Server cluster capacity evaluation method and device, electronic equipment and storage medium
1. A server cluster capacity evaluation method is characterized by comprising the following steps:
acquiring the service per second request quantity, the service average response time and the server central processing unit average load of the current server cluster to be evaluated;
estimating the peak value of the request quantity per second of service of the current server cluster to be evaluated according to the real-time function relationship between the average load of the central processing unit of the server and the request quantity per second of service;
and obtaining the current capacity evaluation result of the server cluster to be evaluated according to a preset server cluster capacity evaluation logic based on the request amount per second of service, the peak value of the request amount per second of service, the average response time of service, a preset average response time threshold value of service and the average load of the central processing unit of the server.
2. The method according to claim 1, wherein the estimating a peak per second service request amount of the current server cluster to be evaluated according to a real-time functional relationship between an average load of a central processing unit of the server and the per second service request amount comprises:
acquiring average load data and service per second request data of a central processing unit of a server in a first historical time period before the current time;
establishing a fitting relation between the average load data of the central processing unit of the server and the request data of the service per second, and taking the fitting relation as the real-time function relation;
and based on the real-time function relationship, taking the service request quantity per second when the average load value of the central processing unit of the server is a preset upper limit load value as the peak value of the service request quantity per second.
3. The method of claim 1, further comprising:
acquiring service average response time data in a second historical time period before the current time, and performing statistical analysis to obtain distribution characteristics of the service average response time data;
and determining the preset service average response time threshold according to the distribution characteristics.
4. The method according to claim 1, wherein the obtaining a current capacity evaluation result of the server cluster to be evaluated according to a preset server cluster capacity evaluation logic based on the request amount per second of service, the peak request amount per second of service, the average response time of service, a preset average response time threshold value of service, and the average load of the central processing unit of the server comprises:
calculating the ratio of the service request quantity per second to the service request quantity per second peak value to obtain a first service capacity evaluation parameter;
calculating the ratio of the average service response time to the preset average service response time threshold to obtain a second service capacity evaluation parameter;
and respectively calculating the products of the first service capacity evaluation parameter, the second service capacity evaluation parameter and the average load of the service central processing unit and the corresponding parameter weight coefficients, and taking the sum of the obtained product results as the current capacity evaluation result of the server cluster to be evaluated.
5. The method according to claim 4, wherein the determining of the parameter weighting coefficients corresponding to the first service capacity assessment parameter, the second service capacity assessment parameter and the average load of the service central processing unit respectively comprises:
acquiring a preset quantity group parameter weight coefficient value;
evaluating the capacity of the server cluster according to the server cluster capacity evaluation logics under the weight coefficient values of each group of parameters;
and taking the evaluation results of the server cluster capacity evaluation logics under the parameter weight coefficient values of all groups as comparison groups within a preset time period, performing regression testing, and determining the final parameter weight coefficient value according to the regression testing results.
6. The method of claim 1, further comprising:
inputting the service request quantity per second data in a third history time period before the current time into a preset request quantity prediction model to obtain a service request quantity per second result at the preset time;
and calculating the difference value between the peak value of the request quantity per second of the service and the request quantity per second of the service at the preset moment, and adjusting the server resources of the server cluster to be evaluated according to the difference value.
7. The method of claim 1, wherein before obtaining the requests per second for service, the average response time for service, and the average load of the central processing units of the servers of the current cluster to be evaluated, the method further comprises:
acquiring server cluster operation data from the server cluster operation monitoring data and storing the server cluster operation data in a preset database;
extracting preset field data from the preset database, and preprocessing the extracted data to obtain a server cluster capacity evaluation data set, wherein the preset field comprises data acquisition time, service request amount per second, service average response time and service central processing unit average load;
correspondingly, the obtaining of the service request amount per second, the service average response time and the server central processing unit average load of the current server cluster to be evaluated includes:
and in the server cluster capacity evaluation data set, acquiring the service per second request quantity, the service average response time and the server central processing unit average load of the current server cluster to be evaluated.
8. A server cluster capacity evaluation apparatus, comprising:
the data acquisition module is used for acquiring the service request quantity per second, the service average response time and the server central processing unit average load of the current server cluster to be evaluated;
the peak value estimation module is used for estimating the peak value of the request quantity per second of service of the current server cluster to be evaluated according to the real-time function relationship between the average load of the central processing unit of the server and the request quantity per second of service;
and the capacity evaluation module is used for obtaining the current capacity evaluation result of the server cluster to be evaluated according to a preset server cluster capacity evaluation logic based on the request amount per second of service, the request amount peak value per second of service, the average response time of service, a preset average response time threshold value and the average load of the central processing unit of the server.
9. An electronic device, characterized in that the electronic device comprises:
one or more processors;
a storage device for storing one or more programs,
when executed by the one or more processors, cause the one or more processors to implement the server cluster capacity assessment method of any of claims 1-7.
10. A storage medium containing computer-executable instructions for performing the server cluster capacity assessment method of any one of claims 1-7 when executed by a computer processor.
Background
In internet services, a service architecture is expanded according to service growth, and meanwhile, in order to ensure service stability, capacity of a server cluster needs to be managed. Currently, in order to predict service traffic pressure at a node in a major activity or a special time and evaluate the capacity of a server cluster to guide resource adjustment such as service expansion or contraction, the capacity and performance of the server cluster are usually evaluated through a server cluster pressure test.
However, the operation cost for obtaining the server cluster capacity evaluation by the pressure test is high, the test process is easy to influence the service performance, an online fault is caused, and the timeliness is low.
Disclosure of Invention
The embodiment of the disclosure provides a server cluster capacity evaluation method and device, electronic equipment and a storage medium, which can efficiently evaluate the capacity of a server cluster in real time.
In a first aspect, an embodiment of the present disclosure provides a server cluster capacity evaluation method, including:
acquiring the service per second request quantity, the service average response time and the server central processing unit average load of the current server cluster to be evaluated;
estimating the peak value of the request quantity per second of service of the current server cluster to be evaluated according to the real-time function relationship between the average load of the central processing unit of the server and the request quantity per second of service;
and obtaining the current capacity evaluation result of the server cluster to be evaluated according to a preset server cluster capacity evaluation logic based on the request amount per second of service, the peak value of the request amount per second of service, the average response time of service, a preset average response time threshold value of service and the average load of the central processing unit of the server.
