Cloud computing platform evaluation system based on data analysis

Abstract

The invention discloses a cloud computing platform evaluation system based on data analysis, which belongs to the field of platform evaluation systems and comprises a computing performance evaluation module, a failure rate evaluation module, a storage performance evaluation module, a recovery capability evaluation module, a network performance evaluation module and a visual early warning module. The computing performance evaluation module is used for evaluating the computing performance of the cloud computing platform and sending an evaluation result to the visual early warning module; the fault rate evaluation module is used for evaluating the fault rate of the cloud computing platform and sending an evaluation result to the visual early warning module; and the storage performance evaluation module is used for evaluating the storage performance of the cloud computing platform and sending an evaluation result to the visual early warning module. According to the invention, multi-angle and all-around evaluation can be carried out on the cloud computing platform from computing performance, failure rate, storage performance, recovery capability and network performance, so that the overall performance of the cloud computing platform can be accurately evaluated.

Description

Technical Field

[0001] The present invention relates to the field of platform evaluation systems, and in particular to a cloud computing platform evaluation system based on data analysis. Background Art

[0002] Cloud computing platform refers to services based on hardware resources and software resources. It uses virtualization technology to encapsulate computing resources, storage resources and network resources into an independent virtual environment, and is designed to provide enterprise users and developers with fast deployment, pay-as-you-go, and elastically scalable IT services. With the continuous development of Internet technology, cloud computing platform has become an important tool for enterprises and organizations to handle large-scale data processing and storage.

[0003] In order to ensure that the performance of the cloud computing platform is maintained at a normal level during use, it is necessary to evaluate the various performance of the cloud computing platform in real time. The current common evaluation method is to evaluate the failure rate and storage performance of the cloud computing platform, which is not comprehensive enough for the overall performance evaluation of the cloud computing platform and cannot accurately evaluate the overall performance of the cloud computing platform.

[0004] Therefore, those skilled in the art provide a cloud computing platform evaluation system based on data analysis to solve the problems raised in the above background technology. Summary of the invention

[0005] The purpose of the present invention is to provide a cloud computing platform evaluation system based on data analysis, which can conduct a multi-angle and comprehensive evaluation of the cloud computing platform from the perspectives of computing performance, failure rate, storage performance, recovery capability, and network performance, so as to accurately evaluate the overall performance of the cloud computing platform to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A cloud computing platform evaluation system based on data analysis, including a computing performance evaluation module, a failure rate evaluation module, a storage performance evaluation module, a recovery capability evaluation module, a network performance evaluation module and a visual early warning module;

[0008] The computing performance evaluation module is used to evaluate the computing performance of the cloud computing platform and send the evaluation results to the visual warning module; the failure rate evaluation module is used to evaluate the failure rate of the cloud computing platform and send the evaluation results to the visual warning module; the storage performance evaluation module is used to evaluate the storage performance of the cloud computing platform and send the evaluation results to the visual warning module; the recovery capability evaluation module is used to evaluate the recovery capability of the cloud computing platform and send the evaluation results to the visual warning module; the network performance evaluation module is used to evaluate the network performance of the cloud computing platform and send the evaluation results to the visual warning module; the visual warning module is used to receive all evaluation results and display them visually, and issue a warning when an abnormality occurs in the evaluation results.

[0009] As a further solution of the present invention: the specific evaluation process of the computing performance of the cloud computing platform is:

[0010] When a user sends a request to the cloud computing platform, the real-time response time of the cloud computing platform in the response phase is collected and marked as T 响应 ;

[0011] At time T 响应 The average CPU utilization of the cloud computing platform is collected and marked as A, the average memory utilization of the cloud computing platform is collected and marked as B, the average database resource utilization of the cloud computing platform is collected and marked as C, the average virtualization resource utilization of the cloud computing platform is collected and marked as D, and the average IO utilization of the cloud computing platform is collected and marked as E;

