Cloud service resource monitoring system and method

Abstract

The present invention provides a cloud service resource supervision system and method, including: a cloud resource management layer, a monitoring data management layer and a cloud service evaluation layer; the cloud resource management layer is used to determine the cloud resources corresponding to each cloud service and perform data monitoring on each cloud resource; the monitoring data management layer is used to obtain monitoring data of a number of cloud resources from the cloud resource management layer, generate and store a first data summary of the monitoring data; the cloud service evaluation layer is used to determine the reputation score of the cloud service according to the monitoring data, and supervise the cloud service according to the reputation score. The present invention supervises the heterogeneous resources of multiple cloud service providers by constructing a cloud resource management layer, a monitoring data management layer and a cloud service evaluation layer, thereby saving the cost of human supervision, improving the efficiency of supervision and evaluation, and realizing the trusted supervision of cloud computing services.

Description

Technical Field

[0001] The present invention relates to the field of cloud computing technology, and in particular to a cloud service resource supervision system and method. Background Art

[0002] In recent years, with the continuous development of cloud computing and the increasing demand for diversified cloud services, the concept of cloud computing has emerged. As a new generation of cloud computing model, cloud computing emphasizes multi-cloud open collaboration and resource integration and sharing among cloud service providers, providing users with flexible, cost-effective and customizable cloud services that can meet large-scale and diversified service needs.

[0003] However, cloud computing faces the challenges of supervision and governance. On the one hand, a reasonable evaluation of cloud service resources in cloud scenarios is a prerequisite for achieving supervision and governance. At present, the evaluation methods for cloud service resources include evaluation methods based on QoS (Quality of Service), etc. However, in the multi-dimensional and heterogeneous cloud computing scenarios, cloud services are diverse and user needs are changing. The above methods are not suitable for such complex and changing scenarios.

[0004] On the other hand, in the cloud computing scenario, there is also a competitive relationship between the cooperating cloud service providers. In order to avoid confusion, a reasonable scoring mechanism is needed as a reference for unified supervision and governance. At the same time, in order to avoid disputes, an implementable method is needed to ensure the authenticity and traceability of regulatory data, thereby achieving credible supervision and governance. Summary of the invention

[0005] The present invention provides a cloud service resource supervision system and method, which are used to solve the defect of the prior art that there is a lack of supervision methods for multiple competing cloud service providers, and to realize a cloud service supervision system and method that is highly authentic, traceable and quantifiable.

[0006] The present invention provides a cloud service resource supervision system, comprising: a cloud resource management layer, a monitoring data management layer and a cloud service evaluation layer;

[0007] The cloud resource management layer is used to determine the cloud resources corresponding to each cloud service and to perform data monitoring on each cloud resource;

[0008] The monitoring data management layer is used to obtain monitoring data of a plurality of cloud resources from the cloud resource management layer, and generate and store a first data summary of the monitoring data;

[0009] The cloud service evaluation layer is used to retrieve the monitoring data stored in the monitoring data management layer and generate a second data summary of the monitoring data. When the second data summary is the same as the first data summary, the reputation score of the cloud service is determined according to the monitoring data, and the cloud service is supervised according to the reputation score.

[0010] According to a cloud service resource supervision system provided by the present invention, the monitoring data includes basic resource information data, operation status data and abnormal event data of the cloud resources.

[0011] According to a cloud service resource supervision system provided by the present invention, the monitoring data management layer further includes a blockchain, and the monitoring data management layer is specifically used for:

[0012] Acquire monitoring data of a plurality of cloud resources from the cloud resource management layer, aggregate the monitoring data of each cloud resource according to the type of the monitoring data, generate a first data summary of the aggregated monitoring data, and store the first data summary in the blockchain.

[0013] According to a cloud service resource supervision system provided by the present invention, the monitoring data management layer is specifically used for:

[0014] Grouping the aggregated monitoring data according to a first time window, calculating a first data summary of each group of monitoring data and storing the summary in the blockchain;

[0015] The cloud service evaluation layer is specifically used to obtain the monitoring data in the second time window, and after grouping the monitoring data in the second time window based on the first time window, calculate a second data summary of each group of monitoring data.

[0016] According to a cloud service resource supervision system provided by the present invention, the blockchain is also used for:

[0017] Storing service level agreement data sent by the cloud resource management layer, wherein the service level agreement data represents a service agreement corresponding to the cloud service;

[0018] In response to the calling instruction of the cloud service evaluation layer, the service level agreement data of the inter-cloud service in the calling instruction is sent to the cloud service evaluation layer.

[0019] According to a cloud service resource supervision system provided by the present invention, the monitoring data management layer further includes a time series database, and the monitoring data management layer is specifically used for:

[0020] The aggregated monitoring data is stored in the time series database, and a plurality of copies of the aggregated monitoring data are generated;

[0021] The cloud service evaluation layer is specifically used to, when the second data summary is different from the first data summary, read the replica data in the time series database and calculate a third data summary of the replica data;

[0022] In a case where the third data summary is identical to the first data summary, a reputation score of the cloud service is determined according to the copy data, and the cloud service is supervised according to the reputation score.