In a second aspect, an embodiment of the present disclosure further provides a server cluster capacity evaluation apparatus, including:
the data acquisition module is used for acquiring the service request quantity per second, the service average response time and the server central processing unit average load of the current server cluster to be evaluated;
the peak value estimation module is used for estimating the peak value of the request quantity per second of service of the current server cluster to be evaluated according to the real-time function relationship between the average load of the central processing unit of the server and the request quantity per second of service;
and the capacity evaluation module is used for obtaining the current capacity evaluation result of the server cluster to be evaluated according to a preset server cluster capacity evaluation logic based on the request amount per second of service, the request amount peak value per second of service, the average response time of service, a preset average response time threshold value and the average load of the central processing unit of the server.
In a third aspect, an embodiment of the present disclosure further provides an electronic device, where the electronic device includes:
one or more processors;
a storage device for storing one or more programs,
when executed by the one or more processors, cause the one or more processors to implement a server cluster capacity assessment method as in any of the embodiments of the present disclosure.
In a fourth aspect, the embodiments of the present disclosure also provide a storage medium containing computer-executable instructions, which when executed by a computer processor, are used to perform the server cluster capacity assessment method according to any one of the embodiments of the present disclosure.
According to the technical scheme of the embodiment of the disclosure, the service per second request quantity, the service average response time and the server central processing unit average load of the current server cluster to be evaluated are obtained in real time; then, estimating the peak value of the request quantity per second of service of the current server cluster to be evaluated according to the real-time function relation between the average load of the central processing unit of the server and the request quantity per second of service; finally, based on service capacity evaluation parameters such as the service request amount per second, the service request amount per second peak value, the service average response time, the preset service average response time threshold and the server central processing unit average load, the current capacity evaluation result of the server cluster to be evaluated is obtained according to the preset server cluster capacity evaluation logic, the problems that the evaluation cost of the prior art center on the server cluster performance is high and the timeliness is low are solved, the server cluster capacity can be efficiently and dynamically evaluated in real time, the evaluation cost on the server cluster performance is reduced, and the server resource management is efficiently guided.
Drawings
The above and other features, advantages and aspects of various embodiments of the present disclosure will become more apparent by referring to the following detailed description when taken in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numbers refer to the same or similar elements. It should be understood that the drawings are schematic and that elements and features are not necessarily drawn to scale.
Fig. 1 is a flowchart of a method for evaluating capacity of a server cluster according to a first embodiment of the present disclosure;
FIG. 2 is a dynamic graph of data for a requested amount of data per second serviced over a specified historical period of time provided by a first embodiment of the present disclosure;
FIG. 3 is a dynamic graph of average load data of a central processing unit of a server over a specified historical period of time, according to an embodiment of the present disclosure;
FIG. 4 is a diagram illustrating a linear fit result between request amount per second data and average load data of a central processing unit of a server served in a simultaneous period of time according to an embodiment of the present disclosure;
fig. 5 is a flowchart of a server cluster capacity evaluation method provided in the second embodiment of the present disclosure;
fig. 6 is a schematic flowchart of a method for evaluating capacity of a server cluster according to a third embodiment of the present disclosure;
fig. 7 is a schematic structural diagram of a server cluster capacity evaluation apparatus according to a fourth embodiment of the present disclosure;
fig. 8 is a schematic structural diagram of an electronic device according to a fifth embodiment of the disclosure.
Detailed Description
Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While certain embodiments of the present disclosure are shown in the drawings, it is to be understood that the present disclosure may be embodied in various forms and should not be construed as limited to the embodiments set forth herein, but rather are provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the disclosure are for illustration purposes only and are not intended to limit the scope of the disclosure.
It should be understood that the various steps recited in the method embodiments of the present disclosure may be performed in a different order, and/or performed in parallel. Moreover, method embodiments may include additional steps and/or omit performing the illustrated steps. The scope of the present disclosure is not limited in this respect.
The term "include" and variations thereof as used herein are open-ended, i.e., "including but not limited to". The term "based on" is "based, at least in part, on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Relevant definitions for other terms will be given in the following description.
It should be noted that the terms "first", "second", and the like in the present disclosure are only used for distinguishing different devices, modules or units, and are not used for limiting the order or interdependence relationship of the functions performed by the devices, modules or units.
It is noted that references to "a", "an", and "the" modifications in this disclosure are intended to be illustrative rather than limiting, and that those skilled in the art will recognize that "one or more" may be used unless the context clearly dictates otherwise.
Example one
Fig. 1 is a schematic flow chart of a method for evaluating capacity of a server cluster according to a first embodiment of the present disclosure, which is applicable to a situation of evaluating an operation state of a server cluster, and is particularly applicable to a situation of requiring service resource management according to the operation state of the server cluster. The method may be performed by a server cluster capacity evaluation apparatus, which may be implemented in software and/or hardware, and may be configured in an electronic device, such as a mobile terminal or a server device.
As shown in fig. 1, the method for evaluating the capacity of a server cluster provided in this embodiment includes:
s110, obtaining the service request quantity per second, the service average response time and the server central processing unit average load of the current server cluster to be evaluated.
The server cluster refers to a plurality of servers processing the same service, and the cluster can use a plurality of computers to perform parallel computation so as to obtain high computation speed, and can also use a plurality of computers to perform backup. For a client, a service cluster is equivalent to only one server. Generally, the performance of a server cluster may be evaluated differently, including parameters such as request per second (QPS), average response time of service (COST), and average load of server central processing units (CPU _ rate), especially for CPU-intensive services.
The average response time of the service reflects the quality of the service to a certain extent, and in order to ensure the quality of the service, a user has good service experience and sets an upper limit value of the average response time of the service. The Processing capacity of the service cluster is shown when the requests per second are served, the larger the requests per second can be processed by the server cluster, the better, but after the requests per second reach a certain level, the average corresponding time of the services will become longer, and the load of a Central Processing Unit (CPU) of the server is too large. That is, if the service request amount is too large, which results in that the service response time consumption reaches a certain level, it is determined that the server cluster cannot work with normal performance, or when the CPU load reaches a certain level, the performance of the entire service cannot work with normal performance due to the limitation of the hardware bottleneck, and under the condition that the cluster size is fixed, the request amount that the service can carry has an upper limit. Therefore, the capacity of the service cluster is evaluated by combining the factors of the request amount per second of the service, the average response time of the service and the average load of the central processing unit of the server.
