target configuration item, wherein the network causal model is used to predict and analyze configuration parameters of configuration items based on causal relationships from configuration items to business indicator variables constructed based on multiple types of data analysis; and according to the tuning strategy corresponding to the cloud-based distributed cache database, adjusting the configuration parameters of each target configuration item of the cloud-based distributed cache database to their respective corresponding adjustment parameters. In this technical solution, since the network causal model is used to construct the causal relationship from configuration items to business indicator variables based on various types of data analysis, and predicts the configuration parameters of the configuration items based on the expected values ​​of the business indicators, the network causal model can determine the adjustment parameters of at least one target configuration item corresponding to the expected value of the business indicator, implement counterfactual reasoning to generate the tuning strategy of the Yunji distributed cache database, and realize the adaptive tuning of the Yunji distributed cache database, thereby improving the optimization efficiency of the business indicators.

[0060] It should be noted that in the present disclosed embodiment, we first analyze and model the three different types of data obtained by the distributed cache database during the test period and the operation period, and then construct a network causal model through an innovative three-layer heterogeneous causal model fusion method.

[0061] In some embodiments, Figure 4 As shown, in response to a tuning request for a cloud-based distributed cache database, before obtaining the expected value of at least one business indicator carried in the tuning request, three types of data collected can be used to construct heterogeneous causal models using different modeling methods, and multiple constructed causal models can be further integrated to generate a network causal model. The generation method of the network causal model includes:

[0062] S401. Obtain a monitoring indicator data set of the Yunji distributed cache database, where the monitoring indicator data set includes verification data participating in the Yunji distributed cache database system test and cloud observation data and experience data during the operation of the Yunji distributed cache database.

[0063] In this step, the three types of data include verification data, cloud observation data, and empirical data.

[0064] Among them, the verification data can refer to the verification data collected through group random control experiments, or it can be obtained from cloud observation data through methods to eliminate bias, such as do-operator (operator overloading method). Exemplarily, the impact of turning on or off the memory defragmentation of the cloud distributed cache database (Redis) on the operating indicators such as the CPU usage, memory usage and cache hit rate of the Redis service node. The cache hit rate refers to the ratio of the number of times the requested data is found in the cache to the total number of requests. Among them, the configuration item is memory defragmentation, and the test parameters of the configuration item include: memory defragmentation is turned on or off. Among them, the operating indicators to be verified include the service node CPU usage, memory usage and cache hit rate. This type of data is small in size and it is easy to remove the influence of bias and confounding variables to obtain relative stability information, which is of greater value in generating safe and reliable causal relationships.

[0065] Among them, cloud observation data can be the operating indicator data directly observed during the system operation phase. Optionally, cloud observation data includes multiple operating indicators. This type of data is large in volume, changes rapidly, and records real-time detailed information about the internal performance and stability fluctuations of the system. However, cloud observation data does not include data caused by intervention actions such as configuration changes and instance migration. Due to the existence of bias and the influence of confounding variables introduced by external intervention (such as changes in load and configuration changes in the cloud basic operating environment), it is difficult for us to generate stable and reliable causal relationships directly from cloud observation data.

[0066] Among them, empirical data can be hidden variable data that cannot be directly observed. Optionally, empirical data includes multiple business indicators. For example: customer experience Apdex, system availability, it is necessary to define a mathematical model in combination with historical statistical data, industry standards and business needs. We call this type of data empirical data. Business operation empirical data is usually used to define the mapping relationship between observed indicator variables and hidden target indicator variables. By determining mathematical formulas or non-deterministic probabilistic reasoning (such as Bayesian reasoning and other methods), the quantitative value of the target hidden variable is obtained for decision-making reference.

[0067] For example, Figure 3As shown, the monitoring indicator data collector can be used to collect indicator data of the system test phase and operation phase from the cloud environment, and the time series event data extracted from the cloud environment log data to obtain the monitoring indicator data set of the cloud distributed cache database, and store the monitoring indicator data set in the monitoring database. Through the monitoring structure data collector, the deployment relationship of each component of the cache database distributed in the cloud environment is collected from monitoring systems such as CMDB and APM, and the mutual call relationship during the query execution process is used to preliminarily associate the operating indicators of different components and store them in the monitoring database. Among them, the monitoring database is used to centrally store the monitoring data of the cloud distributed database during the test period and the operation period, so as to drive the subsequent Causal AI causal discovery algorithm to generate a causal model.

[0068] S402: Generate a network causal model based on the monitoring indicator data set.

[0069] In some embodiments, the cloud observation data includes multiple operating indicators, and the empirical data includes multiple business indicators. The network causal model includes a three-layer heterogeneous causal model; wherein the first layer causal model of the three-layer heterogeneous causal model is used to determine the causal relationship between the configuration items and the operating indicators, the second layer causal model of the three-layer heterogeneous causal model is used to determine the causal relationship between the operating indicators, and the third layer causal model of the three-layer heterogeneous causal model is used to determine the causal relationship between the operating indicators and the business indicators; accordingly, this step may include the following steps (1) to (2):

[0070] (1) Generate a first-layer causal model corresponding to the cloud-based distributed cache database based on the intervention effect of the test parameters of the configuration items in each verification data and the test value of at least one operating indicator to be verified; and generate a second-layer causal model corresponding to the cloud-based distributed cache database based on the dependency and subordinate relationships among multiple operating indicators; and generate a third-layer causal model corresponding to the cloud-based distributed cache database based on multiple business indicators, multiple operating indicators and preset functional relationships.

[0071] In the implementation of this disclosure, if Figure 3 As shown, through the RCM causal analysis model generator, based on the configuration item (knob) variable set provided by the distributed cache database and the monitoring indicator variable set collected in the monitoring database, the RCM causal analysis method is used to fit the historical monitoring data to find an RCM model with significant causal relationship.

