Intelligent test method and system for cluster server

Abstract

The application is suitable for the technical field of cluster server, and provides an intelligent testing method and system for a cluster server. The application matches real-time monitoring and virtual testing of a target cluster server, intercepts virtual testing data, extracts testing requirement data, analyzes the virtual testing, calculates the execution performance value and the queuing performance value of the target cluster server, compares and analyzes the execution performance value and the queuing performance value, and generates testing feedback information. The application can match real-time monitoring data with virtual testing, intercept virtual testing data, extract testing requirement data, analyze the virtual testing, calculate the execution performance value and the queuing performance value, compare and analyze the testing, generate testing feedback information, and does not need to suspend the daily operation of the cluster server, but needs to arrange a separate testing period for relevant detection. The testing process is simple and professional, effectively guarantees the availability of the cluster server, and improves the efficiency of the cluster server testing.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of cluster servers, and particularly relates to an intelligent testing method and system for a cluster server. BACKGROUND

[0002] A cluster server is a form of server organization, and is a plurality of servers interconnected through a network to work together to achieve high performance, high availability and scalability, and is mainly applied to cloud computing, big data processing, website hosting and virtualization scenarios.

[0003] Cluster server testing is an activity of verifying and evaluating a cluster server system.

[0004] In the prior art, for the testing of a cluster server that has been put into operation, the daily operation of the cluster server needs to be suspended, and a separate testing period is specially arranged for related detection. The testing process is complicated, and a large amount of time is needed for recording and analysis, which affects the normal business operation of the cluster server, reduces the availability of the cluster server, and there is a significant deficiency in testing efficiency. SUMMARY

[0005] The purpose of the embodiment of the application is to provide an intelligent testing method and system for a cluster server, which aims to solve the technical problems existing in the prior art mentioned in the background.

[0006] The embodiment of the application is implemented as follows:

[0007] An intelligent testing method for a cluster server, which specifically comprises the following steps:

[0008] Real-time monitoring of a target cluster server is performed to obtain real-time monitoring data, and the real-time monitoring data is subjected to virtual testing matching to intercept virtual testing data.

[0009] Data type identification is performed on the virtual testing data, and test requirement data is extracted from the virtual testing data.

[0010] Virtual testing analysis is performed according to the test requirement data, and an execution performance value and a queuing performance value of the target cluster server are calculated.

[0011] Based on preset standard performance data, the execution performance value and the queuing performance value are subjected to testing comparison and analysis to generate testing feedback information.

[0012] As a further limitation of the technical scheme of the embodiment of the application, the real-time monitoring of the target cluster server to obtain real-time monitoring data, and the virtual testing matching of the real-time monitoring data to intercept virtual testing data specifically comprises the following steps:

[0013] determining a target cluster server with a test requirement;

[0014] real-time monitoring and recording the target cluster server to obtain real-time monitoring data;

[0015] performing virtual test matching on the real-time monitoring data to determine a test matching period;

[0016] according to the test matching period, cutting virtual test data from the real-time monitoring data.

[0017] As a further limitation of the technical scheme of the embodiment of the application, the virtual test matching on the real-time monitoring data to determine a test matching period specifically includes the following steps:

[0018] performing task matching on the real-time monitoring data to determine a task matching period with a single task;

[0019] performing running matching on the real-time monitoring data to determine a running matching period with complete running;

[0020] determining a plurality of overlapping periods according to the task matching period and the running matching period;

[0021] according to preset period requirements, selecting a test matching period from the plurality of overlapping periods.

[0022] As a further limitation of the technical scheme of the embodiment of the application, the data type identification on the virtual test data to extract test requirement data from the virtual test data specifically includes the following steps:

[0023] identifying the virtual test data to determine a test task;

[0024] identifying a plurality of test requests corresponding to the test task;

[0025] according to the plurality of test requests, extracting relevant test data from the virtual test data;

[0026] according to a plurality of preset requirement data types, extracting test requirement data from the relevant test data.