[0023] According to a cloud service resource supervision system provided by the present invention, the cloud service evaluation layer further includes an evaluation module, and the evaluation module is used to:

[0024] Determine a service quality indicator score for any of the cloud services based on the monitoring data;

[0025] Determine a security score of any of the cloud services based on the abnormal event data and the operating status data;

[0026] Determining a performance score of any of the cloud services based on the service level agreement data;

[0027] A weighted sum of the service quality index score, the security score and the fulfillment score is used as the reputation score of any of the cloud services.

[0028] The present invention also provides a cloud service resource supervision method, comprising:

[0029] Determine the cloud resources corresponding to each cloud service through the cloud resource management layer, and perform data monitoring on each of the cloud resources;

[0030] Acquire monitoring data of a plurality of cloud resources from the cloud resource management layer through the monitoring data management layer, and generate and store a first data summary of the monitoring data;

[0031] The monitoring data stored in the monitoring data management layer is retrieved through the cloud service evaluation layer and a second data summary of the monitoring data is generated. When the second data summary is the same as the first data summary, the reputation score of the cloud service is determined according to the monitoring data, and the cloud service is supervised according to the reputation score.

[0032] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the cloud service resource supervision method as described in any one of the above is implemented.

[0033] The present invention also provides a non-transitory computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the cloud service resource supervision method described in any one of the above is implemented.

[0034] The present invention also provides a computer program product, including a computer program, which, when executed by a processor, implements any of the above-mentioned cloud service resource supervision methods.

[0035] The cloud service resource supervision system and method provided by the present invention, by constructing a cloud resource management layer, a monitoring data management layer and a cloud service evaluation layer, monitors and manages the protocol data and monitoring data generated by users and cloud service providers from the signing of a service level agreement to the service implementation process, implements quantitative scoring of cloud services through protocol data and monitoring data, and supervises heterogeneous resources of multiple cloud service providers based on the scoring, thereby saving the cost of human supervision, improving the efficiency of supervision and evaluation, and realizing trusted supervision of cloud computing services. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0037] Figure 1 It is a structural diagram of the cloud service resource supervision system provided by the present invention;

[0038] Figure 2 It is a flow chart of the cloud service resource supervision method provided by the present invention;

[0039] Figure 3 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are 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.

[0041] Combine the following Figure 1 The cloud service resource supervision system of the present invention is described, comprising: a cloud resource management layer, a monitoring data management layer and a cloud service evaluation layer;

[0042] The cloud resource management layer is used to determine the cloud resources corresponding to each cloud service and to perform data monitoring on each cloud resource;

[0043] like Figure 1 As shown, the cloud resource management layer is a cloud resource platform that implements basic functions, including multiple different cloud resources, and is responsible for monitoring the status of each cloud resource and the signing of the cloud service level agreement of the cloud service provider.

[0044] Among them, different cloud service providers provide different cloud services after signing service level agreements, and each cloud service uses different cloud resources. The cloud resources required by cloud services are different due to the different types and contents of cloud services.

[0045] Optionally, the cloud resource management layer allocates corresponding cloud resources to each cloud service according to the category of the cloud service, the signed service level agreement, and the occupancy of the cloud resources.

[0046] The cloud resource management layer obtains the monitoring data of each cloud resource by monitoring each cloud resource, and then obtains the monitoring data of all cloud resources corresponding to each cloud service to determine the monitoring data of the cloud service, which serves as the basis for quantitative scoring of the cloud service.

[0047] Furthermore, the cloud resource management layer also provides a standardized monitoring data query interface to the upper platform, which is the monitoring data management layer in the present invention, so that the upper platform can obtain the monitoring data of each cloud service in the cloud resource management layer and score the cloud service. When receiving a data query request from the monitoring data management layer, the current monitoring data of the cloud resources is obtained and returned through the interface after statistics.

[0048] At the same time, the cloud resource management layer also provides the upper-level platform with the agreement data of the cloud service agreement signed between the user and the cloud service provider.

[0049] Optionally, the agreement data at least includes SLA (Service Level Agreement) data. The SLA data characterizes the performance, availability, support and penalty clauses of the cloud service that the cloud service provider needs to provide to the user.

[0050] The monitoring data management layer is used to obtain monitoring data of a plurality of cloud resources from the cloud resource management layer, and generate and store a first data summary of the monitoring data;

[0051] like Figure 1 As shown in FIG, the monitoring data management layer is the upper layer platform of the cloud resource management layer, which is responsible for statistics, aggregation, analysis and storage of monitoring data, and provides a query interface to the upper layer platform.

[0052] The monitoring data represents the cloud resources of each cloud service and the service status of the cloud resources.

[0053] The monitoring data management layer regularly obtains monitoring data of cloud resources through the standard interface provided by the cloud resource management layer.

[0054] Optionally, the cloud resources for which monitoring data needs to be obtained are determined according to the cloud services to be evaluated.