[0012] Generate a real-time radar chart P in the response phase based on CPU utilization A, memory utilization B, database resource utilization C, virtualization resource utilization D, and IO utilization E 实时 ;

[0013] The average CPU utilization, average memory utilization, average database resource utilization, average virtualization resource utilization, and average IO utilization of the cloud computing platform in each previous response stage are retrieved, and then a radar chart of the cloud computing platform in the previous response stage is generated. 以往 i, i=1···n, where n is a positive integer;

[0014] P 实时 With P 以往 i performs similarity comparison, if P 以往 i and P 实时 If the similarity is greater than the first preset value, then P 以往 i plans to the first sequence, if P 以往 i and P 实时 If the similarity is less than or equal to the first preset value, then P 以往 i Planning to the second sequence;

[0015] The first sequence P 以往 The previous response stages corresponding to i are screened out, and then the cloud computing platform response time corresponding to the screened previous response stages is extracted, marked as Ti, i = 1···k, where k is a positive integer and k≤n;

[0016] T 响应 Match with Ti, if T 响应 If the absolute value of the difference between any Ti is greater than the second preset value, it means that the computing performance of the cloud computing platform is abnormal. 响应 If the absolute value of the difference with any Ti is less than or equal to the second preset value, it means that the computing performance of the cloud computing platform is normal at this time.

[0017] As a further solution of the present invention: the specific evaluation process of the failure rate of the cloud computing platform is:

[0018] Divide a day into eight time periods, each of which is three hours long and are marked as ti, i = 1···8;

[0019] Collect the number of failures that occur on the cloud computing platform in time period ti and mark it as X, and then collect the duration of the failures that occur on the cloud computing platform in time period ti and mark it as Y;

[0020] Generate a bar chart of each time period ti based on the number of faults X and the duration of the faults Y;

[0021] Calculate the sum of the bar chart areas of each time period ti and mark it as Z, and then mark the sum of the preset bar chart areas as z;

[0022] The failure rate V=(Z / z) of the cloud computing platform is calculated. If V is less than the third preset value, it means that the failure rate of the cloud computing platform is normal. If V is greater than or equal to the third preset value, it means that the failure rate of the cloud computing platform is abnormal.

[0023] As a further solution of the present invention: the specific evaluation process of the storage performance of the cloud computing platform is:

[0024] When storing data on the cloud computing platform, the read and write speed of the data storage is collected and marked as F, the throughput of the data storage is collected and marked as G, and the consistency of the data after the data storage is marked as S;

[0025] Generate a radar chart Q based on the read / write speed F, throughput G, and data consistency S. The radar chart Q is a triangle with three angles f, g, and s.

[0026] Obtain radar charts corresponding to the read / write speed, throughput, and data consistency of the cloud computing platform in the past when storing data. Plan the radar charts with read / write speed equal to F to the first cluster, plan the radar charts with throughput equal to G to the second cluster, and plan the radar charts with data consistency equal to S to the third cluster. The same radar chart can be planned to different clusters at the same time.

[0027] Label the radar chart in the first cluster as qi, label the radar chart in the second cluster as qqi, label the radar chart in the third cluster as qqqi, i=1···n;

[0028] The graphic area of ​​Q is compared with the graphic areas of the radar charts in the first cluster, the second cluster, and the third cluster, respectively, to evaluate whether the storage performance of the cloud computing platform is abnormal.

[0029] As a further solution of the present invention: the specific process of evaluating whether the storage performance of the cloud computing platform is abnormal is:

[0030] Compare the graphic area of ​​Q with the graphic areas of each qi, count the number of qi in the first cluster whose graphic area is larger than the graphic area of ​​Q, and mark them as Δq1. If Δq1 is larger than a fourth preset value, it indicates that the storage performance of the cloud computing platform is abnormal. If Δq1 is less than or equal to the fourth preset value, proceed to the next step.