[0027] As a further limitation of the technical scheme of the embodiment of the application, the virtual test analysis according to the test requirement data to calculate the execution performance value and the queuing performance value of the target cluster server specifically includes the following steps:

[0028] performing request service analysis of virtual test according to the test requirement data to obtain a request service rate;

[0029] According to the test requirement data, request arrival analysis of virtual testing is performed to obtain a request arrival rate;

[0030] Based on the test requirement data, an execution performance value of the target cluster server is calculated;

[0031] Based on the request service rate and the request arrival rate, a queuing performance value of the target cluster server is calculated.

[0032] As a further limitation of the technical scheme of the embodiment of the application, the calculation formula of the execution performance value is:

[0033]

[0034] Wherein, A is the execution performance value, i represents the ith test request, n is the total number of test requests, Tr i is the response time of the ith test request, Ts i is the service time of the ith test request.

[0035] The calculation formula of the queuing performance value is:

[0036]

[0037] Wherein, B is the queuing performance value, μ is the request service rate, and λ is the request arrival rate.

[0038] As a further limitation of the technical scheme of the embodiment of the application, the test comparison analysis of the execution performance value and the queuing performance value based on the preset standard performance data generates test feedback information, which specifically includes the following steps:

[0039] Based on the preset standard performance data, the test comparison analysis of the execution performance value and the queuing performance value is performed, and the test comparison result is recorded;

[0040] Importing a test feedback template;

[0041] According to the test feedback template, the test comparison result is sorted and filled to generate test feedback information.

[0042] An intelligent testing system for a cluster server, the system comprising a real-time monitoring processing module, a test data extraction module, a virtual test analysis module, and a test comparison analysis module, wherein:

[0043] The real-time monitoring processing module is used for real-time monitoring of a target cluster server, obtaining real-time monitoring data, and performing virtual test matching on the real-time monitoring data to intercept virtual test data;

[0044] The test data extraction module is configured to perform data type identification on the virtual test data, and extract test requirement data from the virtual test data.

[0045] The virtual test analysis module is configured to perform virtual test analysis on the target cluster server according to the test requirement data, and calculate an execution performance value and a queuing performance value of the target cluster server.

[0046] The test comparison analysis module is configured to perform test comparison analysis on the execution performance value and the queuing performance value based on preset standard performance data, and generate test feedback information.

[0047] As a further limitation of the technical scheme of the embodiment of the present application, the real-time monitoring processing module specifically comprises:

[0048] The cluster server determination unit is configured to determine a target cluster server having a test requirement.

[0049] The monitoring recording unit is configured to perform real-time monitoring and recording on the target cluster server, and obtain real-time monitoring data.

[0050] The virtual test matching unit is configured to perform virtual test matching on the real-time monitoring data, and determine a test matching period.

[0051] The data interception unit is configured to intercept virtual test data from the real-time monitoring data according to the test matching period.

[0052] As a further limitation of the technical scheme of the embodiment of the present application, the test data extraction module specifically comprises:

[0053] The task identification unit is configured to identify the virtual test data, and determine a test task.

[0054] The request identification unit is configured to identify a plurality of test requests corresponding to the test task.

[0055] The relevant test data extraction unit is configured to extract relevant test data from the virtual test data according to the plurality of test requests.

[0056] The test requirement data extraction unit is configured to extract test requirement data from the relevant test data according to a plurality of preset requirement data types.