[0072] In some embodiments, the first-layer causal model includes a potential causal RCM model. Accordingly, according to the intervention effect of the test parameters of the configuration items in each verification data and the test value of at least one operating indicator to be verified, the first-layer causal model corresponding to the cloud-based distributed cache database is generated, including: for each test parameter of the configuration item in the verification data and the test value of at least one operating indicator to be verified, determining the intervention effect of the test parameters of the configuration item on the test value of at least one operating indicator to be verified; according to the intervention effect, determining the correspondence between at least one configuration item, the adjustment parameters of each configuration item and at least one operating indicator, the value range of each operating indicator, and generating the RCM model corresponding to the cloud-based distributed cache database.

[0073] Optionally, the intervention effect comprises an average intervention effect and / or a conditional intervention effect.

[0074] It should be noted that verification data collection needs to be done during the cache database development and testing phase, based on data from a specific verification environment and load stress test. During the stress test, since the potential unobservable variables on the client side and the operating environment dependency side that are uncontrollable in the production phase can be analyzed, we design test cases and analyze the Average Treatment Effect (ATE) and / or Conditional Average Treatment Effect (CATE) based on the Potential Outcome framework to obtain a relatively stable RCM causal relationship model between the configuration item mapping the control variable to the observed variable.

[0075] In some embodiments, the second-layer causal model includes a Bayesian network causal CBN model. Accordingly, based on the dependencies and subordinate relationships between multiple operating indicators, a second-layer causal model corresponding to the cloud-interconnected distributed cache database is generated, including: determining the dependencies and subordinate relationships between multiple operating indicators corresponding to the target event based on the application performance management APM and the configuration management database CMDB, wherein the target event includes a scheduling event or an upgrade event; based on the dependencies and subordinate relationships between multiple operating indicators corresponding to the target event and the conditional probability table CPT, a CBN model corresponding to the cloud-interconnected distributed cache database is generated.

[0076] For example, Figure 3 As shown, through the SCM (Structural Causal Model, structural causal graph model) generator: based on the monitoring indicator variable set collected in the monitoring database, the SCM algorithm is used to fit the historical monitoring data to generate a CBN model.

[0077] It should be noted that cloud observation data is the observation variables extracted from the indicators, logs and other information collected by monitoring tools during the operation phase of the cache database. Due to the rapid change speed and the impact of changes in potential unobservable variables in the business load and cloud operating environment, we cannot use the RCM method to build a causal model. In response to the above technical problems, we propose to first collect event logs including scheduling and upgrading from the environment operation log; then extract the event in the log as an observation variable reflecting the change of the state of the environment event, and merge it with the limited observation variables obtained from the monitoring indicators; finally, the Probability Graph Model (PGM) modeling method is used to generate a causal Bayesian network graph model (directed graph model. During the generation process, the deployment architecture collected from data sources such as APM (Application Performance Management) and CMDB (Configuration Management Database) is used as the basic structure of the relationship between indicators, and the causal relationship is generated on this basis.

[0078] In some embodiments, the third-layer causal model includes a structural equation SEM model. The preset functional relationship includes a first functional relationship and a second functional relationship. Accordingly, according to multiple business indicators, multiple operating indicators and the preset functional relationship, a third-layer causal model corresponding to the cloud-based distributed cache database is generated, including: for each business indicator, constructing a first functional relationship between the value of the business indicator and the value range of at least one first target operating indicator; determining at least one second target operating indicator that is linearly related to the first target operating indicator from multiple operating indicators, and determining the second functional relationship between the first target operating indicator and the second target operating indicator; according to the first functional relationship and the second functional relationship, establishing a correlation relationship between the value range of at least one second target operating indicator and the value of the business indicator, and generating a SEM model corresponding to the cloud-based distributed cache database.

[0079] The empirical data comes from target indicators that cannot be directly observed on the business system side, such as Apdex user experience and RPS average request throughput per second. These indicators cannot be directly observed and require established calculation logic to be calculated from observable indicators on the business side. For example, the calculation formula for Apdex is:

[0080] Apdex index = (expected number of request responses + tolerable number of request responses / 2) / total number of request samples

[0081] During operation, the cloud cache database and the client-side business system pass through the public network whose running status is unobservable, and the network delay, congestion and jitter are unknown. However, since the client-side request delay y in the Apdex calculation formula is calculable, the execution delay x in the database is observable, and it conforms to the linear correlation, we can use linear regression to build a causal influence relationship based on the collected cloud observation data: Y = ax + e; where a is the correlation coefficient, and e is the interference caused by factors such as the public network connection status.

[0082] According to the known influence relationship between the business target variables and observation variables of the distributed cache database, combined with the collected monitoring indicator data, a structural causal model (Structure Equation Model, SEM model) is generated, and a linear causal relationship is constructed between the distributed cache database operation period monitoring indicator variables and the business target variables in the form of a path diagram.

[0083] For example, Figure 3 As shown, the SEM structural equation model generator can be used to build a SEM model by fitting a linear function using a regression algorithm based on a set of pre-defined business target variables (such as Apdex, Availability, etc.) and the most relevant observed variables in the CBM that are associated with business experience.

[0084] (2) The first-layer causal model, the second-layer causal model, and the third-layer causal model are integrated to generate a network causal model.

[0085] Optionally, this step may include: fusing the first-layer causal model with the second-layer causal model with multiple third operating indicators as the connection point, fusing the second-layer causal model with the third-layer causal model with multiple fourth operating indicators as the connection point, and generating a network causal model. The third operating indicator is an operating indicator causally related to the test parameter of the configuration item determined based on the first-layer causal model; the third operating indicator is an operating indicator causally related to the expected value of the business indicator determined based on the third-layer causal model.

[0086] In some embodiments, the first-layer causal model includes a potential causal RCM model, the second-layer causal model includes a Bayesian network causal CBN model, and the third-layer causal model includes a structural equation SEM model. Exemplarily, in the process of generating a network causal model, we first start from the configuration item, find the observed variables with the highest causal significance associated with the configuration item, and use this set of observed variables as a combination point to splice the RCM model and the CBN model. Secondly, starting from the business indicators, find a set of causally related observed variables, and use this set as a combination point to splice the SEM model and the CBN model. By integrating the RCM model, the CBN model and the SEM model, a causal inference model (i.e., a network causal model) is obtained that can go from the target effect to the intervention cause and from the intervention cause to the target effect.