[0031] Compare the graphic area of ​​Q with the graphic areas of each qqi, count the number of qqi in the second cluster whose graphic area is larger than the graphic area of ​​Q, and mark them as Δq2. If Δq2 is larger than the fifth preset value, it indicates that the storage performance of the cloud computing platform is abnormal. If Δq2 is less than or equal to the fifth preset value, proceed to the next step.

[0032] Compare the graphic area of ​​Q with the graphic areas of each qqqi, count the number of qqqi in the third cluster whose graphic area is larger than the graphic area of ​​Q, and mark them as Δq3. If Δq3 is greater than the fourth preset value, it means that the storage performance of the cloud computing platform is abnormal. If Δq3 is less than or equal to the fourth preset value, it means that the storage performance of the cloud computing platform is normal.

[0033] As a further solution of the present invention: the specific evaluation process of the recovery capability of the cloud computing platform is:

[0034] The real-time failure of the cloud computing platform is marked as W, and the recovery speed after the failure is marked as R, and the recovery degree is marked as U;

[0035] Extract the records of the same type of fault as W from the previous fault records of the cloud computing platform, and then use them to obtain the recovery speed ri and recovery degree ui after each fault in the extracted fault records, i = 1···n;

[0036] Match R and U with ri and ui respectively. If any set of ri and ui matches R and U, it means that the recovery capability of the cloud computing platform is normal. Conversely, if any set of ri and ui does not match R and U, it means that the recovery capability of the cloud computing platform is abnormal.

[0037] As a further solution of the present invention: the specific evaluation process of the network performance of the cloud computing platform is:

[0038] When data is transmitted on the cloud computing platform, the collected network bandwidth is marked as J, the collected delay parameter is marked as M, and the collected packet loss rate is marked as L;

[0039] Compare J with the preset value j, M with the preset value m, and L with the preset value l;

[0040] If the comparison results show any two of J>j, M<m, L<l, it means that the network performance of the cloud computing platform is normal. If the comparison results show any two of J≤j, M≥m, L≥l, it means that the network performance of the cloud computing platform is abnormal.

[0041] As a further solution of the present invention: the visual warning module includes a display screen, and the display interface of the display screen is divided into five areas, which are used to display the evaluation results of the computing performance evaluation module, the failure rate evaluation module, the storage performance evaluation module, the recovery capability evaluation module, and the network performance evaluation module respectively. If the evaluation result is abnormal, the edge of the corresponding display interface area flashes red.

[0042] Compared with the prior art, the present invention has the following beneficial effects:

[0043] 1. This application can conduct a comprehensive and multi-angle evaluation of the cloud computing platform from the perspectives of computing performance, failure rate, storage performance, recovery capability, and network performance, so as to accurately evaluate the overall performance of the cloud computing platform. When evaluating computing performance, more representative data is first selected from CPU utilization, memory utilization, database resource utilization, virtualization resource utilization, and IO utilization, and then the response time of the cloud computing platform is evaluated to determine the abnormality.

[0044] 2. This application uses the statistics of the number of failures and failure times that occur during the operation of the platform to quickly and accurately assess whether the failure rate is abnormal by combining the constructed bar chart.

[0045] 3. When evaluating the storage performance of the cloud computing platform, the present application uses the three corners of the constructed triangular radar chart as reference points to select three comparison clusters respectively, and then uses this to quickly evaluate whether the storage performance of the cloud computing platform is abnormal. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 This is a structural diagram of a cloud computing platform evaluation system based on data analysis;

[0047] Figure 2 A schematic diagram of a bar chart in a cloud computing platform evaluation system based on data analysis;

[0048] Figure 3 A schematic diagram of a radar chart Q in a cloud computing platform evaluation system based on data analysis. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] As mentioned in the background technology of this application, research has found that the common existing evaluation method is to evaluate the failure rate and storage performance of the cloud computing platform, which is not comprehensive enough for the overall performance evaluation of the cloud computing platform, and cannot accurately evaluate the overall performance of the cloud computing platform, and has certain defects.