The acquisition of data such as the service request amount per second, the service average response time, the server central processing unit average load and the like can be directly read from the server cluster operation monitoring data, or after the data to be acquired is integrated through a data processing function, the corresponding data is read in the integrated data set.
And S120, estimating the peak value of the request quantity per second of the service of the current server cluster to be evaluated according to the real-time function relationship between the average load of the central processing unit of the server and the request quantity per second of the service.
For a server cluster, the performance limit of the server cluster is usually obtained by a pressure test method, but with time change, optimization adjustment of service logic and framework change, the limit performance of the cluster cannot be obtained dynamically in real time by the pressure test, and the pressure test is long in operation time and high in risk.
In this embodiment, a functional relationship between the average load of the central processing unit of the server and the requested data per second of the service is established, and the functional relationship between the central processing unit of the server and the requested data per second of the service can be updated according to the dynamically collected new data. And estimating the peak value of the request volume per second of the service of the current server cluster to be evaluated according to the linear relation established between the average load of the central processing unit of the server and the request volume per second of the service.
Specifically, referring to fig. 2 and fig. 3, a graph of data of requests per second of a server cluster and a graph of data of average load of central processing units (i.e., CPU utilization) of servers in the same time period are captured. It can be observed from the graph that the trend of the average load data of the central processing unit of the server is consistent with the trend of the request data per second, that is, there is an obvious association relationship between the QPS and the CPU _ rate, which can be described as that, in a certain time range, the CPU _ rate increases with the increase of the QPS, and the overall trends of the two keep consistent. Therefore, in this embodiment, curve fitting is performed through the average load of the central processing unit of the server and the request amount data per second of service, and a fitting relationship, that is, a functional relationship, between the two is determined.
In the data fitting process, the server central processor average load data and the requested amount of data per second of the service in the first history time period before the current time can be obtained for two relationship fitting. The first historical time period may be seven days, 14 days or other time lengths before the current time, and may be determined comprehensively according to data calculation amount, calculation efficiency and other indexes. In a specific example, the data in fig. 2 and 3 are taken as an example for fitting calculation, and there are 288 sets of raw data, wherein the training data for linear fitting includes 230 sets of data, and there are 58 sets of data for fitting result test, and the fitting result can refer to the best fit line (best line) shown in fig. 4, in which the abscissa is CPU-rate and the ordinate is QPS. The best fit line has an intercept of 40845.11716930033, a regression coefficient of 2554071.37846435, and an evaluation model score of 0.9855894892189563. By this best fit line, the QPS is 807067.0 when the CPU-rate is evaluated to be 30%; the QPS was 1062474.0 when the CPU-rate was evaluated to be 40%; the QPS was 1317881.0 when the CPU-rate was evaluated to be 50%; when the CPU-rate was evaluated to be 60%, the QPS was 1573288.0. Therefore, the request volume per second of service when the average load value of the central processing unit of the server is the preset upper limit load value can be used as the peak value of the request volume per second of service according to the fitting relation. The predetermined upper limit load value is typically 70% or other value. If a processor such as a GPU is provided in the server, the preset upper limit load value may also be a higher value adaptively.
And S130, obtaining a current capacity evaluation result of the server cluster to be evaluated according to a preset server cluster capacity evaluation logic based on the request amount per second of service, the request amount peak value per second of service, the average response time of service, a preset average response time threshold value of service and the average load of the central processing unit of the server.
Specifically, the service cluster capacity evaluation process may be that, first, a ratio of a service request amount per second to a service request amount per second peak is calculated to obtain a first service capacity evaluation parameter; then, calculating the ratio of the average service response time to a preset average service response time threshold value to obtain a second service capacity evaluation parameter; and finally, respectively calculating the products of the first service capacity evaluation parameter, the second service capacity evaluation parameter, the average load of the service central processing unit and the corresponding parameter weight coefficient, and taking the sum of the obtained product results as the current capacity evaluation result of the server cluster to be evaluated. Expressed by a mathematical formula, the server cluster capacity usage water level value at time t can be expressed as: wherein, QPStThe request per second of service, QPS, denoted as time tpeakIndicating the peak per second request volume for service at time t, COSTtMean response time of service, COST, at time tthreadRepresents a preset service average response time threshold, cputThe average load of the central processing unit of the server at the time t is represented by A, B and C, which respectively represent the first service capacity evaluation parameter, the second service capacity evaluation parameter, the average load of the central processing unit of the server and the corresponding parameter weight coefficient, LeveltThe value range of (A) is 0-100%.
Further, the values of A, B and C are set and selected based on empirical values. The degrees of influence of the request amount per second of service, the average time consumption of service response and the average load of a central processor of the server on the overload capacity of the server cluster are different. Specifically, for CPU-intensive services, the following situations exist:
under the condition of not considering the cache, the QPS value rises, and the CPU-rate value theoretically rises; QPS value rises, CPU-rate value rises within a reasonable range, and COST value does not necessarily rise; when an exception (such as thread blocking/accumulation) occurs at the downstream of the service cluster, the QPS value does not rise, the COST value rises, and the CPU-rate value rises; an exception occurs downstream of the service cluster, but the thread is not blocked, the QPS value does not rise, the COST value rises, and the CPU-rate value does not rise.
Under normal conditions, the abnormal rising of the CPU-rate value is generally 2-3 times higher than that of the normal state, for example, the normal response time is 50ms, and the abnormal time can be 300 ms; secondly, temporary abnormal fluctuation of the QPS value can also occur, and in the initial adjustment process of the weight coefficient, the weight of the factor with large fluctuation can be adjusted to be low, so that abnormal interference is reduced as far as possible. Based on the assumptions, a plurality of parameter weight coefficient groups can be set, the parameter weight coefficient groups are mutually compared in the actual on-line regression process, and the final coefficient is determined according to the fitting condition after running for a period of time. The server cluster capacity is evaluated according to the server cluster capacity evaluation logics under the weight coefficient values of all the groups of parameters; and taking the evaluation results of the server cluster capacity evaluation logics under the parameter weight coefficient values of all groups as comparison groups within a preset time period, performing regression testing, and determining the final parameter weight coefficient value according to the regression testing results. Illustratively, the values of the parameter weighting factors may be set as follows, a, B, C [ [0.25,0.25,0.5], [0.3,0.2,0.5], [0.3,0.15,0.55], [0.4,0.1,0.5], [0.3,0.1,0.6], [0.6,0.1,0.3], [0.7,0.1,0.2] ].