[0055] Optionally, the frequency of acquiring the monitoring data is determined according to the memory occupancy of the monitoring data management layer itself.

[0056] On this basis, before storing the monitoring data, a first data summary is generated for the monitoring data to determine the authenticity of the monitoring data when the upper-layer platform calls the monitoring data. It can be understood that the first data summary can be one or more, and the acquired monitoring data is grouped in different ways, and each group of data generates a corresponding first data summary. In this application, the first data summary is generated by calculating the hash value of a group of data.

[0057] Optionally, the monitoring data of multiple cloud resources belonging to the same cloud service acquired at one time are integrated into a group of data to generate a first data summary.

[0058] Optionally, monitoring data of multiple cloud resources belonging to the same cloud service acquired multiple times within a period of time are integrated into a group of data to generate a first data summary.

[0059] Optionally, the monitoring data of multiple cloud resources belonging to the same cloud service acquired at one time are grouped according to a fixed time window to generate a first data summary for each group of data.

[0060] Optionally, the monitoring data of a cloud resource is determined as a group of data, and a first data summary of each group of data is generated.

[0061] Furthermore, the monitoring data management layer also receives and stores the protocol data uploaded by the cloud resource management layer for invocation by the upper-level platform.

[0062] The cloud service evaluation layer is used to retrieve the monitoring data stored in the monitoring data management layer and generate a second data summary of the monitoring data. When the second data summary is the same as the first data summary, the reputation score of the cloud service is determined according to the monitoring data, and the cloud service is supervised according to the reputation score.

[0063] like Figure 1 As shown, the cloud service evaluation layer is the upper platform of the monitoring data management layer, and obtains the monitoring data and protocol data of the cloud service from the monitoring data management layer.

[0064] The various indicators that the cloud service needs to complete as determined in the agreement data are used as the criteria for scoring the cloud service. The actual completion status of each indicator in the cloud service is obtained from the monitoring data, and the reputation score of the cloud service is calculated based on the criteria and completion status.

[0065] Among them, before calculating the reputation score of the cloud service, it is necessary to verify the authenticity of the monitoring data obtained from the self-monitoring data management, so as to ensure the reliability of the reputation score.

[0066] Specifically, the monitoring data stored in the monitoring data management layer is retrieved regularly according to the scoring requirements to generate a second data summary of the monitoring data. The second data summary is generated in the same manner as the first data summary, and in this application, a hash value of a set of monitoring data is calculated.

[0067] It is understandable that the second data summary can be one or more, and by grouping the called monitoring data, each group of data generates a corresponding second data summary. It should be noted that the way of grouping the called monitoring data at this time needs to correspond to the grouping method when generating the first data summary, so as to ensure that the first data summary stored by the monitoring data management layer and the second data summary generated by the cloud service evaluation layer are the same when each group of monitoring data is not tampered with.

[0068] Furthermore, when the second data summary is identical to the first data summary, it is confirmed that the monitoring data called from the monitoring data management layer is identical to the monitoring data obtained from the cloud resource management layer of the monitoring data management layer, that is, the monitoring data that has not been tampered with is obtained.

[0069] On this basis, data related to the indicators specified in the service level agreement are obtained from the monitoring data, and the score of each indicator is determined according to the completion rate of each indicator. The weighted sum of all indicator scores of a cloud service is used as the reputation score of the cloud service.

[0070] Optionally, different weight allocation strategies are used when calculating the reputation score of each cloud service according to the category of the cloud service.

[0071] The service providers of each cloud service are supervised based on the reputation score of each cloud service.

[0072] Specifically, the cloud service assessment layer issues instructions to the cloud resource management layer based on the reputation score of each cloud service, and implements a reward and punishment mechanism for cloud service providers that provide cloud services, thereby achieving supervision of cloud service resources.

[0073] The present invention constructs a cloud resource management layer, a monitoring data management layer and a cloud service evaluation layer to monitor and manage the protocol data and monitoring data generated from the signing of a service level agreement to the service implementation between users and cloud service providers. The present invention implements quantitative scoring of cloud services through protocol data and monitoring data, and supervises heterogeneous resources of multiple cloud service providers based on the scoring, thereby saving the cost of human supervision, improving the efficiency of supervision and evaluation, and realizing trusted supervision of cloud computing services.

[0074] In the cloud service resource supervision system of the present invention, the monitoring data includes basic resource information data, operation status data and abnormal event data of the cloud resources.

[0075] The monitoring data management layer obtains monitoring data from the cloud resource management layer, including basic resource information data, operation status data, and abnormal event data of each cloud resource.

[0076] Optionally, the basic resource information data includes the resource ID (Identification, number) of the cloud resource, the resource type and the SLA agreement number n to which the cloud service corresponding to the cloud resource belongs. SLA wait.

[0077] Optionally, the operating status data includes CPU usage, memory, storage, operating bandwidth, and operating latency of cloud resources.

[0078] Optionally, the abnormal event data includes relevant information such as the abnormal event type, threat level, and abnormal time when an abnormal event occurs in cloud resources.