[0051] In order to solve the above-mentioned defects, the present application discloses a cloud computing platform evaluation system based on data analysis, which can conduct a multi-angle and comprehensive evaluation of the cloud computing platform from the aspects of computing performance, failure rate, storage performance, recovery capability, and network performance, so as to accurately evaluate the overall performance of the cloud computing platform.

[0052] The following will describe in detail how the solution of the present application solves the above technical problems in conjunction with the accompanying drawings.

[0053] See also Figure 1In an embodiment of the present invention, a cloud computing platform evaluation system based on data analysis includes a computing performance evaluation module, a failure rate evaluation module, a storage performance evaluation module, a recovery capability evaluation module, a network performance evaluation module and a visual warning module; the computing performance evaluation module is used to evaluate the computing performance of the cloud computing platform and send the evaluation result to the visual warning module; the failure rate evaluation module is used to evaluate the failure rate of the cloud computing platform and send the evaluation result to the visual warning module; the storage performance evaluation module is used to evaluate the storage performance of the cloud computing platform and send the evaluation result to the visual warning module; the recovery capability evaluation module is used to evaluate the recovery capability of the cloud computing platform and send the evaluation result to the visual warning module; the network performance evaluation module is used to evaluate the network performance of the cloud computing platform and send the evaluation result to the visual warning module; the visual warning module is used to receive all evaluation results and display them visually, and issue an early warning when an abnormality occurs in the evaluation result. The present application can evaluate the cloud computing platform from multiple angles and in an all-round manner from the perspectives of computing performance, failure rate, storage performance, recovery capability and network performance, so as to accurately evaluate the overall performance of the cloud computing platform.

[0054] In this embodiment, the specific evaluation process of the computing performance of the cloud computing platform is as follows: when a user sends a request to the cloud computing platform, the real-time response time of the cloud computing platform in the response phase is collected and marked as T 响应 ; At time T 响应 The average CPU utilization of the cloud computing platform is collected and marked as A, the average memory utilization of the cloud computing platform is collected and marked as B, the average database resource utilization of the cloud computing platform is collected and marked as C, the average virtualization resource utilization of the cloud computing platform is collected and marked as D, and the average IO utilization of the cloud computing platform is collected and marked as E; based on CPU utilization A, memory utilization B, database resource utilization C, virtualization resource utilization D, IO utilization E, a real-time radar chart P is generated in the response stage 实时 ; retrieve the average CPU utilization, average memory utilization, average database resource utilization, average virtualization resource utilization, and average IO utilization of the cloud computing platform in each previous response stage, and then generate a radar chart of the cloud computing platform in the previous response stage. 以往 i, i = 1···n, where n is a positive integer; P 实时 With P 以往 i performs similarity comparison, if P 以往 i and P 实时 If the similarity is greater than the first preset value, then P 以往 i plans to the first sequence, if P 以往 i and P 实时 If the similarity is less than or equal to the first preset value, then P 以往 i is planned to the second sequence; P in the first sequence以往 The previous response stages corresponding to i are screened out, and then the cloud computing platform response time corresponding to the screened previous response stages is extracted, marked as Ti, i = 1···k, where k is a positive integer and k≤n; T 响应 Match with Ti, if T 响应 If the absolute value of the difference between any Ti is greater than the second preset value, it means that the computing performance of the cloud computing platform is abnormal. 响应 If the absolute value of the difference with any Ti is less than or equal to the second preset value, it means that the computing performance of the cloud computing platform is normal at this time. CPU utilization is a key indicator for measuring CPU (central processing unit) workload. It indicates the proportion of time that the CPU is occupied by the program to execute instructions in a specific time period. The level of CPU utilization directly affects the response speed and performance of the system. Memory utilization refers to the proportion of the amount of used memory in the system to the total amount of memory. It is an important indicator for measuring the efficiency of system memory usage and is of great significance for system performance tuning and resource management. Database resource utilization refers to the effective utilization of the allocated resources by the database system. It covers multiple indicators, which jointly reflect the performance and health of the database system. Common indicators include disk utilization and connection utilization, cache hit rate, and read and write page utilization. Virtualization resource utilization is an indicator for measuring the effective utilization of physical resources (such as CPU, memory, storage, etc.) in a virtualization environment. In virtualization technology, a physical server is divided into multiple virtual machines, each of which can independently run different operating systems and applications and share physical resources. The level of virtualization resource utilization directly affects the overall performance and resource cost of the system. IO utilization is a key indicator for measuring the workload of the input and output system. It indicates the proportion of the IO system resources actually used in a specific period of time. The IO system includes devices or components such as disks and networks that are responsible for data input and output.