[0087] For example, Figure 3 As shown, the causal inference engine integrates the already constructed RCM causal model, CBN causal model and SEM causal model to generate the above causal inference model. In the loading inference engine, the inference service calling interface is provided in the form of service.

[0088] The generation process of the NeoCG network causal model is explained below through a specific example.

[0089] (1) Collect distributed cache data operation period monitoring data and environmental log data, define the database configurable knob as the intervention variable (Treatment Variable), the monitoring indicator collection data as the observed variable (ObservableVariable), and the unobservable environmental state and customer-side experience-related state as the latent variable (Latent Variable).

[0090] (2) Through the potential outcome analysis (Potential Outcome Analysis) of the database runtime configuration variables, the average ATE impact effect between the database configuration items and the observed variables is found, and the causal relationship between the database configuration items and the target observed variables is determined;

[0091] (3) Based on the physical deployment architecture read out by CMDB, APM and other systems, the relationship between each observed variable is associated, and the events in the operating environment log are extracted as an observable indicator to reflect the events that occurred in the hidden variables of the environment. On this basis, combined with the monitoring data, the probabilistic correlation and causal relationship between the observed variables are found through the causal architecture discovery algorithm and the Bayesian network generation algorithm;

[0092] (4) Based on the structural equation model (SEM), the structural equation between the observed variables and the target variables of customer-side experience-related states is constructed, and the corresponding parameters are generated by fitting the cloud observation data to construct the relationship between the observed variables and the target hidden variables;

[0093] (5) The above models, which are modeled by potential outcome analysis, structural causal network and structural equation model, are integrated into a three-layer hybrid causal graph model, which we call NeoCG network causal model (analogous to the six-layer hybrid network structure of the human brain's neocortex, the prefrontal cortex, hence the name Neo Causal Graph).

[0094] It should be noted that the network causal model can also be automatically updated after the state of the latent variables in the operating environment changes.

[0095] In some embodiments, after receiving the tuning strategy, the configuration change manager executes the configuration change on the target cache database, and observes through the monitoring system whether the tuning strategy makes the target variable meet the expectations; if it does not meet the expectations, it calls to update the monitoring indicator data set and regenerate the model. Correspondingly, according to the tuning strategy corresponding to the Yunji distributed cache database, after adjusting each target configuration item of the Yunji distributed cache database to the corresponding adjustment parameters, it also includes: if the empirical data corresponding to the business indicator does not meet the expected value of the business indicator, then the monitoring indicator data set is updated; through the updated monitoring indicator data set, a new network causal model is generated. Exemplarily, the expected value of the business indicator Apdex is 0.8. If the empirical data corresponding to the business indicator Apdex is 0.7, which does not meet the expected value of the business indicator, then the monitoring indicator data set is updated.

[0096] In this step, the amount of data in the updated monitoring indicator data set is larger than the amount of data in the original monitoring indicator data set, so that the network causal model can be incrementally updated.

[0097] For example, Figure 5 As shown, the steps of automatically updating the network causal model include:

[0098] (1) Define the target variable expected value vector based on the data administrator's expectations of the indicators of the client-side cache database customer experience (such as the round-trip request response time RTT and the request processing throughput per second RPT).

[0099] (2) After receiving the expected value vector, the system tuning controller sends it to the configuration change strategy generator, which is then forwarded to the NeoCG inference engine.

[0100] (3) The NeoCG inference engine infers the set of variable Knob items and encapsulates their values ​​into a vector and returns it to the configuration change strategy generator.

[0101] (4) The configuration change strategy generator obtains the inference results and then generates a tuning strategy for the target cache database.

[0102] (5) After receiving the tuning strategy, the configuration change manager executes the configuration change on the target cache database and observes through the monitoring system whether the tuning strategy makes the target variables meet expectations.

[0103] (6) If the expectation is not met, the monitoring indicator dataset is updated and the CBN model is regenerated.

[0104] (7) After the NeoCG inference engine obtains the updated CBN model, it incrementally updates the NeoCG model again and continues with step (3).

[0105] In the disclosed embodiment, since we have added a process for judging the execution effect of the tuning strategy after the database tuning strategy is generated, a mechanism for automatically updating the network causal model can be implemented after the state of the potential variables in the operating environment changes, thereby improving the robustness of the system.

[0106] Figure 6 A schematic diagram of the structure of the cloud-based distributed cache database adaptive tuning system provided by the embodiment of the present disclosure is shown in FIG. Figure 6 As shown, the cloud-based distributed cache database adaptive tuning system includes:

[0107] The acquisition unit 601 is used to respond to a tuning request for the cloud-based distributed cache database and acquire an expected value of at least one business indicator carried in the tuning request;

[0108] A determination unit 602 is used to determine a tuning strategy corresponding to the cloud-based distributed cache database according to the expected value of the at least one business indicator and the network causal model corresponding to the cloud-based distributed cache database, wherein the tuning strategy includes at least one target configuration item and an adjustment parameter for each target configuration item, wherein the network causal model is used to predict and analyze the configuration parameters of the configuration item based on the causal relationship between the configuration item and the business indicator variable constructed based on multiple types of data analysis;

[0109] The adjustment unit 603 is used to adjust the configuration parameters of each target configuration item of the cloud-based distributed cache database to their corresponding adjustment parameters according to the tuning strategy corresponding to the cloud-based distributed cache database.

[0110] According to one or more embodiments of the present disclosure, the network causal model includes a multi-layer heterogeneous causal model constructed based on multiple types of data, wherein different causal models are constructed and generated based on different types of data; wherein the multiple types of data include verification data, cloud observation data, and empirical data.

[0111] According to one or more embodiments of the present disclosure, the network causal model includes a three-layer heterogeneous causal model constructed based on multiple types of data; wherein, the first layer causal model of the three-layer heterogeneous causal model is constructed based on verification data in the multiple types of data, and is used to determine the causal relationship between configuration items and operating indicators, the second layer causal model of the three-layer heterogeneous causal model is constructed based on cloud observation data in the multiple types of data, and is used to determine the causal relationship between operating indicators, and the third layer causal model of the three-layer heterogeneous causal model is constructed based on cloud observation data and experience data in the multiple types of data, and is used to determine the causal relationship between operating indicators and business indicators.