[0079] Among them, abnormal event types may include resource unavailability and various related security events.

[0080] Specifically, the basic information of cloud resources is used to match the resources with the SLA agreements to which they belong, thereby determining the service indicators and indicator standards agreed upon in the SLA agreement for each cloud resource.

[0081] The actual operation data of the cloud resources corresponding to each cloud service is obtained through the operation status data of the cloud resources, and the occupancy of the server during the operation of the cloud resources is used as one of the bases for calculating the reputation score of the cloud service.

[0082] Through the operation status data of cloud resources and the corresponding SLA agreement data, the completion degree of the indicators agreed in the SLA agreement of cloud services is determined, and it is used as one of the bases for calculating the reputation score of cloud services.

[0083] The abnormal event data of cloud resources is used to obtain the abnormal event occurrence of each cloud service corresponding to the cloud resources, and it is used as one of the bases for calculating the cloud service reputation score.

[0084] The weighted sum of the three bases is calculated as the reputation score of the cloud service, so that the reputation score of each cloud service can reflect the performance of the cloud service in these three different dimensions. At the same time, by adjusting the weights corresponding to each type of basis for different types of cloud services, the reputation score finally calculated can overcome the influence of different types of cloud services on the reputation score calculation to a certain extent, so that the reputation score can be used to measure the service quality of different types of cloud services and realize the unified supervision of heterogeneous cloud resources of multiple cloud providers.

[0085] In the cloud service resource supervision system of the present invention, the monitoring data management layer also includes a blockchain, and the monitoring data management layer is specifically used for:

[0086] Acquire monitoring data of a plurality of cloud resources from the cloud resource management layer, aggregate the monitoring data of each cloud resource according to the type of the monitoring data, generate a first data summary of the aggregated monitoring data, and store the first data summary in the blockchain.

[0087] After obtaining the monitoring data of each cloud resource in the cloud resource management layer through regular requests, the monitoring data needs to be aggregated first.

[0088] Optionally, the monitoring data is aggregated according to the categories of each item of data in the monitoring data, and the aggregated data is stored, thereby achieving classified statistics of the data and reducing the amount of data during storage.

[0089] Optionally, the operating status data in the monitoring data is aggregated based on time; and the abnormal event data in the monitoring data is aggregated based on events.

[0090] Specifically, for the running status data, the mean, extreme value, and variance of the running status data within a certain time window are calculated. If the variance and the difference between the extreme values ​​are less than the corresponding preset thresholds, the mean is output as the aggregated running status data, and the size of the time window at this time is recorded, otherwise the size of the time window is adjusted and recalculated.

[0091] The certain time window may be a rolling window or a sliding window.

[0092] For abnormal event data, the repeated abnormal event data within a certain time window is removed and the duration is calculated. Finally, the type, start time and duration of the abnormal event are output as the aggregated abnormal event data.

[0093] The certain time window may be the same as or different from the time window of the operation status data.

[0094] On this basis, a piece of aggregated monitoring data includes basic resource information of a cloud resource, aggregated operating status data and aggregated abnormal event data.

[0095] The aggregated monitoring data is grouped, and the first data summary of each group of data is calculated and stored in the blockchain to realize a distributed monitoring data management layer structure.

[0096] In the cloud service resource supervision system of the present invention, the monitoring data management layer is specifically used for:

[0097] Grouping the aggregated monitoring data according to a first time window, calculating a first data summary of each group of monitoring data and storing the summary in the blockchain;

[0098] like Figure 1 As shown, the monitoring data management layer also includes a data aggregation module, which is also used to group the aggregated data and generate a first data summary for each group of data and send it to the blockchain for storage.

[0099] Specifically, a fixed first time window is used to group the monitoring data of each cloud resource.

[0100] Optionally, the length of the first time window is determined according to demand.

[0101] Optionally, the first time window may be a rolling window or a sliding window.

[0102] For example, when the first time window is a rolling window with a length of 12 hours, for the aggregated monitoring data of cloud resource 1, the multiple aggregated monitoring data between 0-12 o'clock are divided into a group, and the corresponding first data summary is calculated; the multiple aggregated monitoring data between 12-24 o'clock are divided into a group, and another corresponding first data summary is calculated.

[0103] The cloud service evaluation layer is specifically used to obtain the monitoring data in the second time window, and after grouping the monitoring data in the second time window based on the first time window, calculate a second data summary of each group of monitoring data.

[0104] On this basis, in order for the cloud service evaluation layer to verify the monitoring data based on the first data summary, it is necessary to obtain the aggregated monitoring data in a fixed second time window, wherein the length of the second time window is greater than the length of the first time window.

[0105] Furthermore, before calculating the second data summary, the acquired data needs to be grouped so that the monitoring data used in calculating the second data summary corresponds to the data used in calculating the first data summary.