In a preferred embodiment, the preset service average response time threshold may be adjusted in stages, and the distribution characteristics of the service average response time data may be obtained by obtaining the service average response time data in a second historical time period before the current time, and performing statistical analysis; and then, determining a new preset service average response time threshold according to the distribution characteristics, and taking the newly determined numerical value as the current preset service average response time threshold. For example, if the service mean response time is determined to conform to the normal distribution theory according to the threshold calculation theory of the service mean response time, and the historical service mean response time monitoring data of a period of time (for example, 7 days of the current time) is substituted for calculation and updated according to the day level, then the preset service mean response time threshold is equal to the sum of the arithmetic mean of the service mean response times in the statistical period of time and three times the standard deviation of the service mean response time.
Of course, the preset response timeout threshold of the service may also be used as the preset average response time threshold of the service. However, considering that the response timeout thresholds defined by various services are not uniform and have subjective deviation, the service limit performance cannot be reflected, or the time consumption threshold for calculating the average response of the service is better in the above preferred embodiment.
According to the technical scheme of the embodiment of the disclosure, the service per second request quantity, the service average response time and the server central processing unit average load of the current server cluster to be evaluated can be obtained in real time; then, estimating the peak value of the request quantity per second of service of the current server cluster to be evaluated according to the real-time function relation between the average load of the central processing unit of the server and the request quantity per second of service; finally, based on service capacity evaluation parameters such as the service request amount per second, the service request amount per second peak value, the service average response time, the preset service average response time threshold and the server central processing unit average load, the current capacity evaluation result of the server cluster to be evaluated is obtained according to the preset server cluster capacity evaluation logic, the problems that the evaluation cost of the prior art center on the server cluster performance is high and the timeliness is low are solved, the server cluster capacity can be efficiently and dynamically evaluated in real time, the evaluation cost on the server cluster performance is reduced, and the server resource management is efficiently guided. In the whole server cluster capacity evaluation process, relevant data are obtained in real time to carry out capacity evaluation, the request pressure of the server cluster can be sensed in an hour level or even a minute level, when the water level value of the server cluster capacity is higher than a warning value, an alarm can be given in time to inform capacity risks, and the method is fully automatic and does not need human intervention. Compared with the situation of evaluating the load risk of the service cluster through empirical analysis, the method has the advantages that the data are used as the evaluation basis, the accuracy is high, and the evaluation results are all under a unified standard.
Example two
The embodiments of the present disclosure and various alternatives in the server cluster capacity evaluation method provided in the above embodiments may be combined. The method for evaluating the capacity of the server cluster provided in this embodiment further describes a process of predicting the service request amount at a future time and adjusting the server cluster resources according to the prediction result and the capacity evaluation result.
Fig. 5 is a schematic flow chart of a server cluster capacity evaluation method provided in the second embodiment of the present disclosure. As shown in fig. 5, the method for evaluating the capacity of a server cluster provided in this embodiment includes:
s210, inputting the service request quantity per second data in the third history time period before the current time into a preset request quantity prediction model to obtain the result of the service request quantity per second at the preset time.
Specifically, the preset request amount prediction model may be a Prophet prediction model, which is a time sequence prediction model. Specifically, the model formula may be expressed as y (t) ═ g (t) + s (t) + h (t) + epsilon (t). Where g (t) belongs to a trend term, s (t) belongs to a seasonal term, h (t) belongs to a holiday term, and ε (t), an error term or residual term, represents the unpredicted fluctuation of the model. When the service per second request quantity at any future preset time is predicted, the time stamp of the known time sequence, the corresponding service per second request quantity value and the length of the time sequence to be predicted need to be input into a prediction model; the predictive model will then output future time series trends based on the input data.
For example, assuming that the request amount per second monitoring data of 24 months of the past is used as a training sample in 12 months and 13 days of 2020, model training is performed to predict the peak value of the request amount per second of service for 7 days in the future. In the last two years, the date of each day and the corresponding service request amount per second of each day are input into the prediction model, and the length of the time series needing to be predicted is 7 days, so that the output result shown in table 1 is obtained.
TABLE 1 prediction of requests per second for 7 days of future service on a day scale
Where ds denotes a date, yhat denotes a predicted value in a corresponding time series, yhat _ lower denotes a lower bound of the predicted value, and yhat _ upper denotes an upper bound of the predicted value.
For another example, in 12/22/2020, model training is performed with hourly service hourly demand monitoring data of 14 historical days as training samples to predict the peak value of the service hourly demand and the trend of the demand in the future 6 hours. In the last 14 days, the service request amount per second in each hour of each day is input into the prediction model, and the length of the time series needing to be predicted is 6 hours, so that the output result shown in table 2 is obtained.
TABLE 2 prediction of future 6-hour service requests per second on an hourly scale
Through the prediction of the service request quantity per second, the service pressure of the service cluster at a certain time in the future can be known, so that the capacity of the service cluster can be adjusted in time, and the service provided by the server cluster is prevented from being out of order.
S220, in the dynamic server cluster capacity evaluation result, a service request quantity per second peak value of the server cluster to be evaluated at the service request quantity per second reference time is preset.
In order to further guide the adjustment of the server cluster resources, the peak value of the request amount per second of the service at a reference time can be selected as a reference to determine whether to perform the server resource adjustment. For example, at the time of the fifth afternoon of a certain day of a certain month of a year, the requested amount per second of the service at 8 pm today is predicted by step S210, and can be regarded as the peak requested amount per second of the service at 8 pm today. The estimated peak per second requests for service that a server cluster can withstand yesterday, the previous day, or other historical times can be used as a reference to determine whether the server cluster can take over the predicted per second requests for service at 8 pm. Alternatively, the requested amount per second of service at a time near eight pm may be used as a reference. Assuming now 6 o' clock and 10 min later today, the peak value of the estimated service requests per second corresponding to the current time can be used as a reference.