[0112] According to one or more embodiments of the present disclosure, the determination unit 602 determines the tuning strategy corresponding to the cloud-based distributed cache database based on the expected value of the at least one business indicator and the network causal model corresponding to the cloud-based distributed cache database, specifically including: calling each layer of the causal model in turn according to the calling order from the bottom layer to the top layer of the multi-layer heterogeneous causal model in the network causal model, using the output of the current layer of the causal model as the input of the causal model of the previous layer connected to it, obtaining at least one target configuration item output by the top-level causal model and the adjustment parameters of each target configuration item as the tuning strategy corresponding to the cloud-based distributed cache database.

[0113] According to one or more embodiments of the present disclosure, the network causal model includes a three-layer heterogeneous causal model that communicates layer by layer from the bottom layer to the top layer, wherein the first layer causal model includes a latent causal RCM model, the second layer causal model includes a Bayesian network causal CBN model, and the third layer causal model includes a structural equation SEM model.

[0114] According to one or more embodiments of the present disclosure, the cloud-based distributed cache database adaptive tuning system also includes: a model generation unit; the model generation unit is used to obtain a monitoring indicator data set of the cloud-based distributed cache database, the monitoring indicator data set includes verification data participating in the cloud-based distributed cache database system test and cloud observation data and experience data during the operation of the cloud-based distributed cache database; based on the monitoring indicator data set, the network causal model is generated.

[0115] According to one or more embodiments of the present disclosure, the cloud-based distributed cache database adaptive tuning system also includes: an update unit; the update unit is used to update the monitoring indicator data set if the empirical data corresponding to the business indicator does not meet the expected value of the business indicator; and generate a new network causal model through the updated monitoring indicator data set.

[0116] According to one or more embodiments of the present disclosure, the model generation unit generates the network causal model based on the monitoring indicator data set, including: generating a first-layer causal model corresponding to the cloud-based distributed cache database according to the intervention effect of the test parameters of the configuration items in each verification data and the test value of at least one operation indicator to be verified, and generating a second-layer causal model corresponding to the cloud-based distributed cache database according to the dependency and subordination between the multiple operation indicators, and generating a third-layer causal model corresponding to the cloud-based distributed cache database according to the multiple business indicators, the multiple operation indicators and preset functional relationships; fusing the first-layer causal model, the second-layer causal model and the third-layer causal model to generate the network causal model.

[0117] According to one or more embodiments of the present disclosure, the model generation unit fuses the first-layer causal model, the second-layer causal model and the third-layer causal model to generate the network causal model, including: fusing the first-layer causal model with the second-layer causal model with a plurality of third operating indicators as a connection point, and fusing the second-layer causal model with the third-layer causal model with a plurality of fourth operating indicators as a connection point to generate the network causal model; wherein the third operating indicator is an operating indicator causally related to the test parameters of the configuration item determined based on the first-layer causal model; and the third operating indicator is an operating indicator causally related to the expected value of the business indicator determined based on the third-layer causal model.

[0118] According to one or more embodiments of the present disclosure, the first-layer causal model includes a potential causal RCM model; accordingly, the model generation unit generates a first-layer causal model corresponding to the Yunji distributed cache database according to the intervention effect of the test parameters of the configuration items in each verification data and the test value of at least one operating indicator to be verified, including: for each test parameter of the configuration item in the verification data and the test value of at least one operating indicator to be verified, determining the intervention effect of the test parameter of the configuration item on the test value of the at least one operating indicator to be verified; according to the intervention effect, determining the correspondence between at least one configuration item, the adjustment parameter of each configuration item and at least one operating indicator, the value range of each operating indicator, and generating the potential causal model corresponding to the Yunji distributed cache database.

[0119] According to one or more embodiments of the present disclosure, the second-layer causal model includes a Bayesian network causal CBN model; accordingly, the model generation unit generates a second-layer causal model corresponding to the cloud-interconnected distributed cache database according to the dependencies and subordination between the multiple operating indicators, including: determining the dependencies and subordination between the multiple operating indicators corresponding to the target event according to the application performance management and configuration management database, wherein the target event includes a scheduling event or an upgrade event; generating a Bayesian network causal model corresponding to the cloud-interconnected distributed cache database according to the dependencies, subordination and conditional probability table between the multiple operating indicators corresponding to the target event.

[0120] According to one or more embodiments of the present disclosure, the third-layer causal model includes a structural equation SEM model; accordingly, the model generation unit generates a third-layer causal model corresponding to the Yunji distributed cache database according to the multiple business indicators, the multiple operation indicators and the preset functional relationship, including: for each business indicator, constructing a first functional relationship between the numerical value of the business indicator and the value range of at least one first target operation indicator; determining at least one second target operation indicator linearly related to the first target operation indicator from the multiple operation indicators, and determining a second functional relationship between the first target operation indicator and the second target operation indicator; according to the first functional relationship and the second functional relationship, establishing an association relationship between the value range of the at least one second target operation indicator and the numerical value of the business indicator, and generating a structural equation model corresponding to the Yunji distributed cache database.

[0121] refer to Figure 7 , which shows a schematic diagram of the structure of an electronic device 700 suitable for implementing the embodiment of the present disclosure, and the electronic device 700 may be a terminal device or a server. The terminal device may include but is not limited to mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, personal digital assistants (PDAs), tablet computers (Portable Android Devices, PADs), portable multimedia players (PMPs), vehicle terminals (such as vehicle navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 7 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0122] like Figure 7As shown, the electronic device 700 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 701, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 702 or a program loaded from a storage device 708 to a random access memory (RAM) 703. Various programs and data required for the operation of the electronic device 700 are also stored in the RAM 703. The processing device 701, the ROM 702, and the RAM 703 are connected to each other via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0123] Typically, the following devices may be connected to the I / O interface 705: input devices 706 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; output devices 707 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 708 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 709. The communication devices 709 may allow the electronic device 700 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 7 The electronic device 700 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0124] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device 709, or installed from a storage device 708, or installed from a ROM 702. When the computer program is executed by the processing device 701, the above-mentioned functions defined in the method of the embodiment of the present disclosure are executed.