[0106] In a feasible implementation, when the length of the first time window is 12 hours, the second time window is 24 hours, that is, the aggregated monitoring data is obtained every 24 hours. The data statistical analysis module in the cloud service evaluation layer divides the obtained 24-hour data into two groups based on the length of the first time window, so that each group of data corresponds to the data in the first time window, generates two second data summaries respectively, and compares the second data summaries with the corresponding first data summaries to determine whether the data stored in the monitoring data management layer has changed.

[0107] In another feasible implementation, when the length of the first time window is 6 hours, the second time window is 20 hours. In this case, within the time length corresponding to the second time window, 3 groups of 6-hour aggregated monitoring data can be fully acquired. Therefore, the acquired monitoring data is divided into three groups of 6 hours each, and the start and end time points of the three groups of data correspond to the time points of the first time window, so that each group of data corresponds to the data of the first time window, and three second data summaries are generated respectively. The second data summaries are compared with the corresponding first data summaries to determine whether the data stored in the monitoring data management layer has changed.

[0108] In the cloud service resource supervision system of the present invention, the blockchain is also used for:

[0109] Storing service level agreement data sent by the cloud resource management layer, wherein the service level agreement data represents a service agreement corresponding to the cloud service;

[0110] In response to the calling instruction of the cloud service evaluation layer, the service level agreement data of the inter-cloud service in the calling instruction is sent to the cloud service evaluation layer.

[0111] When users obtain cloud resources, they need to sign a service level agreement (SLA) with the cloud service provider to clarify the specific service content, indicators and other information. Whenever an agreement is signed, the cloud resource management layer will generate standardized SLA data, that is, service level agreement data, including the start and end dates of the cloud service, information of all parties, service description, performance indicators, security standards, breach of contract penalties, etc., which are simplified and stored in the form of fields. Subsequently, the cloud resource management layer will send the SLA data to the blockchain for trusted evidence.

[0112] On this basis, when the blockchain stores the SLA data of each cloud service, it is numbered according to the SLA agreement number n. SLA Divide the storage location corresponding to each cloud service.

[0113] When the data aggregation module sends the first data summary to the blockchain, according to n in the aggregated monitoring data SLAand resource ID, and stores the first data summary to its own n SLA The corresponding storage location so that the cloud service evaluation layer can be used in n SLA The reputation score of a cloud service is calculated based on monitoring data of a cloud service and a corresponding first data summary.

[0114] Specifically, when the cloud service evaluation layer needs to evaluate a certain cloud service within a certain period of time, based on the n SLA Send a call instruction to the monitoring data management layer, and the blockchain responds to the call instruction and sends the corresponding n SLA The first data summary and service level agreement data are stored below.

[0115] In the cloud service resource supervision system of the present invention, the monitoring data management layer also includes a time series database, and the monitoring data management layer is specifically used for:

[0116] The aggregated monitoring data is stored in the time series database, and a plurality of copies of the aggregated monitoring data are generated;

[0117] like Figure 1 As shown, the monitoring data management layer also includes a time series database. After the data aggregation module generates a first data summary and sends it to the blockchain, the aggregated monitoring data is sent to the time series database for storage.

[0118] At the same time, after receiving the aggregated monitoring data, the time series database generates several copies of the received data for backup storage.

[0119] Optionally, the number of copies of the replica data stored is determined according to the capacity of the time series database and the size of the aggregated monitoring data.

[0120] The backup copy data is used to restore the data and achieve fault tolerance when the monitoring data called in the cloud service evaluation layer is damaged.

[0121] The cloud service evaluation layer is specifically used to, when the second data summary is different from the first data summary, read the replica data in the time series database and calculate a third data summary of the replica data;

[0122] When the second data summary is different from the first data summary, the cloud service evaluation layer determines that the retrieved aggregated monitoring data is damaged. At this time, the corresponding replica data stored in the time series database can be retrieved for data recovery, thereby evaluating the cloud service based on the replica data.

[0123] Before evaluation, the authenticity of the copy data also needs to be confirmed, so the third data summary of the copy data needs to be calculated. The calculation method is the same as the calculation method of the second data summary, so it will not be repeated here.

[0124] In a case where the third data summary is identical to the first data summary, a reputation score of the cloud service is determined according to the copy data, and the cloud service is supervised according to the reputation score.

[0125] The third data summary is the same as the first data summary, indicating that the called copy data is normal. Therefore, the reputation score of the corresponding cloud service is calculated based on the called copy data, and then the cloud service is supervised by the reputation score.

[0126] If the third data digest is different from the first data digest, other stored copy data are continuously called for verification in the same manner, and when it is determined that other copy data are not damaged, the reputation score of the cloud service is calculated using other copy data.

[0127] In the cloud service resource supervision system of the present invention, the cloud service evaluation layer further includes an evaluation module, and the evaluation module is used to:

[0128] Determine a service quality indicator score for any of the cloud services based on the monitoring data;

[0129] like Figure 1 As shown, the cloud service evaluation layer also includes a data statistics analysis module, which is used to perform statistics and analysis on the called monitoring data before evaluation.

[0130] Specifically, the data statistics analysis module uses the same SLA agreement number n SLA The corresponding monitoring data and SLA data are obtained from the monitoring data management layer based on a group of cloud resources under the same cloud.