[0055] In this embodiment, the specific evaluation process of the failure rate of the cloud computing platform is as follows: a day is evenly divided into eight time periods, each of which is three hours, and each is marked as ti, i=1...8; the number of failures that occur in the cloud computing platform in the time period ti is collected and marked as X, and then the duration of the failure of the cloud computing platform in the time period ti is collected and marked as Y; a bar chart of each time period ti is generated based on the number of failures X and the duration of the failure Y, such as Figure 2As shown, the greater the width of the bar graph, the greater the duration Y of the corresponding fault; the sum of the bar graph areas of each time period ti is calculated and marked as Z, and then the sum of the preset bar graph areas is marked as z; the failure rate V of the cloud computing platform is calculated as (Z / z), if V is less than the third preset value, it means that the failure rate of the cloud computing platform is normal, if V is greater than or equal to the third preset value, it means that the failure rate of the cloud computing platform is abnormal. By counting the number of failures and failure time that occur during the operation of the platform, it is evaluated whether its failure rate is abnormal.

[0056] In this embodiment, the specific evaluation process of the storage performance of the cloud computing platform is as follows: when the cloud computing platform performs data storage, the read and write speed of the data storage is collected and marked as F, the throughput of the data storage is collected and marked as G, and then the consistency of the data after the data storage is marked as S; based on the read and write speed F, the throughput G, and the data consistency S, a radar chart Q is generated. The radar chart Q is a triangle with three angles f, g, and s, respectively. Figure 3 As shown; obtain the radar chart corresponding to the read and write speed, throughput, and data consistency of the cloud computing platform in the past when storing data, and plan the radar chart with the read and write speed equal to F to the first cluster, the radar chart with the throughput equal to G to the second cluster, and the radar chart with the data consistency equal to S to the third cluster. The same radar chart can be planned to different clusters at the same time; mark the radar chart in the first cluster as qi, the radar chart in the second cluster as qqi, and the radar chart in the third cluster as qqqi, i=1···n; compare the graphic area of ​​Q with the graphic area of ​​the radar chart in the first cluster, the second cluster, and the third cluster, respectively, to evaluate whether the storage performance of the cloud computing platform is abnormal. When evaluating the storage performance of the cloud computing platform, the three corners of the constructed triangular radar chart are used as reference points to select three comparison clusters respectively, and then quickly evaluate whether the storage performance of the cloud computing platform is abnormal. Radar charts are also called spider charts, star charts, spider web charts, irregular polygons, polar coordinate charts, or Kiviat charts. They are a graphical method for displaying multivariate data. It starts from a central point and extends multiple rays outward from the central point. Each ray represents a specific variable or indicator, and the points or line segments on the rays represent the values ​​or scores of the variable in different dimensions.