[0125] It should be noted that the computer-readable medium disclosed above may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0126] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0127] The computer-readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device executes the method shown in the above embodiment.

[0128] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0129] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0130] The units involved in the embodiments described in the present disclosure may be implemented by software or hardware. The name of a unit does not limit the unit itself in some cases. For example, the first acquisition unit may also be described as a "unit for acquiring at least two Internet Protocol addresses".

[0131] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0132] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0133] In a first aspect, according to one or more embodiments of the present disclosure, a method for adaptively tuning a cloud-based distributed cache database is provided, comprising:

[0134] In response to a tuning request for a cloud-based distributed cache database, obtaining an expected value of at least one business indicator carried in the tuning request;

[0135] Determine a tuning strategy corresponding to the Yunji distributed cache database according to the expected value of the at least one business indicator and the network causal model corresponding to the Yunji distributed cache database, wherein the tuning strategy includes at least one target configuration item and adjustment parameters of each target configuration item, wherein the network causal model is used to predict and analyze the configuration parameters of the configuration items based on the causal relationship between the configuration items and the business indicator variables constructed based on multiple types of data analysis;

[0136] According to the tuning strategy corresponding to the cloud-based distributed cache database, the configuration parameters of each target configuration item of the cloud-based distributed cache database are adjusted to their corresponding adjustment parameters.

[0137] According to one or more embodiments of the present disclosure, the network causal model includes a multi-layer heterogeneous causal model constructed based on multiple types of data, wherein different causal models are constructed and generated based on different types of data; wherein the multiple types of data include verification data, cloud observation data, and empirical data.

[0138] According to one or more embodiments of the present disclosure, the network causal model includes a three-layer heterogeneous causal model constructed based on multiple types of data; wherein, the first layer causal model of the three-layer heterogeneous causal model is constructed based on verification data in the multiple types of data, and is used to determine the causal relationship between configuration items and operating indicators, the second layer causal model of the three-layer heterogeneous causal model is constructed based on cloud observation data in the multiple types of data, and is used to determine the causal relationship between operating indicators, and the third layer causal model of the three-layer heterogeneous causal model is constructed based on cloud observation data and experience data in the multiple types of data, and is used to determine the causal relationship between operating indicators and business indicators.

[0139] According to one or more embodiments of the present disclosure, the tuning strategy corresponding to the cloud-based distributed cache database is determined based on the expected value of the at least one business indicator and the network causal model corresponding to the cloud-based distributed cache database, including: calling each layer of the causal model in turn according to the calling order of the multi-layer heterogeneous causal models from the bottom layer to the top layer in the network causal model, using the output of the causal model of the current layer as the input of the causal model of the previous layer connected to it, obtaining at least one target configuration item output by the top-level causal model and the adjustment parameters of each target configuration item as the tuning strategy corresponding to the cloud-based distributed cache database.

[0140] According to one or more embodiments of the present disclosure, the network causal model includes a three-layer heterogeneous causal model that communicates layer by layer from the bottom layer to the top layer, wherein the first layer causal model includes a latent causal model, the second layer causal model includes a Bayesian network causal model, and the third layer causal model includes a structural equation model.

[0141] According to one or more embodiments of the present disclosure, before the method responds to a tuning request for a Yunji distributed cache database and obtains the expected value of at least one business indicator carried in the tuning request, the method also includes: obtaining a monitoring indicator data set for the Yunji distributed cache database, the monitoring indicator data set including verification data participating in the Yunji distributed cache database system test and cloud observation data and experience data during the operation of the Yunji distributed cache database; generating the network causal model based on the monitoring indicator data set.

[0142] According to one or more embodiments of the present disclosure, after adjusting each target configuration item of the Yunji distributed cache database to the corresponding adjustment parameters according to the tuning strategy corresponding to the Yunji distributed cache database, it also includes: if the empirical data corresponding to the business indicator does not meet the expected value of the business indicator, updating the monitoring indicator data set; generating a new network causal model through the updated monitoring indicator data set.

[0143] According to one or more embodiments of the present disclosure, generating the network causal model based on the monitoring indicator data set includes: generating a first-layer causal model corresponding to the cloud-interconnected distributed cache database based on the intervention effect of the test parameters of the configuration items in each verification data and the test value of at least one operation indicator to be verified, and generating a second-layer causal model corresponding to the cloud-interconnected distributed cache database based on the dependency and subordination between the multiple operation indicators, and generating a third-layer causal model corresponding to the cloud-interconnected distributed cache database based on the multiple business indicators, the multiple operation indicators and preset functional relationships; fusing the first-layer causal model, the second-layer causal model and the third-layer causal model to generate the network causal model.

[0144] According to one or more embodiments of the present disclosure, the fusing of the first-layer causal model, the second-layer causal model and the third-layer causal model to generate the network causal model includes: fusing the first-layer causal model with the second-layer causal model with a plurality of third operating indicators as a connection point, and fusing the second-layer causal model with the third-layer causal model with a plurality of fourth operating indicators as a connection point to generate the network causal model; wherein the third operating indicator is an operating indicator causally related to a test parameter of a configuration item determined based on the first-layer causal model; and the third operating indicator is an operating indicator causally related to an expected value of a business indicator determined based on the third-layer causal model.

[0145] According to one or more embodiments of the present disclosure, the first-layer causal model includes a potential causal model; accordingly, the first-layer causal model corresponding to the Yunji distributed cache database is generated according to the intervention effect of the test parameters of the configuration items in each verification data and the test value of at least one operating indicator to be verified, including: for each test parameter of the configuration item in the verification data and the test value of at least one operating indicator to be verified, determining the intervention effect of the test parameter of the configuration item on the test value of the at least one operating indicator to be verified; according to the intervention effect, determining the correspondence between at least one configuration item, the adjustment parameter of each configuration item and at least one operating indicator, the value range of each operating indicator, and generating the potential causal model corresponding to the Yunji distributed cache database.