[0131] Based on the aggregated running status data in the called monitoring data, the QoS indicator (service quality indicator) data of the cloud resources to which it belongs is extracted, including the bandwidth, latency and packet loss rate of the cloud resources. The extracted data is sent to the evaluation module, which calculates the score of each indicator based on the received data.

[0132] Furthermore, a QoS scoring table is predetermined, which distinguishes the types of cloud resources, and uses different indicators and scoring intervals for different types of resources. The scores of the QoS scoring table are used as service quality indicator scores.

[0133] Determine a security score of any of the cloud services based on the abnormal event data and the operating status data;

[0134] The data statistical analysis module extracts the aggregated abnormal event data from the called monitoring data, including the type of abnormal event, threat level and duration, etc. The called data is then sent to the evaluation module, which calculates the score of each indicator based on the received data.

[0135] Furthermore, an abnormal event scoring table is predetermined, which scores events based on three aspects: threat level, duration, and event frequency, and serves as the security score for cloud services.

[0136] Determining a performance score of any of the cloud services based on the service level agreement data;

[0137] The performance score of cloud services represents the performance of agreements when cloud resources provide services.

[0138] The data statistical analysis module extracts the operating status data, abnormal event data and SLA data stored in the blockchain from the called monitoring data, and then sends the called data to the evaluation module, which calculates the score of each indicator based on the received data.

[0139] Determine the various indicators in the completion score sheet in advance based on the indicators in the SLA agreement, and determine whether the actual service provided by Yunji Service meets the requirements of the SLA agreement based on the operating status data and abnormal event data, and then calculate and determine the completion score sheet.

[0140] Optionally, if necessary, user evaluations can also be used as the basis for scoring the completion score sheet.

[0141] When determining the service quality index score, fulfillment score and security score of Yunji Services, that is, calculating the specific scores of the three scoring tables, the values ​​of each indicator in the scoring table are obtained in turn based on the called monitoring data and SLA data, and the score of the indicator is given according to the interval of the corresponding value in the scoring table. Finally, the scores of each indicator are added up to obtain the final score of the evaluation item; when the evaluation item contains multiple groups of data (such as multiple resource instances), the average of the multiple groups of scores is taken.

[0142] Optionally, the intervals of corresponding values ​​in the scoring table are pre-divided according to the specific values ​​of the indicators, that is, each indicator in the scoring table is scored in the form of points.

[0143] The final service quality index score is A1, the security score is A2, and the performance score is A3.

[0144] A weighted sum of the service quality index score, the security score and the fulfillment score is used as the reputation score of any of the cloud services.

[0145] A pre-trained weight calculation model is embedded in the evaluation module, which is used to determine the weight vectors W1, W2 and W3 corresponding to the service quality indicator score A1, the security score A2 and the fulfillment score A3 respectively.

[0146] On this basis, the reputation score of Yunji Service is S=W1×A1+W2×A2+W3×A3.

[0147] Among them, the weight calculation model is used to dynamically calculate the scoring weights of each evaluation item. The model includes a prediction module and a calculation module. The prediction module will predict the load and abnormal events based on the overall statistical data of the cloud platform through deep learning and other methods (such as LSTM, long short-term memory network); the calculation module will adjust the initial weight according to the prediction results to obtain the final weight score. For example, when it is predicted that the load will increase significantly, the weight of the QoS score will be increased, thereby giving high-performance resources a higher score; through the reward and punishment mechanism of the cloud platform, providers of high-performance resources will get more task allocation opportunities, which will also help alleviate the load pressure of the cloud platform. In addition to the prediction results, the weight calculation module also uses alarm signals as the basis for weight calculation to deal with unexpected high loads, high-frequency abnormal events and other emergencies.

[0148] It should be noted that the model will perform a weight calculation every certain period of time, and the weight obtained will be applied to all statistical data within that period of time to ensure the fairness of the evaluation. When judging the time period of the statistical data, the timestamp stored in the data (i.e. the time when the data was recorded) should be referred to, rather than the time when the score was calculated.

[0149] The present invention uses different preset standards to score different types of resources through the cloud service evaluation layer, and at the same time performs weighted calculation on the reputation scores of cloud service resources from multiple dimensions, thereby improving the fairness of resource evaluation; at the same time, the cloud service evaluation layer regards multiple resources that execute the same SLA agreement as a whole to further judge the agreement fulfillment when the resources provide services. This strategy is more suitable for cloud computing scenarios with multi-cloud collaboration.

[0150] The inter-cloud service resource supervision method provided by the present invention is described below. The inter-cloud service resource supervision method described below and the inter-cloud service resource supervision system described above can be referenced to each other.

[0151] like Figure 2 As shown in the figure, the cloud service resource supervision method includes:

[0152] Step 201, determining the cloud resources corresponding to each cloud service through the cloud resource management layer, and performing data monitoring on each of the cloud resources;

[0153] The cloud resource management layer is a cloud resource platform that implements basic functions, including multiple different cloud resources. It is responsible for monitoring the status of each cloud resource and the signing of cloud service level agreements with cloud service providers.