[0057] In this embodiment, the specific process of evaluating whether the storage performance of the cloud computing platform is abnormal is as follows: comparing the graphic area of ​​Q with the graphic area of ​​each qi, counting the number of qi in the first cluster whose graphic area is larger than the graphic area of ​​Q, and marking them as Δq1. If Δq1 is larger than the fourth preset value, it means that the storage performance of the cloud computing platform is abnormal. If Δq1 is less than or equal to the fourth preset value, proceed to the next step; comparing the graphic area of ​​Q with the graphic area of ​​each qqi, counting the number of qqi in the second cluster whose graphic area is larger than the graphic area of ​​Q, and marking them as Δq2. If Δq2 is larger than the fifth preset value, it means that the storage performance of the cloud computing platform is abnormal. If Δq2 is less than or equal to the fifth preset value, proceed to the next step; comparing the graphic area of ​​Q with the graphic area of ​​each qqqi, counting the number of qqqi in the third cluster whose graphic area is larger than the graphic area of ​​Q, and marking them as Δq3. If Δq3 is larger than the fourth preset value, it means that the storage performance of the cloud computing platform is abnormal. If Δq3 is less than or equal to the fourth preset value, it means that the storage performance of the cloud computing platform is normal. This setting can quickly assess whether the storage performance of the cloud computing platform is abnormal.

[0058] In this embodiment, the specific evaluation process of the recovery capability of the cloud computing platform is as follows: mark the real-time fault of the cloud computing platform as W, then mark the recovery speed after this fault as R, and the recovery degree as U; extract the records of the same type of fault as W from the previous fault records of the cloud computing platform, and then obtain the recovery speed ri and recovery degree ui after each fault in the extracted fault records, i=1···n; match R and U with ri and ui respectively. If any set of ri and ui matches R and U, it means that the recovery capability of the cloud computing platform is normal. On the contrary, if any set of ri and ui does not match R and U, it means that the recovery capability of the cloud computing platform is abnormal. This setting can quickly determine whether the recovery capability of the cloud computing platform is abnormal.

[0059] In this embodiment, the specific evaluation process of the network performance of the cloud computing platform is as follows: when the cloud computing platform performs data transmission, the collected network bandwidth is marked as J, the collected delay parameter is marked as M, and the collected packet loss rate is marked as L; J is compared with the preset value j, M is compared with the preset value m, and L is compared with the preset value l; if the comparison result shows any two of J>j, M<m, and L<l, it means that the network performance of the cloud computing platform is normal; if the comparison result shows any two of J≤j, M≥m, and L≥l, it means that the network performance of the cloud computing platform is abnormal. This setting can quickly determine whether the network performance of the cloud computing platform is abnormal.

[0060] In this embodiment, the visual warning module includes a display screen, and the display interface of the display screen is divided into five areas, which are used to display the evaluation results of the computing performance evaluation module, the failure rate evaluation module, the storage performance evaluation module, the recovery capability evaluation module, and the network performance evaluation module, respectively. If the evaluation result is abnormal, the edge of the corresponding display interface area flashes red. This setting is convenient for showing the evaluation results clearly and intuitively to the staff, and is convenient for the staff to quickly identify when an abnormality occurs.

[0061] The present invention can evaluate the cloud computing platform from multiple angles and in an all-round manner from the perspectives of computing performance, failure rate, storage performance, recovery capability, and network performance, thereby accurately evaluating the overall performance of the cloud computing platform. When evaluating computing performance, more representative data is first screened out from CPU utilization, memory utilization, database resource utilization, virtualization resource utilization, and IO utilization, and then the response time of the cloud computing platform is evaluated to be abnormal. The present application uses the statistical data of the number of failures and failure times that occur during the operation of the platform, combined with the constructed bar chart, to quickly and accurately evaluate whether its failure rate is abnormal. When evaluating the storage performance of the cloud computing platform, the present application uses the three corners of the constructed triangular radar chart as reference points to screen out three comparison clusters respectively, and then quickly evaluates whether the storage performance of the cloud computing platform is abnormal.