[0057] Compared with the prior art, the present application has the following advantages:

[0058] The embodiment of the present application can perform real-time monitoring and virtual test matching on the target cluster server, intercept virtual test data, extract test requirement data, perform virtual test analysis, calculate the execution performance value and the queuing performance value of the target cluster server, perform test comparison and analysis on the execution performance value and the queuing performance value, and generate test feedback information. The real-time monitoring data can be matched with virtual test data, the virtual test data can be intercepted, the test requirement data can be extracted, the virtual test analysis can be performed, the execution performance value and the queuing performance value can be calculated, the test comparison and analysis can be performed, and the test feedback information can be generated. The daily operation of the cluster server does not need to be paused, and a separate test period is arranged for related detection. The test process is simple and professional, the availability of the cluster server is effectively ensured, and the efficiency of the cluster server test is improved. BRIEF DESCRIPTION OF DRAWINGS

[0059] Figure 1 A flowchart of the intelligent test method of the cluster server is shown;

[0060] Figure 2 A flowchart of the virtual test data interception in the method is shown;

[0061] Figure 3 A flowchart of the virtual test matching in the method is shown;

[0062] Figure 4 A flowchart of the test requirement data extraction in the method is shown;

[0063] Figure 5 A flowchart of the virtual test analysis in the method is shown;

[0064] Figure 6 A flowchart of the test comparison and analysis in the method is shown;

[0065] Figure 7 An application architecture diagram of the intelligent test system of the cluster server is shown;

[0066] Figure 8 A structural block diagram of the real-time monitoring processing module in the system is shown;

[0067] Figure 9 A structural block diagram of the test data extraction module in the system is shown. DETAILED DESCRIPTION

[0068] In order to make the objectives, technical solutions and advantages of the present application clearer, further detailed description will be given to the present application in combination with the accompanying drawings and examples. It should be understood that the specific examples described herein are only used to explain the present application and not to limit the present application.

[0069] It can be understood that, in the prior art, for the test of the cluster server which has been put into operation, the daily operation of the cluster server needs to be suspended, and a separate test period is specially arranged for the related detection, and the test process is complicated, and a large amount of time is needed for recording and analysis, which not only affects the normal business operation of the cluster server, but also reduces the availability of the cluster server, and the test efficiency is obviously insufficient.

[0070] In order to solve the above problems, the embodiment of the present application discloses an intelligent test method and system for a cluster server, which acquires real-time monitoring data by monitoring a target cluster server in real time, and performs virtual test matching on the real-time monitoring data to intercept virtual test data; the data type of the virtual test data is identified, and test requirement data is extracted from the virtual test data; virtual test analysis is performed according to the test requirement data to calculate the execution performance value and the queuing performance value of the target cluster server; and the execution performance value and the queuing performance value are compared and analyzed based on the preset standard performance data to generate test feedback information. The real-time monitoring data can be virtually tested and matched to intercept virtual test data, extract test requirement data, perform virtual test analysis, calculate execution performance value and queuing performance value, and generate test feedback information through comparison and analysis, without suspending the daily operation of the cluster server and specially arranging a separate test period for related detection, and the test process is simple and professional, which effectively guarantees the availability of the cluster server and improves the efficiency of the cluster server test.

[0071] Specifically, Figure 1 A flowchart of an intelligent test method for a cluster server provided by the embodiment of the present application is shown.

[0072] In a preferred embodiment provided by the present application, an intelligent test method for a cluster server specifically includes the following steps:

[0073] Step S101, a target cluster server is monitored in real time to acquire real-time monitoring data, and the real-time monitoring data is virtually tested and matched to intercept virtual test data.

[0074] In the embodiment of the present application, the target cluster server with test requirements is determined, the real-time monitoring data is obtained by monitoring and recording the running state, task processing, node state, business index and the like used in daily operation of the target cluster server, the real-time monitoring data is matched, the working period in which the target cluster server only executes a single task is determined and marked as a task matching period, the running matching is performed on the real-time monitoring data, the working period in which all nodes in the target cluster server are in a running state is determined and marked as a running matching period, the task matching period and the running matching period are analyzed for period overlap, a plurality of overlapping periods that meet the task matching period and the running matching period are determined, one overlapping period that meets the period requirement is selected from the plurality of overlapping periods according to a preset period requirement, and the selected overlapping period is marked as a test matching period, and then the virtual test data is intercepted from the real-time monitoring data according to the test matching period.

[0075] It can be understood that the period requirement can include a requirement for the period length and a time requirement from the current time (for example, the period length is 1-2 hours, and the time from the current time is not more than 30 days), so that the selected test matching period has a certain length of test research and a freshness of not being far from the current time.