And S230, calculating a difference value between the peak value of the request quantity per second of the service and the request quantity per second of the service at the preset moment, and adjusting the server resources of the server cluster to be evaluated according to the difference value.
Specifically, in the running process of the server cluster, a load balancing algorithm is adopted, so that each server can equally bear service requests, and when the server resources are adjusted, the upper limit threshold of the service requests per second that each server can bear can be determined through the peak value of the service requests per second as a reference and the number of existing servers in the server cluster to be evaluated. Then, the number of servers required is determined by the predicted service request amount per second and the upper threshold of the service request amount per second that each server can support. If the required number of the servers is larger than the existing number of the servers, the capacity of the server cluster to be evaluated needs to be expanded, and if the required number of the servers is smaller than the existing number of the servers, the capacity of the server cluster to be evaluated can be reduced or no processing is carried out, so that the server resource adjustment of the server cluster is completed.
According to the technical scheme of the embodiment of the disclosure, the server cluster real-time capacity risk assessment and the capacity expansion and contraction guidance can be realized by predicting the service request quantity per second at a preset time in the future, comparing the predicted service request quantity per second with the estimated service request quantity per second peak value corresponding to the set reference time, and further adjusting the server resources in time according to the comparison result. Particularly, the service request amount and the service cluster capacity are estimated by the time node in major activities or special holidays, data guidance can be provided in the aspects of service maintenance and resource management, normal service resource structure optimization can be performed daily, and redundant resource diving guidance is provided. The method and the system solve the problems of high cost and low timeliness of the prior art center for evaluating the performance of the server cluster, can efficiently and dynamically evaluate the capacity of the server cluster in real time, reduce the cost for evaluating the performance of the server cluster, and efficiently guide the resource management of the server.
EXAMPLE III
The embodiments of the present disclosure and various alternatives in the server cluster capacity evaluation method provided in the above embodiments may be combined. The server cluster capacity evaluation method provided by this embodiment further illustrates a specific process of storing and managing server cluster operation data, and realizes an overall operation scheme for acquiring server capacity evaluation from data.
Fig. 6 is a schematic flow chart of a server cluster capacity evaluation method provided in the third embodiment of the present disclosure. As shown in fig. 6, the method for evaluating the capacity of a server cluster provided in this embodiment includes:
s310, obtaining server cluster operation data from the server cluster operation monitoring data and storing the server cluster operation data in a preset database.
Specifically, in the process of server operation, server cluster operation data can be monitored through Metrics, and the server cluster operation data is stored in a preset database. In the preset database, the stored operation data is all parameters that can be monitored by the service cluster in the operation process, such as data of service content, service request amount, service average response time, number of servers in the server cluster, and CPU-rate.
S320, extracting preset field data from the preset database, and preprocessing the extracted data to obtain a server cluster capacity evaluation data set.
In this step, it is equivalent to establish a personalized database on the basis of the preset database, for storing the parameters read from the preset database under each field. Illustratively, the data fields read include those shown in Table 3:
TABLE 3 read data field schematic
Further, after the data is read, the data needs to be preprocessed, for example, data formats are converted, formats of the data are unified into a format used in the server cluster capacity evaluation process, and repeated or invalid data are deleted.
S330, in the server cluster capacity evaluation data set, acquiring the service per second request quantity, the service average response time and the server central processing unit average load of the current server cluster to be evaluated.
And S340, estimating the peak value of the request quantity per second of service of the current server cluster to be evaluated according to the real-time function relationship between the average load of the central processing unit of the server and the request quantity per second of service.
And S350, obtaining a current capacity evaluation result of the server cluster to be evaluated according to a preset server cluster capacity evaluation logic based on the request amount per second of service, the request amount per second of service peak value, the average response time of service, a preset average response time threshold value and the average load of the central processing unit of the server.
The steps S330 to S350 refer to the related contents in the first embodiment and the second embodiment.
In one embodiment, a complete closed loop of resource evaluation, resource redundancy reclamation, and resource reuse may also be implemented. Specifically, the service capacity of a large number of normal redundant resources is measured, and the redundant resources are recycled as mobile resources under the condition of keeping a certain margin. Meanwhile, the services with resource gaps are backfilled by combining with real-time capacity evaluation, so that resource recycling is realized, the utilization rate of the whole resources is improved, the fragmentation of the resources is reduced, and the problem of insufficient normal resources is solved.
According to the technical scheme of the embodiment of the disclosure, the operation data of the server cluster are monitored in real time, the operation data are extracted and processed statically, and then the capacity of the server cluster is evaluated based on the processed data, so that a whole set of solution for data acquisition, data storage management, query display, evaluation logic establishment, model evaluation and estimation result output is provided. The method and the system solve the problems of high cost and low timeliness of the prior art center for evaluating the performance of the server cluster, can efficiently and dynamically evaluate the capacity of the server cluster in real time, reduce the cost for evaluating the performance of the server cluster, and efficiently guide the resource management of the server.
Example four
Fig. 7 is a schematic structural diagram of a server cluster capacity evaluation apparatus according to a fourth embodiment of the present disclosure. The server cluster capacity evaluation device provided by the embodiment is suitable for evaluating the running state of the server cluster.
As shown in fig. 7, the server cluster capacity evaluation apparatus includes: a data acquisition module 410, a peak estimation module 420, and a capacity estimation module 430.
The data acquisition module 410 is configured to acquire a service request per second, an average service response time, and an average load of a central processing unit of a server of a current server cluster to be evaluated; the peak value estimation module 420 is configured to estimate a peak value of the request amount per second of service of the current server cluster to be evaluated according to a real-time functional relationship between an average load of a central processing unit of the server and the request amount per second of service; and the capacity evaluation module 430 obtains a current capacity evaluation result of the server cluster to be evaluated according to a preset server cluster capacity evaluation logic based on the request amount per second of service, the request amount per second of service peak value, the average response time of service, a preset average response time threshold value and the average load of the central processing unit of the server.