[0062] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

[0063] What is described above is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A cloud computing platform evaluation system based on data analysis, characterized in that: It includes computing performance evaluation module, failure rate evaluation module, storage performance evaluation module, recovery capability evaluation module, network performance evaluation module and visual early warning module; The computing performance evaluation module is used to evaluate the computing performance of the cloud computing platform and send the evaluation results to the visual warning module; The failure rate evaluation module is used to evaluate the failure rate of the cloud computing platform and send the evaluation result to the visual warning module; The storage performance evaluation module is used to evaluate the storage performance of the cloud computing platform and send the evaluation results to the visual warning module; The recovery capability assessment module is used to assess the recovery capability of the cloud computing platform and send the assessment result to the visual warning module; The network performance evaluation module is used to evaluate the network performance of the cloud computing platform and send the evaluation results to the visual early warning module; The visual warning module is used to receive all evaluation results and display them visually, and to issue a warning when an evaluation result is abnormal.

2. A cloud computing platform evaluation system based on data analysis according to claim 1, characterized in that: The specific evaluation process of the computing performance of the cloud computing platform is as follows: When a user sends a request to the cloud computing platform, the real-time response time of the cloud computing platform in the response phase is collected and marked as T 响应 ; At time T 响应 The average CPU utilization of the cloud computing platform is collected and marked as A, the average memory utilization of the cloud computing platform is collected and marked as B, the average database resource utilization of the cloud computing platform is collected and marked as C, the average virtualization resource utilization of the cloud computing platform is collected and marked as D, and the average IO utilization of the cloud computing platform is collected and marked as E; Generate a real-time radar chart P in the response phase based on CPU utilization A, memory utilization B, database resource utilization C, virtualization resource utilization D, and IO utilization E 实时 ; The average CPU utilization, average memory utilization, average database resource utilization, average virtualization resource utilization, and average IO utilization of the cloud computing platform in each previous response stage are retrieved, and then a radar chart of the cloud computing platform in the previous response stage is generated. 以往 i, i=1···n, where n is a positive integer; P 实时 With P 以往 i performs similarity comparison, if P 以往 i and P 实时 If the similarity is greater than the first preset value, then P 以往 i plans to the first sequence, if P 以往 i and P 实时 If the similarity is less than or equal to the first preset value, then P 以往 i Planning to the second sequence; The first sequence P 以往 The previous response stages corresponding to i are screened out, and then the cloud computing platform response time corresponding to the screened previous response stages is extracted, marked as Ti, i = 1···k, where k is a positive integer and k≤n; T 响应 Match with Ti, if T 响应 If the absolute value of the difference between any Ti is greater than the second preset value, it means that the computing performance of the cloud computing platform is abnormal. 响应 If the absolute value of the difference with any Ti is less than or equal to the second preset value, it means that the computing performance of the cloud computing platform is normal at this time.

3. A cloud computing platform evaluation system based on data analysis according to claim 2, characterized in that: The specific evaluation process of the failure rate of the cloud computing platform is as follows: Divide a day into eight time periods, each of which is three hours long and are marked as ti, i = 1···8; Collect the number of failures that occur on the cloud computing platform in time period ti and mark it as X, and then collect the duration of the failures that occur on the cloud computing platform in time period ti and mark it as Y; Generate a bar chart of each time period ti based on the number of faults X and the duration of the faults Y; Calculate the sum of the bar chart areas of each time period ti and mark it as Z, and then mark the sum of the preset bar chart areas as z; The failure rate V=(Z / z) of the cloud computing platform is calculated. If V is less than the third preset value, it means that the failure rate of the cloud computing platform is normal. If V is greater than or equal to the third preset value, it means that the failure rate of the cloud computing platform is abnormal.