[0154] Among them, different cloud service providers provide different cloud services after signing service level agreements, and each cloud service uses different cloud resources. The cloud resources required by cloud services are different due to the different types and contents of cloud services.

[0155] Optionally, the cloud resource management layer allocates corresponding cloud resources to each cloud service according to the category of the cloud service, the signed service level agreement, and the occupancy of the cloud resources.

[0156] The cloud resource management layer obtains the monitoring data of each cloud resource by monitoring each cloud resource, and then obtains the monitoring data of all cloud resources corresponding to each cloud service to determine the monitoring data of the cloud service, which serves as the basis for quantitative scoring of the cloud service.

[0157] Furthermore, the cloud resource management layer also provides a standardized monitoring data query interface to the upper platform, which is the monitoring data management layer in the present invention, so that the upper platform can obtain the monitoring data of each cloud service in the cloud resource management layer and score the cloud service. When receiving a data query request from the monitoring data management layer, the current monitoring data of the cloud resources is obtained and returned through the interface after statistics.

[0158] At the same time, the cloud resource management layer also provides the upper-level platform with the agreement data of the cloud service agreement signed between the user and the cloud service provider.

[0159] Optionally, the agreement data at least includes SLA (Service Level Agreement) data. The SLA data characterizes the performance, availability, support and penalty clauses of the cloud service that the cloud service provider needs to provide to the user.

[0160] Step 202, obtaining monitoring data of a plurality of cloud resources from the cloud resource management layer through the monitoring data management layer, and generating and storing a first data summary of the monitoring data;

[0161] The monitoring data management layer is the upper layer platform of the cloud resource management layer. It is responsible for statistics, aggregation, analysis and storage of monitoring data, and provides a query interface to the upper layer platform.

[0162] The monitoring data represents the cloud resources of each cloud service and the service status of the cloud resources.

[0163] The monitoring data management layer regularly obtains monitoring data of cloud resources through the standard interface provided by the cloud resource management layer.

[0164] Optionally, the cloud resources for which monitoring data needs to be obtained are determined according to the cloud services to be evaluated.

[0165] Optionally, the frequency of acquiring the monitoring data is determined according to the memory occupancy of the monitoring data management layer itself.

[0166] On this basis, before storing the monitoring data, a first data summary is generated for the monitoring data to determine the authenticity of the monitoring data when the upper-layer platform calls the monitoring data. It can be understood that the first data summary can be one or more, and the acquired monitoring data is grouped in different ways, and each group of data generates a corresponding first data summary. In this application, the first data summary is generated by calculating the hash value of a group of data.

[0167] Optionally, the monitoring data of multiple cloud resources belonging to the same cloud service acquired at one time are integrated into a group of data to generate a first data summary.

[0168] Optionally, monitoring data of multiple cloud resources belonging to the same cloud service acquired multiple times within a period of time are integrated into a group of data to generate a first data summary.

[0169] Optionally, the monitoring data of multiple cloud resources belonging to the same cloud service acquired at one time are grouped according to a fixed time window to generate a first data summary for each group of data.

[0170] Optionally, the monitoring data of a cloud resource is determined as a group of data, and a first data summary of each group of data is generated.

[0171] Furthermore, the monitoring data management layer also receives and stores the protocol data uploaded by the cloud resource management layer for invocation by the upper-level platform.

[0172] Step 203, retrieve the monitoring data stored in the monitoring data management layer through the cloud service evaluation layer and generate a second data summary of the monitoring data. When the second data summary is the same as the first data summary, determine the reputation score of the cloud service based on the monitoring data, and supervise the cloud service based on the reputation score.

[0173] The cloud service evaluation layer is the upper platform of the monitoring data management layer, and obtains the monitoring data and protocol data of the cloud service from the monitoring data management layer.

[0174] The various indicators that the cloud service needs to complete as determined in the agreement data are used as the criteria for scoring the cloud service. The actual completion status of each indicator in the cloud service is obtained from the monitoring data, and the reputation score of the cloud service is calculated based on the criteria and completion status.

[0175] Among them, before calculating the reputation score of the cloud service, it is necessary to verify the authenticity of the monitoring data obtained from the self-monitoring data management, so as to ensure the reliability of the reputation score.

[0176] Specifically, the monitoring data stored in the monitoring data management layer is retrieved regularly according to the scoring requirements to generate a second data summary of the monitoring data. The second data summary is generated in the same manner as the first data summary, and in this application, a hash value of a set of monitoring data is calculated.

[0177] It is understandable that the second data summary can be one or more, and by grouping the called monitoring data, each group of data generates a corresponding second data summary. It should be noted that the way of grouping the called monitoring data at this time needs to correspond to the grouping method when generating the first data summary, so as to ensure that the first data summary stored by the monitoring data management layer and the second data summary generated by the cloud service evaluation layer are the same when each group of monitoring data is not tampered with.