[0146] According to one or more embodiments of the present disclosure, the second-layer causal model includes a Bayesian network causal model; accordingly, the second-layer causal model corresponding to the cloud-interconnected distributed cache database is generated according to the dependencies and subordination between the multiple operating indicators, including: determining the dependencies and subordination between the multiple operating indicators corresponding to the target event according to the application performance management and configuration management database, wherein the target event includes a scheduling event or an upgrade event; generating the Bayesian network causal model corresponding to the cloud-interconnected distributed cache database according to the dependencies, subordination and conditional probability table between the multiple operating indicators corresponding to the target event.

[0147] According to one or more embodiments of the present disclosure, the third-layer causal model includes a structural equation model; accordingly, the third-layer causal model corresponding to the cloud-based distributed cache database is generated based on the multiple business indicators, the multiple operating indicators and the preset functional relationship, including: for each business indicator, constructing a first functional relationship between the numerical value of the business indicator and the value range of at least one first target operating indicator; determining at least one second target operating indicator linearly related to the first target operating indicator from the multiple operating indicators, and determining a second functional relationship between the first target operating indicator and the second target operating indicator; based on the first functional relationship and the second functional relationship, establishing an association relationship between the value range of the at least one second target operating indicator and the numerical value of the business indicator, and generating the structural equation model corresponding to the cloud-based distributed cache database.

[0148] In a second aspect, according to one or more embodiments of the present disclosure, a cloud-based distributed cache database adaptive tuning system is provided, including:

[0149] An acquisition unit, configured to respond to a tuning request for the cloud-based distributed cache database and acquire an expected value of at least one business indicator carried in the tuning request;

[0150] A determination unit, configured to determine a tuning strategy corresponding to the cloud-based distributed cache database according to the expected value of the at least one business indicator and the network causal model corresponding to the cloud-based distributed cache database, wherein the tuning strategy includes at least one target configuration item and an adjustment parameter for each target configuration item, wherein the network causal model is used to predict and analyze the configuration parameters of the configuration item based on the causal relationship from the configuration item to the business indicator variable constructed based on multiple types of data analysis;

[0151] The adjustment unit is used to adjust the configuration parameters of each target configuration item of the cloud-based distributed cache database to their corresponding adjustment parameters according to the tuning strategy corresponding to the cloud-based distributed cache database.

[0152] According to one or more embodiments of the present disclosure, the network causal model includes a multi-layer heterogeneous causal model constructed based on multiple types of data, wherein different causal models are constructed and generated based on different types of data; wherein the multiple types of data include verification data, cloud observation data, and empirical data.

[0153] According to one or more embodiments of the present disclosure, the network causal model includes a three-layer heterogeneous causal model constructed based on multiple types of data; wherein, the first layer causal model of the three-layer heterogeneous causal model is constructed based on verification data in the multiple types of data, and is used to determine the causal relationship between configuration items and operating indicators, the second layer causal model of the three-layer heterogeneous causal model is constructed based on cloud observation data in the multiple types of data, and is used to determine the causal relationship between operating indicators, and the third layer causal model of the three-layer heterogeneous causal model is constructed based on cloud observation data and experience data in the multiple types of data, and is used to determine the causal relationship between operating indicators and business indicators.

[0154] According to one or more embodiments of the present disclosure, the determination unit determines the tuning strategy corresponding to the Yunji distributed cache database based on the expected value of the at least one business indicator and the network causal model corresponding to the Yunji distributed cache database, specifically including: calling each layer of causal models in turn according to the calling order from the bottom layer to the top layer of the multi-layer heterogeneous causal models in the network causal model, using the output of the current layer of causal model as the input of the causal model of the previous layer connected to it, obtaining at least one target configuration item output by the top-level causal model and the adjustment parameters of each target configuration item as the tuning strategy corresponding to the Yunji distributed cache database.

[0155] According to one or more embodiments of the present disclosure, the network causal model includes a three-layer heterogeneous causal model that communicates layer by layer from the bottom layer to the top layer, wherein the first layer causal model includes a latent causal model, the second layer causal model includes a Bayesian network causal model, and the third layer causal model includes a structural equation model.

[0156] According to one or more embodiments of the present disclosure, the cloud-based distributed cache database adaptive tuning system also includes: a model generation unit; the model generation unit is used to obtain a monitoring indicator data set of the cloud-based distributed cache database, the monitoring indicator data set includes verification data participating in the cloud-based distributed cache database system test and cloud observation data and experience data during the operation of the cloud-based distributed cache database; based on the monitoring indicator data set, the network causal model is generated.

[0157] According to one or more embodiments of the present disclosure, the cloud-based distributed cache database adaptive tuning system also includes: an update unit; the update unit is used to update the monitoring indicator data set if the empirical data corresponding to the business indicator does not meet the expected value of the business indicator; and generate a new network causal model through the updated monitoring indicator data set.

[0158] According to one or more embodiments of the present disclosure, the model generation unit generates the network causal model based on the monitoring indicator data set, including: generating a first-layer causal model corresponding to the cloud-based distributed cache database according to the intervention effect of the test parameters of the configuration items in each verification data and the test value of at least one operation indicator to be verified, and generating a second-layer causal model corresponding to the cloud-based distributed cache database according to the dependency and subordination between the multiple operation indicators, and generating a third-layer causal model corresponding to the cloud-based distributed cache database according to the multiple business indicators, the multiple operation indicators and preset functional relationships; fusing the first-layer causal model, the second-layer causal model and the third-layer causal model to generate the network causal model.

[0159] According to one or more embodiments of the present disclosure, the model generation unit fuses the first-layer causal model, the second-layer causal model and the third-layer causal model to generate the network causal model, including: fusing the first-layer causal model with the second-layer causal model with a plurality of third operating indicators as a connection point, and fusing the second-layer causal model with the third-layer causal model with a plurality of fourth operating indicators as a connection point to generate the network causal model; wherein the third operating indicator is an operating indicator causally related to the test parameters of the configuration item determined based on the first-layer causal model; and the third operating indicator is an operating indicator causally related to the expected value of the business indicator determined based on the third-layer causal model.