[0076] Specifically, Figure 2 A flowchart of intercepting virtual test data in the method provided by the embodiment of the present application is shown.

[0077] In another preferred embodiment provided by the present application, the real-time monitoring of the target cluster server, the acquisition of the real-time monitoring data, and the virtual test matching of the real-time monitoring data to intercept the virtual test data specifically include the following steps:

[0078] Step S1011, determining a target cluster server with test requirements.

[0079] Step S1012, monitoring and recording the target cluster server in real time to obtain real-time monitoring data.

[0080] Step S1013, performing virtual test matching on the real-time monitoring data to determine a test matching period.

[0081] Specifically, Figure 3 A flowchart of performing virtual test matching in the method provided by the embodiment of the present application is shown.

[0082] In another preferred embodiment provided by the present application, the virtual test matching of the real-time monitoring data to determine a test matching period specifically includes the following steps:

[0083] Step S10131, task matching is performed on the real-time monitoring data to determine a task single task matching period.

[0084] Step S10132, operation matching is performed on the real-time monitoring data to determine an operation complete operation matching period.

[0085] Step S10133, a plurality of overlapping periods are determined according to the task matching period and the operation matching period.

[0086] Step S10134, a test matching period is selected from the plurality of overlapping periods according to a preset period requirement.

[0087] Further, the real-time monitoring of the target cluster server, the acquisition of real-time monitoring data, and the virtual test matching of the real-time monitoring data to intercept virtual test data further include the following steps:

[0088] Step S1014, virtual test data is intercepted from the real-time monitoring data according to the test matching period.

[0089] Further, the intelligent test method of the cluster server further includes the following steps:

[0090] Step S102, data type identification is performed on the virtual test data to extract test requirement data from the virtual test data.

[0091] In the embodiment of the application, by identifying the virtual test data, the operation task executed by the target cluster server in the test matching period is determined and is marked as a test task. Since the cluster server will coordinate and distribute tasks to form a plurality of requests and send them to a plurality of nodes, a plurality of test requests corresponding to the test task can be identified, and then according to the plurality of test requests, relevant test data is extracted from the virtual test data, and then according to a plurality of preset requirement data types, test requirement data is extracted from the relevant test data.

[0092] It can be understood that the extraction of relevant test data is a process of preliminary screening of the virtual test data, and data irrelevant to the test request in the virtual test data can be eliminated; the extraction of test requirement data is a further screening of the relevant test data, and data irrelevant to the plurality of requirement data types can be eliminated.

[0093] It can be understood that the test task is actually an execution task in the daily operation of the target cluster server; the test request is actually an execution request for coordinating and distributing the execution task in the daily operation of the target cluster server.

[0094] Specifically, Figure 4A flow chart of extracting test requirement data in the method provided by the embodiment of the application is shown.

[0095] In another preferred embodiment provided by the application, the data type identification of the virtual test data and the extraction of the test requirement data from the virtual test data specifically include the following steps:

[0096] Step S1021, identifying the virtual test data to determine a test task.

[0097] Step S1022, identifying a plurality of test requests corresponding to the test task.

[0098] Step S1023, extracting relevant test data from the virtual test data according to the plurality of test requests.

[0099] Step S1024, extracting test requirement data from the relevant test data according to a plurality of preset requirement data types.

[0100] Further, the intelligent test method of the cluster server further includes the following steps:

[0101] Step S103, performing virtual test analysis according to the test requirement data to calculate an execution performance value and a queuing performance value of the target cluster server.

[0102] In the embodiment of the application, the request service analysis of the virtual test is performed according to the test requirement data to obtain a request service rate, the request arrival analysis of the virtual test is performed to obtain a request arrival rate, the execution performance value of the target cluster server is calculated based on the test requirement data, and the queuing performance value of the target cluster server is calculated based on the request service rate and the request arrival rate. Specifically, the calculation formula of the execution performance value is:

[0103]

[0104] wherein A is the execution performance value, i represents the i th test request, n is the total number of test requests, Tr i is the response time of the i th test request, Ts i is the service time of the i th test request.