According to the technical scheme of the embodiment of the disclosure, the service per second request quantity, the service average response time and the server central processing unit average load of the current server cluster to be evaluated are obtained in real time; then, estimating the peak value of the request quantity per second of service of the current server cluster to be evaluated according to the real-time function relation between the average load of the central processing unit of the server and the request quantity per second of service; finally, based on service capacity evaluation parameters such as the service request amount per second, the service request amount per second peak value, the service average response time, the preset service average response time threshold and the server central processing unit average load, the current capacity evaluation result of the server cluster to be evaluated is obtained according to the preset server cluster capacity evaluation logic, the problems that the evaluation cost of the prior art center on the server cluster performance is high and the timeliness is low are solved, the server cluster capacity can be efficiently and dynamically evaluated in real time, the evaluation cost on the server cluster performance is reduced, and the server resource management is efficiently guided.
In some alternative implementations, the peak estimation module 420 is specifically configured to:
acquiring average load data and service per second request data of a central processing unit of a server in a first historical time period before the current time;
establishing a fitting relation between the average load data of the central processing unit of the server and the request data of the server per second;
and based on the fitting relation, taking the service request quantity per second when the average load value of the central processing unit of the server is a preset upper limit load value as the peak value of the service request quantity per second.
In some optional implementations, the server cluster capacity evaluation apparatus further includes a service average response time threshold determination module, configured to:
acquiring service average response time data in a second historical time period before the current time, and performing statistical analysis to obtain distribution characteristics of the service average response time data;
and determining the preset service average response time threshold according to the distribution characteristics.
In some optional implementations, the capacity evaluation module 430 is specifically configured to:
calculating the ratio of the service request quantity per second to the service request quantity per second peak value to obtain a first service capacity evaluation parameter;
calculating the ratio of the average service response time to the preset average service response time threshold to obtain a second service capacity evaluation parameter;
and respectively calculating the products of the first service capacity evaluation parameter, the second service capacity evaluation parameter and the average load of the service central processing unit and the corresponding parameter weight coefficients, and taking the sum of the obtained product results as the current capacity evaluation result of the server cluster to be evaluated.
In some optional implementations, the server cluster capacity evaluation apparatus further includes a parameter weighting factor determination module, configured to:
acquiring a preset quantity group parameter weight coefficient value;
evaluating the capacity of the server cluster according to the server cluster capacity evaluation logics under the weight coefficient values of each group of parameters;
and taking the evaluation results of the server cluster capacity evaluation logics under the parameter weight coefficient values of all groups as comparison groups within a preset time period, performing regression testing, and determining the final parameter weight coefficient value according to the regression testing results.
In some optional implementations, the server cluster capacity evaluation apparatus further includes a service resource adjustment module, configured to:
inputting the service request quantity per second data in a third history time period before the current time into a preset request quantity prediction model to obtain a service request quantity per second result at the preset time;
and calculating the difference value between the peak value of the request quantity per second of the service and the request quantity per second of the service at the preset moment, and adjusting the server resources of the server cluster to be evaluated according to the difference value.
In some optional implementations, the server cluster capacity evaluation apparatus further includes a parameter management module, configured to:
acquiring server operation data from the server operation monitoring data and storing the server operation data in a preset database;
extracting preset field data from the preset database, and preprocessing the extracted data to obtain a server cluster capacity evaluation data set, wherein the preset field comprises data acquisition time, service request amount per second, service average response time and service central processing unit average load;
correspondingly, the data obtaining module 410 is further configured to:
and in the server cluster capacity evaluation data set, acquiring the service per second request quantity, the service average response time and the server central processing unit average load of the current server cluster to be evaluated.
The server cluster capacity evaluation device provided by the embodiment of the disclosure can execute the server cluster capacity evaluation method provided by any embodiment of the disclosure, and has corresponding functional modules and beneficial effects of the execution method.
It should be noted that, the units and modules included in the apparatus are merely divided according to functional logic, but are not limited to the above division as long as the corresponding functions can be implemented; in addition, specific names of the functional units are only used for distinguishing one functional unit from another, and are not used for limiting the protection scope of the embodiments of the present disclosure.
EXAMPLE five
Referring now to fig. 8, a schematic diagram of an electronic device (e.g., the terminal device or the server in fig. 8) 500 suitable for implementing embodiments of the present disclosure is shown. The terminal device in the embodiments of the present disclosure may include, but is not limited to, a mobile terminal such as a mobile phone, a notebook computer, a digital broadcast receiver, a PDA (personal digital assistant), a PAD (tablet computer), a PMP (portable multimedia player), a vehicle terminal (e.g., a car navigation terminal), and the like, and a stationary terminal such as a digital TV, a desktop computer, and the like. The electronic device shown in fig. 8 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present disclosure.
As shown in fig. 8, the electronic device 500 may include a processing means (e.g., central processor, graphics processor, etc.) 501 that may perform various appropriate actions and processes in accordance with a program stored in a Read-Only Memory (ROM) 502 or a program loaded from a storage means 506 into a Random Access Memory (RAM) 503. In the RAM 503, various programs and data necessary for the operation of the electronic apparatus 500 are also stored. The processing device 501, the ROM502, and the RAM 503 are connected to each other through a bus 504. An input/output (I/O) interface 505 is also connected to bus 504.
Generally, the following devices may be connected to the I/O interface 505: input devices 506 including, for example, a touch screen, touch pad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; output devices 507 including, for example, a Liquid Crystal Display (LCD), speakers, vibrators, and the like; storage devices 508 including, for example, magnetic tape, hard disk, etc.; and a communication device 509. The communication means 509 may allow the electronic device 500 to communicate with other devices wirelessly or by wire to exchange data. While fig. 8 illustrates an electronic device 500 having various means, it is to be understood that not all illustrated means are required to be implemented or provided. More or fewer devices may alternatively be implemented or provided.
In particular, according to an embodiment of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program carried on a non-transitory computer readable medium, the computer program containing program code for performing the method illustrated by the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network via the communication means 509, or installed from the storage means 506, or installed from the ROM 502. The computer program performs the above-described functions defined in the server cluster capacity evaluation method of the embodiment of the present disclosure when executed by the processing device 501.