4. The cloud computing platform evaluation system based on data analysis according to claim 3, characterized in that: The specific evaluation process of the storage performance of the cloud computing platform is as follows: When storing data on the cloud computing platform, the read and write speed of the data storage is collected and marked as F, the throughput of the data storage is collected and marked as G, and the consistency of the data after the data storage is marked as S; Generate a radar chart Q based on the read / write speed F, throughput G, and data consistency S. The radar chart Q is a triangle with three angles f, g, and s. Obtain radar charts corresponding to the read / write speed, throughput, and data consistency of the cloud computing platform in the past when storing data. Plan the radar charts with read / write speed equal to F to the first cluster, plan the radar charts with throughput equal to G to the second cluster, and plan the radar charts with data consistency equal to S to the third cluster. The same radar chart can be planned to different clusters at the same time. Label the radar chart in the first cluster as qi, label the radar chart in the second cluster as qqi, label the radar chart in the third cluster as qqqi, i=1···n; The graphic area of ​​Q is compared with the graphic areas of the radar charts in the first cluster, the second cluster, and the third cluster, respectively, to evaluate whether the storage performance of the cloud computing platform is abnormal.

5. The cloud computing platform evaluation system based on data analysis according to claim 4, characterized in that: The specific process of evaluating whether the storage performance of the cloud computing platform is abnormal is as follows: Compare the graphic area of ​​Q with the graphic areas of each qi, count the number of qi in the first cluster whose graphic area is larger than the graphic area of ​​Q, and mark them as Δq1. If Δq1 is larger than a fourth preset value, it indicates that the storage performance of the cloud computing platform is abnormal. If Δq1 is less than or equal to the fourth preset value, proceed to the next step. Compare the graphic area of ​​Q with the graphic areas of each qqi, count the number of qqi in the second cluster whose graphic area is larger than the graphic area of ​​Q, and mark them as Δq2. If Δq2 is larger than the fifth preset value, it indicates that the storage performance of the cloud computing platform is abnormal. If Δq2 is less than or equal to the fifth preset value, proceed to the next step. Compare the graphic area of ​​Q with the graphic areas of each qqqi, count the number of qqqi in the third cluster whose graphic area is larger than the graphic area of ​​Q, and mark them as Δq3. If Δq3 is greater than the fourth preset value, it means that the storage performance of the cloud computing platform is abnormal. If Δq3 is less than or equal to the fourth preset value, it means that the storage performance of the cloud computing platform is normal.

6. The cloud computing platform evaluation system based on data analysis according to claim 5, characterized in that: The specific evaluation process of the recovery capability of the cloud computing platform is as follows: The real-time failure of the cloud computing platform is marked as W, and the recovery speed after the failure is marked as R, and the recovery degree is marked as U; Extract the records of the same type of fault as W from the previous fault records of the cloud computing platform, and then use them to obtain the recovery speed ri and recovery degree ui after each fault in the extracted fault records, i = 1···n; Match R and U with ri and ui respectively. If any set of ri and ui matches R and U, it means that the recovery capability of the cloud computing platform is normal. Conversely, if any set of ri and ui does not match R and U, it means that the recovery capability of the cloud computing platform is abnormal.

7. The cloud computing platform evaluation system based on data analysis according to claim 6, characterized in that: The specific evaluation process of the network performance of the cloud computing platform is as follows: When data is transmitted on the cloud computing platform, the collected network bandwidth is marked as J, the collected delay parameter is marked as M, and the collected packet loss rate is marked as L; Compare J with the preset value j, M with the preset value m, and L with the preset value l; If the comparison results show any two of J>j, M<m, L<l, it means that the network performance of the cloud computing platform is normal. If the comparison results show any two of J≤j, M≥m, L≥l, it means that the network performance of the cloud computing platform is abnormal.

8. The cloud computing platform evaluation system based on data analysis according to claim 7, characterized in that: The visual warning module includes a display screen, and the display interface of the display screen is divided into five areas, which are used to display the computing performance evaluation module, the failure rate evaluation module, the storage performance evaluation module, The evaluation results of the recovery capability evaluation module and the network performance evaluation module. If the evaluation results are abnormal, The corresponding edge of the display interface area flashes red.

Images