[0160] According to one or more embodiments of the present disclosure, the first-layer causal model includes a potential causal RCM model; accordingly, the model generation unit generates a first-layer causal model corresponding to the Yunji distributed cache database according to the intervention effect of the test parameters of the configuration items in each verification data and the test value of at least one operating indicator to be verified, including: for each test parameter of the configuration item in the verification data and the test value of at least one operating indicator to be verified, determining the intervention effect of the test parameter of the configuration item on the test value of the at least one operating indicator to be verified; according to the intervention effect, determining the correspondence between at least one configuration item, the adjustment parameter of each configuration item and at least one operating indicator, the value range of each operating indicator, and generating the potential causal model corresponding to the Yunji distributed cache database.

[0161] According to one or more embodiments of the present disclosure, the second-layer causal model includes a Bayesian network causal CBN model; accordingly, the model generation unit generates a second-layer causal model corresponding to the cloud-interconnected distributed cache database according to the dependencies and subordination between the multiple operating indicators, including: determining the dependencies and subordination between the multiple operating indicators corresponding to the target event according to the application performance management and configuration management database, wherein the target event includes a scheduling event or an upgrade event; generating a Bayesian network causal model corresponding to the cloud-interconnected distributed cache database according to the dependencies, subordination and conditional probability table between the multiple operating indicators corresponding to the target event.

[0162] According to one or more embodiments of the present disclosure, the third-layer causal model includes a structural equation SEM model; accordingly, the model generation unit generates a third-layer causal model corresponding to the Yunji distributed cache database according to the multiple business indicators, the multiple operation indicators and the preset functional relationship, including: for each business indicator, constructing a first functional relationship between the numerical value of the business indicator and the value range of at least one first target operation indicator; determining at least one second target operation indicator linearly related to the first target operation indicator from the multiple operation indicators, and determining a second functional relationship between the first target operation indicator and the second target operation indicator; according to the first functional relationship and the second functional relationship, establishing an association relationship between the value range of the at least one second target operation indicator and the numerical value of the business indicator, and generating a structural equation model corresponding to the Yunji distributed cache database.

[0163] In a third aspect, according to one or more embodiments of the present disclosure, there is provided an electronic device, comprising: at least one processor and a memory;

[0164] The memory stores computer-executable instructions;

[0165] The at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor executes the cloud-based distributed cache database adaptive tuning method as described in the first aspect and various possible designs of the first aspect.

[0166] In a fourth aspect, according to one or more embodiments of the present disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer execution instructions. When a processor executes the computer execution instructions, the cloud-based distributed cache database adaptive tuning method as described in the first aspect and various possible designs of the first aspect is implemented.

[0167] In a fifth aspect, according to one or more embodiments of the present disclosure, a computer program product is provided, including a computer program, which, when executed by a processor, implements the cloud-based distributed cache database adaptive tuning method as described in the first aspect and various possible designs of the first aspect.

[0168] The above description is only a preferred embodiment of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present disclosure (but not limited to) by each other to form a technical solution.

[0169] In addition, although each operation is described in a specific order, this should not be understood as requiring these operations to be performed in the specific order shown or in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

[0170] Although the subject matter has been described in language specific to structural features and / or methodological logical actions, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely example forms of implementing the claims.

Claims

1. A cloud-based distributed cache database adaptive tuning method, characterized in that: include: In response to a tuning request for a cloud-based distributed cache database, obtaining an expected value of at least one business indicator carried in the tuning request; Determine a tuning strategy corresponding to the Yunji distributed cache database according to the expected value of the at least one business indicator and the network causal model corresponding to the Yunji distributed cache database, wherein the tuning strategy includes at least one target configuration item and adjustment parameters of each target configuration item, wherein the network causal model is used to predict and analyze the configuration parameters of the configuration items based on the causal relationship between the configuration items and the business indicator variables constructed based on multiple types of data analysis; According to the tuning strategy corresponding to the cloud-based distributed cache database, the configuration parameters of each target configuration item of the cloud-based distributed cache database are adjusted to their corresponding adjustment parameters.

2. The method according to claim 1, characterized in that The network causal model includes a multi-layer heterogeneous causal model constructed based on multiple types of data, wherein different causal models are constructed and generated based on different types of data; wherein the multiple types of data include multiple types of verification data, cloud observation data, and empirical data.

3. The method according to claim 1, characterized in that The network causal model includes a three-layer heterogeneous causal model constructed based on multiple types of data; wherein, the first layer causal model of the three-layer heterogeneous causal model is constructed based on verification data in multiple types of data, and is used to determine the causal relationship between configuration items and operating indicators, the second layer causal model of the three-layer heterogeneous causal model is constructed based on cloud observation data in multiple types of data, and is used to determine the causal relationship between operating indicators, and the third layer causal model of the three-layer heterogeneous causal model is constructed based on cloud observation data and experience data in multiple types of data, and is used to determine the causal relationship between operating indicators and business indicators.

4. The method according to any one of claims 1 to 3, characterized in that Determining the tuning strategy corresponding to the cloud-based distributed cache database according to the expected value of the at least one business indicator and the network causal model corresponding to the cloud-based distributed cache database includes: According to the calling order of the multi-layer heterogeneous causal models from the bottom layer to the top layer in the network causal model, the causal models of each layer are called in turn, and the output of the causal model of the current layer is used as the input of the causal model of the upper layer connected to it, and at least one target configuration item output by the top-level causal model and the adjustment parameters of each target configuration item are obtained as the tuning strategy corresponding to the Yunji distributed cache database.

5. The method according to any one of claims 1 to 3, characterized in that The network causal model includes a three-layer heterogeneous causal model that communicates layer by layer from the bottom layer to the top layer, wherein the first layer causal model includes a potential causal model, the second layer causal model includes a Bayesian network causal model, and the third layer causal model includes a structural equation model.