The functions that can be realized by the electronic device provided by the embodiment of the present disclosure and the server cluster capacity evaluation method provided by the embodiment of the present disclosure belong to the same disclosure concept, and the technical details that are not described in detail in the embodiment of the present disclosure may refer to the embodiment of the present disclosure, and the embodiment of the present disclosure has the same beneficial effects as the embodiment of the present disclosure.
EXAMPLE six
The embodiment of the present disclosure provides a computer storage medium, on which a computer program is stored, and when the program is executed by a processor, the method for evaluating the capacity of a server cluster provided by the above embodiment is implemented.
It should be noted that the computer readable medium in the present disclosure can be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a Read-Only Memory (ROM), an Erasable Programmable Read-Only Memory (EPROM) or FLASH Memory (FLASH), an optical fiber, a portable compact disc Read-Only Memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In contrast, in the present disclosure, a computer readable signal medium may comprise a propagated data signal with computer readable program code embodied therein, either in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, optical cables, RF (radio frequency), etc., or any suitable combination of the foregoing.
In some embodiments, the clients, servers may communicate using any currently known or future developed network Protocol, such as HTTP (Hyper Text Transfer Protocol), and may interconnect with any form or medium of digital data communication (e.g., a communications network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), the Internet (e.g., the Internet), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future developed network.
The computer readable medium may be embodied in the electronic device; or may exist separately without being assembled into the electronic device.
The computer readable medium carries one or more programs which, when executed by the electronic device, cause the electronic device to:
acquiring the service per second request quantity, the service average response time and the server central processing unit average load of the current server cluster to be evaluated;
estimating the peak value of the request quantity per second of service of the current server cluster to be evaluated according to the real-time function relationship between the average load of the central processing unit of the server and the request quantity per second of service;
and obtaining the current capacity evaluation result of the server cluster to be evaluated according to a preset server cluster capacity evaluation logic based on the request amount per second of service, the peak value of the request amount per second of service, the average response time of service, a preset average response time threshold value of service and the average load of the central processing unit of the server.
Computer program code for carrying out operations for the present disclosure may be written in any combination of one or more programming languages, including but not limited to an object oriented programming language such as Java, Smalltalk, C + +, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The units described in the embodiments of the present disclosure may be implemented by software or hardware. The names of the units and modules do not limit the units and modules in some cases, and for example, the data generation module may be described as a "video data generation module".
The functions described herein above 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: a Field Programmable Gate Array (FPGA), an Application Specific Integrated Circuit (ASIC), an Application Specific Standard Part (ASSP), a System On Chip (SOC), a Complex Programmable Logic Device (CPLD), and the like.
In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The 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, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, 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 disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
According to one or more embodiments of the present disclosure, [ example one ] there is provided a server cluster capacity assessment method, the method comprising:
acquiring the service per second request quantity, the service average response time and the server central processing unit average load of the current server cluster to be evaluated;
estimating the peak value of the request quantity per second of service of the current server cluster to be evaluated according to the real-time function relationship between the average load of the central processing unit of the server and the request quantity per second of service;
and obtaining the current capacity evaluation result of the server cluster to be evaluated according to a preset server cluster capacity evaluation logic based on the request amount per second of service, the peak value of the request amount per second of service, the average response time of service, a preset average response time threshold value of service and the average load of the central processing unit of the server.
According to one or more embodiments of the present disclosure, [ example two ] there is provided a server cluster capacity evaluation method, further comprising:
in some optional implementations, the estimating a peak value of the request volume per second of service of the current server cluster to be evaluated according to a real-time functional relationship between an average load of a central processing unit of the server and the request volume per second of service includes:
acquiring average load data and service per second request data of a central processing unit of a server in a first historical time period before the current time;
establishing a fitting relation between the average load data of the central processing unit of the server and the request data of the server per second;
and based on the fitting relation, taking the service request quantity per second when the average load value of the central processing unit of the server is a preset upper limit load value as the peak value of the service request quantity per second.
According to one or more embodiments of the present disclosure, [ example three ] there is provided a server cluster capacity evaluation method, further comprising:
in some optional implementations, the method for evaluating the capacity of a server cluster further includes:
acquiring service average response time data in a second historical time period before the current time, and performing statistical analysis to obtain distribution characteristics of the service average response time data;
and determining the preset service average response time threshold according to the distribution characteristics.
According to one or more embodiments of the present disclosure, [ example four ] there is provided a server cluster capacity evaluation method, further comprising:
in some optional implementation manners, the obtaining a current capacity evaluation result of the server cluster to be evaluated according to a preset server cluster capacity evaluation logic based on the request amount per second of service, the peak value of the request amount per second of service, the average response time of service, a preset threshold value of average response time of service, and the average load of the central processing unit of the server includes:
calculating the ratio of the service request quantity per second to the service request quantity per second peak value to obtain a first service capacity evaluation parameter;
calculating the ratio of the average service response time to the preset average service response time threshold to obtain a second service capacity evaluation parameter;
and respectively calculating the products of the first service capacity evaluation parameter, the second service capacity evaluation parameter and the average load of the service central processing unit and the corresponding parameter weight coefficients, and taking the sum of the obtained product results as the current capacity evaluation result of the server cluster to be evaluated.
According to one or more embodiments of the present disclosure, [ example five ] there is provided a server cluster capacity evaluation method, further comprising:
in some optional implementation manners, the determining process of the parameter weight coefficients corresponding to the first service capacity evaluation parameter, the second service capacity evaluation parameter, and the average load of the service central processing unit respectively includes:
acquiring a preset quantity group parameter weight coefficient value;
evaluating the capacity of the server cluster according to the server cluster capacity evaluation logics under the weight coefficient values of each group of parameters;
and taking the evaluation results of the server cluster capacity evaluation logics under the parameter weight coefficient values of all groups as comparison groups within a preset time period, performing regression testing, and determining the final parameter weight coefficient value according to the regression testing results.
According to one or more embodiments of the present disclosure, [ example six ] there is provided a server cluster capacity evaluation method, further comprising:
in some optional implementations, the server cluster capacity evaluation method further includes:
inputting the service request quantity per second data in a third history time period before the current time into a preset request quantity prediction model to obtain a service request quantity per second result at the preset time;
and calculating the difference value between the peak value of the request quantity per second of the service and the request quantity per second of the service at the preset moment, and adjusting the server resources of the server cluster to be evaluated according to the difference value.