[0105] The calculation formula of the queuing performance value is:

[0106]

[0107] wherein B is the queuing performance value, μ is the request service rate, and λ is the request arrival rate.

[0108] It can be understood that the request service rate is the average number of test request services (executed and completed) per unit of time; and the request arrival rate is the average number of test requests arriving at the corresponding server node of the target cluster server per unit of time.

[0109] It can be understood that the execution performance value is related to the load balancing performance of the target cluster server, and the greater the value of the execution performance value, the heavier the load of the target cluster server; the smaller the value of the execution performance value, the lighter the load of the target cluster server, so the smaller the value of the execution performance value, the better the load balancing performance of the target cluster server, and the test analysis of the target cluster server through the execution performance value can solve the defect that the performance evaluation through the response time is unfair due to large differences in different request service times.

[0110] It can be understood that the queuing performance value is related to the request queuing time of multiple test requests in the target cluster server, and can reflect the request processing efficiency of the target cluster server, and the greater the queuing performance value, the higher the request processing efficiency of the target cluster server; the smaller the queuing performance value, the lower the request processing efficiency of the target cluster server.

[0111] Specifically, Figure 5 A flowchart of the virtual test analysis in the method provided by the embodiment of the application is shown.

[0112] In another preferred embodiment provided by the application, the virtual test analysis according to the test requirement data and the calculation of the execution performance value and the queuing performance value of the target cluster server specifically include the following steps:

[0113] Step S1031, performing request service analysis of virtual test according to the test requirement data to obtain a request service rate.

[0114] Step S1032, performing request arrival analysis of virtual test according to the test requirement data to obtain a request arrival rate.

[0115] Step S1033, calculating an execution performance value of the target cluster server based on the test requirement data.

[0116] Step S1034, calculating a queuing performance value of the target cluster server based on the request service rate and the request arrival rate.

[0117] Further, the intelligent test method of the cluster server further includes the following steps:

[0118] Step S104, performing test comparison analysis on the execution performance value and the queuing performance value based on the preset standard performance data to generate test feedback information.

[0119] In the embodiment of the present application, based on the preset standard performance data, the execution performance value and the queuing performance value are compared in test values, and the test comparison result is recorded, the test feedback template is imported, and the test comparison result is filled and arranged based on the test feedback template, and the test feedback information is generated, and when there is a browsing demand, the test feedback information is displayed.

[0120] It can be understood that, by directly comparing the execution performance value and the queuing performance value in test values, the test is simple and professional, and can accurately reflect the load balancing performance and request processing efficiency of the target cluster server, effectively avoiding a complex test process, and without a large amount of time for recording and analysis, thereby realizing intelligent test of the cluster server quickly and accurately.

[0121] Specifically, Figure 6 A flowchart for test comparison and analysis in the method provided by the embodiment of the present application is shown.

[0122] In another preferred embodiment provided by the present application, the test comparison and analysis of the execution performance value and the queuing performance value based on the preset standard performance data to generate test feedback information specifically includes the following steps:

[0123] Step S1041, based on the preset standard performance data, the execution performance value and the queuing performance value are compared in test values, and the test comparison result is recorded.

[0124] Step S1042, import the test feedback template.

[0125] Step S1043, according to the test feedback template, the test comparison result is filled and arranged to generate test feedback information.

[0126] Further, Figure 7 An application architecture diagram of the intelligent test system of the cluster server provided by the embodiment of the present application is shown.

[0127] Specifically, in another preferred embodiment provided by the present application, an intelligent test system of a cluster server includes:

[0128] The real-time monitoring processing module 101 is used for real-time monitoring of the target cluster server, acquiring real-time monitoring data, and performing virtual test matching on the real-time monitoring data, and intercepting virtual test data.