6. The method according to claim 1, characterized in that Before the step of responding to the tuning request for the cloud-based distributed cache database and obtaining the expected value of at least one business indicator carried in the tuning request, the method further includes: Acquire a monitoring indicator data set of the cloud distributed cache database, wherein the monitoring indicator data set includes verification data participating in the cloud distributed cache database system test and cloud observation data and experience data during the operation of the cloud distributed cache database; The network causal model is generated according to the monitoring indicator data set.

7. The method according to claim 6, characterized in that After adjusting each target configuration item of the cloud-based distributed cache database to a corresponding adjustment parameter according to the tuning strategy corresponding to the cloud-based distributed cache database, the method further includes: If the empirical data corresponding to the business indicator does not meet the expected value of the business indicator, updating the monitoring indicator data set; Generate a new network causal model through the updated monitoring indicator dataset.

8. The method according to claim 6, characterized in that The cloud observation data includes a plurality of operation indicators, and the experience data includes a plurality of business indicators; accordingly, generating the network causal model according to the monitoring indicator data set includes: Generate a first-layer causal model corresponding to the cloud-interconnected distributed cache database according to the intervention effect of the test parameters of the configuration items in each verification data and the test value of at least one operation indicator to be verified, and generate a second-layer causal model corresponding to the cloud-interconnected distributed cache database according to the dependency relationship and the subordinate relationship between the multiple operation indicators, and generate a third-layer causal model corresponding to the cloud-interconnected distributed cache database according to the multiple business indicators, the multiple operation indicators and the preset functional relationship; The first-layer causal model, the second-layer causal model and the third-layer causal model are integrated to generate the network causal model.

9. The method according to claim 8, characterized in that The fusing the first-layer causal model, the second-layer causal model and the third-layer causal model to generate the network causal model includes: The first layer causal model and the second layer causal model are combined with a plurality of third operation indicators as a combination point, and the second layer causal model and the third layer causal model are combined with a plurality of fourth operation indicators as a combination point to generate the network causal model; Among them, the third operating indicator is an operating indicator causally related to the test parameters of the configuration item determined based on the first-level causal model; the third operating indicator is an operating indicator causally related to the expected value of the business indicator determined based on the third-level causal model.

10. The method according to claim 8, characterized in that The first-level causal model includes a potential causal model; Accordingly, generating the first-layer causal model corresponding to the cloud-based distributed cache database according to the intervention effect of the test parameters of the configuration items in each verification data and the test value of at least one operation indicator to be verified includes: For each test parameter of a configuration item in the verification data and a test value of at least one operating indicator to be verified, determining an intervention effect of the test parameter of the configuration item on the test value of the at least one operating indicator to be verified; According to the intervention effect, the corresponding relationship between at least one configuration item, the adjustment parameter of each configuration item and at least one operating indicator, the value range of each operating indicator is determined, and a potential causal model corresponding to the cloud-based distributed cache database is generated.

11. The method according to claim 8, characterized in that The second-layer causal model includes a Bayesian network causal model; Accordingly, generating a second-layer causal model corresponding to the cloud-based distributed cache database according to the dependencies and subordinate relationships among the multiple operating indicators includes: Determining, according to an application performance management and configuration management database, dependencies and subordination relationships between a plurality of operating indicators corresponding to a target event, wherein the target event includes a scheduling event or an upgrade event; According to the dependency relationship, subordination relationship and conditional probability table among the multiple operating indicators corresponding to the target event, a Bayesian network causal model corresponding to the Yunji distributed cache database is generated.

12. The method according to claim 8, characterized in that The third level causal model includes a structural equation model; Accordingly, generating the third-layer causal model corresponding to the cloud-based distributed cache database according to the multiple business indicators, the multiple operation indicators and the preset functional relationship includes: For each business indicator, construct a first functional relationship between a value of the business indicator and a value range of at least one first target operation indicator; Determine at least one second target operating indicator linearly related to the first target operating indicator from the plurality of operating indicators, and determine a second functional relationship between the first target operating indicator and the second target operating indicator; According to the first functional relationship and the second functional relationship, an association relationship between the value range of the at least one second target operating indicator and the value of the business indicator is established to generate a structural equation model corresponding to the cloud-based distributed cache database.

13. A cloud-based distributed cache database adaptive tuning system, characterized in that: include: An acquisition unit, configured to respond to a tuning request for the cloud-based distributed cache database and acquire an expected value of at least one business indicator carried in the tuning request; A determination unit, configured to determine a tuning strategy corresponding to the cloud-based distributed cache database according to the expected value of the at least one business indicator and the network causal model corresponding to the cloud-based distributed cache database, wherein the tuning strategy includes at least one target configuration item and an adjustment parameter for each target configuration item, wherein the network causal model is used to predict and analyze the configuration parameters of the configuration item based on the causal relationship from the configuration item to the business indicator variable constructed based on multiple types of data analysis; The adjustment unit is used to adjust the configuration parameters of each target configuration item of the cloud-based distributed cache database to their corresponding adjustment parameters according to the tuning strategy corresponding to the cloud-based distributed cache database.

14. An electronic device, characterized in that: include: Processor and memory; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory, so that the processor executes the cloud-based distributed cache database adaptive tuning method as described in any one of claims 1 to 12.

15. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions. When the processor executes the computer-executable instructions, the cloud-based distributed cache database adaptive tuning method as described in any one of claims 1 to 12 is implemented.

16. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the cloud-based distributed cache database adaptive tuning method as described in any one of claims 1 to 12 is implemented.

Citation Information

Patent Citations

  • Database parameter adjusting method and device

    CN114546990A

  • Parameter optimization method and device for production process

    CN116050607A

  • Defect identification method based on causal relationship network

    CN116881777A

  • Carbon neutralization steel production process control method based on biomass energy conversion

    CN118095650A

  • Methods and systems for controlled modeling and optimization of a natural language database interface

    WO2022174356A1

Cited By

  • Multi-source data fusion system for synchronously monitoring density and micro water of SF6

    CN120873984A