According to one or more embodiments of the present disclosure, [ example seven ] there is provided a server cluster capacity evaluation method, further comprising:
in some optional implementations, before obtaining the service request amount per second, the service average response time, and the server central processor average load of the current server cluster to be evaluated, the method further includes:
acquiring server cluster operation data from the server cluster operation monitoring data and storing the server cluster operation data in a preset database;
extracting preset field data from the preset database, and preprocessing the extracted data to obtain a server cluster capacity evaluation data set, wherein the preset field comprises data acquisition time, service request amount per second, service average response time and service central processing unit average load;
correspondingly, the obtaining of the service request amount per second, the service average response time and the server central processing unit average load of the current server cluster to be evaluated includes:
and in the server cluster capacity evaluation data set, acquiring the service per second request quantity, the service average response time and the server central processing unit average load of the current server cluster to be evaluated.
According to one or more embodiments of the present disclosure, [ example eight ] there is provided a server cluster capacity evaluation apparatus including:
the data acquisition module is used for acquiring the service request quantity per second, the service average response time and the server central processing unit average load of the current server cluster to be evaluated;
the peak value estimation module is used for estimating the peak value of the request quantity per second of service of the current server cluster to be evaluated according to the real-time function relationship between the average load of the central processing unit of the server and the request quantity per second of service;
and the capacity evaluation module is used for obtaining the current capacity evaluation result of the server cluster to be evaluated according to a preset server cluster capacity evaluation logic based on the request amount per second of service, the request amount peak value per second of service, the average response time of service, a preset average response time threshold value and the average load of the central processing unit of the server.
According to one or more embodiments of the present disclosure, [ example nine ] there is provided a server cluster capacity evaluation apparatus further including:
in some optional implementations, the peak estimation module is specifically configured to:
acquiring average load data and service per second request data of a central processing unit of a server in a first historical time period before the current time;
establishing a fitting relation between the average load data of the central processing unit of the server and the request data of the server per second;
and based on the fitting relation, taking the service request quantity per second when the average load value of the central processing unit of the server is a preset upper limit load value as the peak value of the service request quantity per second.
According to one or more embodiments of the present disclosure, [ example ten ] there is provided a server cluster capacity evaluation apparatus further including:
in some optional implementations, the server cluster capacity evaluation apparatus further includes a service average response time threshold determination module, configured to:
acquiring service average response time data in a second historical time period before the current time, and performing statistical analysis to obtain distribution characteristics of the service average response time data;
and determining the preset service average response time threshold according to the distribution characteristics.
According to one or more embodiments of the present disclosure, [ example eleven ] there is provided a server cluster capacity evaluation apparatus further including:
in some optional implementations, the capacity evaluation module is specifically configured to:
calculating the ratio of the service request quantity per second to the service request quantity per second peak value to obtain a first service capacity evaluation parameter;
calculating the ratio of the average service response time to the preset average service response time threshold to obtain a second service capacity evaluation parameter;
and respectively calculating the products of the first service capacity evaluation parameter, the second service capacity evaluation parameter and the average load of the service central processing unit and the corresponding parameter weight coefficients, and taking the sum of the obtained product results as the current capacity evaluation result of the server cluster to be evaluated.
According to one or more embodiments of the present disclosure, [ example twelve ] there is provided a server cluster capacity evaluation apparatus further comprising:
in some optional implementations, the server cluster capacity evaluation apparatus further includes a parameter weighting factor determination module, configured to:
acquiring a preset quantity group parameter weight coefficient value;
evaluating the capacity of the server cluster according to the server cluster capacity evaluation logics under the weight coefficient values of each group of parameters;
and taking the evaluation results of the server cluster capacity evaluation logics under the parameter weight coefficient values of all groups as comparison groups within a preset time period, performing regression testing, and determining the final parameter weight coefficient value according to the regression testing results.
According to one or more embodiments of the present disclosure, [ example thirteen ] provides a server cluster capacity evaluation apparatus, further including:
in some optional implementations, the server cluster capacity evaluation apparatus further includes a service resource adjustment module, configured to:
inputting the service request quantity per second data in a third history time period before the current time into a preset request quantity prediction model to obtain a service request quantity per second result at the preset time;
and calculating the difference value between the peak value of the request quantity per second of the service and the request quantity per second of the service at the preset moment, and adjusting the server resources of the server cluster to be evaluated according to the difference value.
According to one or more embodiments of the present disclosure, [ example fourteen ] there is provided a server cluster capacity evaluation apparatus further comprising:
in some optional implementations, the server cluster capacity evaluation apparatus further includes a parameter management module, configured to:
acquiring server cluster operation data from the server cluster operation monitoring data and storing the server cluster operation data in a preset database;
extracting preset field data from the preset database, and preprocessing the extracted data to obtain a server cluster capacity evaluation data set, wherein the preset field comprises data acquisition time, service request amount per second, service average response time and service central processing unit average load;
correspondingly, the data obtaining module 410 is further configured to:
and in the server cluster capacity evaluation data set, acquiring the service per second request quantity, the service average response time and the server central processing unit average load of the current server cluster to be evaluated.
The foregoing description is only exemplary of the preferred embodiments of the disclosure and is illustrative of the principles of the technology employed. It will be appreciated by those skilled in the art that the scope of the disclosure herein is not limited to the particular combination of features described above, but also encompasses other embodiments in which any combination of the features described above or their equivalents does not depart from the spirit of the disclosure. For example, the above features and (but not limited to) the features disclosed in this disclosure having similar functions are replaced with each other to form the technical solution.
Further, while operations are depicted in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Likewise, while several specific implementation details are included in the above discussion, these should not be construed as limitations on the scope of the disclosure. Certain features that are described in the context of separate embodiments can also be implemented in combination in a single embodiment. Conversely, various features that are described in the context of a single embodiment can also be implemented in multiple embodiments separately or in any suitable subcombination.
Although the subject matter has been described in language specific to structural features and/or methodological acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are disclosed as example forms of implementing the claims.