[0129] In the embodiment of the present application, the real-time monitoring processing module 101 determines a target cluster server with test requirements, performs real-time monitoring and recording on the running state, task processing, node state, business indicators and the like used in daily operation of the target cluster server, obtains real-time monitoring data, performs task matching on the real-time monitoring data, determines a working period in which the target cluster server only performs a single task, marks it as a task matching period, performs running matching on the real-time monitoring data, determines a working period in which all nodes in the target cluster server are in a running state, marks it as a running matching period, performs period overlap analysis on the task matching period and the running matching period, determines a plurality of overlapping periods that simultaneously meet the task matching period and the running matching period, selects an overlapping period that meets the period requirement from the plurality of overlapping periods according to a preset period requirement, and marks it as a test matching period, and then extracts virtual test data from the real-time monitoring data according to the test matching period.

[0130] Further, Figure 8 The structure block diagram of the real-time monitoring processing module 101 in the system provided by the embodiment of the present application is shown.

[0131] Specifically, in another preferred embodiment provided by the present application, the real-time monitoring processing module 101 specifically includes:

[0132] The cluster server determination unit 1011 is configured to determine a target cluster server with test requirements.

[0133] The monitoring and recording unit 1012 is configured to perform real-time monitoring and recording on the target cluster server, and obtain real-time monitoring data.

[0134] The virtual test matching unit 1013 is configured to perform virtual test matching on the real-time monitoring data, and determine a test matching period.

[0135] The data extraction unit 1014 is configured to extract virtual test data from the real-time monitoring data according to the test matching period.

[0136] Further, the intelligent test system of the cluster server further includes:

[0137] The test data extraction module 102 is configured to perform data type identification on the virtual test data, and extract test requirement data from the virtual test data.

[0138] In the embodiment of the present application, the test data extraction module 102 identifies the virtual test data, determines the operation task performed by the target cluster server in the test matching period, and marks it as a test task. Since the cluster server will coordinate and distribute the task, form multiple requests and send them to multiple nodes, the multiple test requests corresponding to the test task can be identified, and then the relevant test data is extracted from the virtual test data according to the multiple test requests, and the test requirement data is extracted from the relevant test data according to the preset multiple requirement data types.

[0139] Further, Figure 9 The structural block diagram of the test data extraction module 102 in the system provided by the embodiment of the present application is shown.

[0140] Specifically, in another preferred embodiment provided by the present application, the test data extraction module 102 specifically includes:

[0141] The task identification unit 1021 is configured to identify the virtual test data and determine the test task.

[0142] The request identification unit 1022 is configured to identify the multiple test requests corresponding to the test task.

[0143] The relevant test data extraction unit 1023 is configured to extract relevant test data from the virtual test data according to the multiple test requests.

[0144] The test requirement data extraction unit 1024 is configured to extract test requirement data from the relevant test data according to the preset multiple requirement data types.

[0145] Further, the intelligent test system of the cluster server further includes:

[0146] The virtual test analysis module 103 is configured to perform virtual test analysis according to the test requirement data, and calculate the execution performance value and the queuing performance value of the target cluster server.

[0147] In the embodiment of the present application, the virtual test analysis module 103 performs request service analysis of virtual test according to the test requirement data, obtains the request service rate, and performs request arrival analysis of virtual test, obtains the request arrival rate, calculates the execution performance value of the target cluster server based on the test requirement data, and calculates the queuing performance value of the target cluster server based on the request service rate and the request arrival rate. Specifically, the calculation formula of the execution performance value is:

[0148]

[0149] Wherein, A is the execution performance value, i represents the ith test request, n is the total number of test requests, Tr i is the response time of the ith test request, Ts i is the service time of the ith test request.

[0150] The calculation formula of the queuing performance value is:

[0151]

[0152] Wherein, B is the queuing performance value, μ is the request service rate, and λ is the request arrival rate.

[0153] The test comparison and analysis module 104 is configured to perform test comparison and analysis on the execution performance value and the queuing performance value based on the preset standard performance data, and generate test feedback information.

[0154] In the embodiments of the present application, the test comparison and analysis module 104 performs test value comparison on the execution performance value and the queuing performance value based on the preset standard performance data, records the test comparison result, fills in the content of the test comparison result based on the test feedback template, and generates test feedback information. When there is a browsing demand, the test feedback information is displayed.

[0155] It should be understood that, although each step in the flowchart of each embodiment of the present application is displayed in sequence according to the arrow, these steps are not necessarily executed in sequence according to the arrow. Unless otherwise stated herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other sequences. Moreover, at least part of the steps in each embodiment can include multiple sub-steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence of these sub-steps or stages is not necessarily sequential, but can be executed in rotation or alternation with at least part of other steps or sub-steps or stages of other steps.

[0156] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The program can be stored in a non-volatile computer readable storage medium, and when the program is executed, the processes of the above-mentioned embodiments of the methods can be included. Any reference to memory, storage, database or other medium used in the embodiments provided in the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0157] The above-mentioned embodiments only express several embodiments of the present application, which are described in detail and specifically, but cannot be understood as a limitation on the scope of the patent of the present application. It should be noted that for those skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are all within the scope of protection of the present application. Therefore, the scope of protection of the patent of the present application should be subject to the appended claims.

Claims

1. A method for intelligent testing of a cluster server, the method comprising: The method specifically comprises the following steps: Real-time monitoring is performed on the target cluster server to obtain real-time monitoring data, and the real-time monitoring data is subjected to virtual test matching to intercept virtual test data; Task matching is performed on the real-time monitoring data to determine a working period in which the target cluster server only executes a single task, which is marked as a task matching period, and operation matching is performed on the real-time monitoring data to determine a working period in which all nodes in the target cluster server are in an operation state, which is marked as an operation matching period, the task matching period and the operation matching period are subjected to period overlap analysis to determine a plurality of overlap periods that simultaneously meet the task matching period and the operation matching period, then one overlap period that meets a period requirement is selected from the plurality of overlap periods according to the preset period requirement, and the overlap period is marked as a test matching period, and then virtual test data is intercepted from the real-time monitoring data according to the test matching period; the period requirement includes a period length requirement and a time requirement from a current time; Data type identification is performed on the virtual test data to extract test demand data from the virtual test data; By identifying the virtual test data, an operation task executed by the target cluster server in the test matching period is determined and marked as a test task, since the cluster server will coordinate and distribute the task to form a plurality of requests and send the plurality of requests to a plurality of nodes, a plurality of test requests corresponding to the test task are identified, then related test data is extracted from the virtual test data according to the plurality of test requests, and test demand data is extracted from the related test data according to a plurality of preset demand data types; the extraction of the related test data is a preliminary screening process of the virtual test data to eliminate data irrelevant to the test request; the extraction of the test demand data is a further screening process of the related test data to eliminate data irrelevant to the plurality of demand data types; the test task is actually an execution task in daily operation of the target cluster server; the test request is actually an execution request for coordinating and distributing the execution task in daily operation of the target cluster server; The daily operation of the cluster server does not need to be suspended, and a separate test period is specially arranged for related detection; Virtual test analysis is performed according to the test demand data to calculate an execution performance value and a queuing performance value of the target cluster server; Test comparison analysis is performed on the execution performance value and the queuing performance value based on preset standard performance data to generate test feedback information.

2. The intelligent testing method of cluster servers according to claim 1, characterized in that, The virtual test analysis performed according to the test demand data to calculate the execution performance value and the queuing performance value of the target cluster server specifically comprises the following steps: Request service analysis is performed on the virtual test according to the test demand data to obtain a request service rate; Request arrival analysis is performed on the virtual test according to the test demand data to obtain a request arrival rate; The execution performance value of the target cluster server is calculated based on the test demand data; The queuing performance value of the target cluster server is calculated based on the request service rate and the request arrival rate.

3. The intelligent testing method of cluster servers according to claim 2, characterized in that, The calculation formula of the execution performance value is: ； wherein, is the performance value, represents the first test request, is the total number of test requests, is the first response time of the test request, is the first service time of the test request; The calculation formula of the queuing performance value is: ； wherein, is a queuing performance value, is a request service rate, is a request arrival rate.

4. The intelligent testing method of cluster servers according to claim 1, characterized in that, The test feedback information generated by the test comparison and analysis of the execution performance value and the queuing performance value based on the preset standard performance data specifically includes the following steps: Based on the preset standard performance data, the execution performance value and the queuing performance value are tested and compared and analyzed, and the test comparison result is recorded; Import the test feedback template; According to the test feedback template, the test comparison result is arranged and filled to generate test feedback information.

5. An intelligent testing system of cluster servers, characterized in that, The system includes a real-time monitoring processing module, a test data extraction module, a virtual test analysis module, and a test comparison and analysis module, wherein: The real-time monitoring processing module is used for real-time monitoring of the target cluster server, obtaining real-time monitoring data, and performing virtual test matching on the real-time monitoring data to intercept virtual test data; The real-time monitoring processing module performs task matching on the real-time monitoring data, determines the working period of the target cluster server only executing a single task, marks it as a task matching period, and performs operation matching on the real-time monitoring data, determines the working period of all nodes in the target cluster server being in a running state, marks it as an operation matching period, analyzes the period overlap of the task matching period and the operation matching period, determines multiple overlapping periods that meet the task matching period and the operation matching period at the same time, selects an overlapping period that meets the period requirement from the multiple overlapping periods according to the preset period requirement, and marks it as a test matching period, and then according to the test matching period, the virtual test data is intercepted from the real-time monitoring data; The period requirement includes the requirement of the period length and the time requirement from the current time; The test data extraction module is used for data type identification of the virtual test data, and extraction of test requirement data from the virtual test data; The test data extraction module identifies the operation task of the target cluster server in the test matching period by identifying the virtual test data, and marks it as a test task. Since the cluster server will coordinate and distribute tasks to form multiple requests and send them to multiple nodes, multiple test requests corresponding to the test task are identified, and then according to the multiple test requests, related test data is extracted from the virtual test data, and then according to the preset multiple requirement data types, test requirement data is extracted from the related test data; The extraction of related test data is a preliminary screening process of virtual test data, which eliminates data unrelated to test requests in virtual test data; The extraction of test requirement data is a further screening of related test data, which eliminates data unrelated to multiple requirement data types; The test task is actually an execution task in the daily operation of the target cluster server; The test request is actually an execution request for coordinating and distributing the execution task in the daily operation of the target cluster server; Without pausing the daily operation of the cluster server, an independent test period is specially arranged for related detection; The virtual test analysis module is used for virtual test analysis according to the test requirement data, and calculation of the execution performance value and the queuing performance value of the target cluster server; The test comparison and analysis module is configured to perform test comparison and analysis on the execution performance value and the queuing performance value based on preset standard performance data, and generate test feedback information.

6. The intelligent test system of cluster servers according to claim 5, wherein, The real-time monitoring processing module specifically includes: A cluster server determination unit is configured to determine a target cluster server having a test requirement. A monitoring and recording unit is configured to perform real-time monitoring and recording on the target cluster server to obtain real-time monitoring data. A virtual test matching unit is configured to perform virtual test matching on the real-time monitoring data to determine a test matching period. A data interception unit is configured to intercept virtual test data from the real-time monitoring data according to the test matching period.

7. The intelligent test system of cluster servers according to claim 5, wherein, The test data extraction module specifically includes: A task identification unit is configured to identify the virtual test data to determine a test task. A request identification unit is configured to identify a plurality of test requests corresponding to the test task. A related test data extraction unit is configured to extract related test data from the virtual test data according to the plurality of test requests. A test requirement data extraction unit is configured to extract test requirement data from the related test data according to a plurality of preset requirement data types.

Images