[0178] Furthermore, when the second data summary is identical to the first data summary, it is confirmed that the monitoring data called from the monitoring data management layer is identical to the monitoring data obtained from the cloud resource management layer of the monitoring data management layer, that is, the monitoring data that has not been tampered with is obtained.

[0179] On this basis, data related to the indicators specified in the service level agreement are obtained from the monitoring data, and the score of each indicator is determined according to the completion rate of each indicator. The weighted sum of all indicator scores of a cloud service is used as the reputation score of the cloud service.

[0180] Optionally, different weight allocation strategies are used when calculating the reputation score of each cloud service according to the category of the cloud service.

[0181] The service providers of each cloud service are supervised based on the reputation score of each cloud service.

[0182] Specifically, the cloud service assessment layer issues instructions to the cloud resource management layer based on the reputation score of each cloud service, and implements a reward and punishment mechanism for cloud service providers that provide cloud services, thereby achieving supervision of cloud service resources.

[0183] The present invention constructs a cloud resource management layer, a monitoring data management layer and a cloud service evaluation layer to monitor and manage the protocol data and monitoring data generated from the signing of a service level agreement to the service implementation between users and cloud service providers. The present invention implements quantitative scoring of cloud services through protocol data and monitoring data, and supervises heterogeneous resources of multiple cloud service providers based on the scoring, thereby saving the cost of human supervision, improving the efficiency of supervision and evaluation, and realizing trusted supervision of cloud computing services.

[0184] Figure 3 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 3 As shown, the electronic device may include: a processor 310, a communication interface 320, a memory 830 and a communication bus 340, wherein the processor 310, the communication interface 320 and the memory 330 communicate with each other through the communication bus 340. The processor 310 may call the logic instructions in the memory 330 to execute the cloud service resource supervision method, which includes determining the cloud resources corresponding to each cloud service through the cloud resource management layer, and performing data monitoring on each of the cloud resources; obtaining monitoring data of a plurality of cloud resources from the cloud resource management layer through the monitoring data management layer, generating and storing a first data summary of the monitoring data; calling the monitoring data stored in the monitoring data management layer through the cloud service evaluation layer and generating a second data summary of the monitoring data, and determining the reputation score of the cloud service according to the monitoring data when the second data summary is the same as the first data summary, and supervising the cloud service according to the reputation score.

[0185] In addition, the logic instructions in the above-mentioned memory 330 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0186] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the cloud service resource supervision method provided by the above methods, which includes: determining the cloud resources corresponding to each cloud service through the cloud resource management layer, and performing data monitoring on each of the cloud resources; obtaining monitoring data of several cloud resources from the cloud resource management layer through the monitoring data management layer, generating and storing a first data summary of the monitoring data; retrieving the monitoring data stored in the monitoring data management layer through the cloud service evaluation layer and generating a second data summary of the monitoring data; when the second data summary is the same as the first data summary, determining the reputation score of the cloud service based on the monitoring data, and supervising the cloud service based on the reputation score.

[0187] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, it is implemented to execute the cloud service resource supervision method provided by the above-mentioned methods. The method includes: determining the cloud resources corresponding to each cloud service through the cloud resource management layer, and performing data monitoring on each of the cloud resources; obtaining monitoring data of a number of cloud resources from the cloud resource management layer through the monitoring data management layer, generating and storing a first data summary of the monitoring data; retrieving the monitoring data stored in the monitoring data management layer through the cloud service evaluation layer and generating a second data summary of the monitoring data, and when the second data summary is the same as the first data summary, determining the reputation score of the cloud service based on the monitoring data, and supervising the cloud service based on the reputation score.

[0188] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0189] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0190] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A cloud service resource monitoring system, characterized in that: include: Cloud resource management layer, monitoring data management layer and cloud service assessment layer; The cloud resource management layer is used to determine the cloud resources corresponding to each cloud service and to perform data monitoring on each cloud resource; The monitoring data management layer is used to obtain monitoring data of a plurality of cloud resources from the cloud resource management layer, and generate and store a first data summary of the monitoring data; The cloud service evaluation layer is used to retrieve the monitoring data stored in the monitoring data management layer and generate a second data summary of the monitoring data, and when the second data summary is the same as the first data summary, determine the reputation score of the cloud service according to the monitoring data, and supervise the cloud service according to the reputation score; The monitoring data management layer also includes a blockchain, and the monitoring data management layer is specifically used for: Acquire monitoring data of a plurality of cloud resources from the cloud resource management layer, aggregate the monitoring data of each cloud resource according to the type of the monitoring data, generate a first data summary of the aggregated monitoring data, and store the first data summary in the blockchain; The monitoring data management layer is specifically used for: Grouping the aggregated monitoring data according to a first time window, calculating a first data summary of each group of monitoring data and storing the summary in the blockchain; The cloud service evaluation layer is specifically used to obtain the monitoring data within a second time window, and after grouping the monitoring data within the second time window based on the first time window, calculate a second data summary for each group of monitoring data, wherein the length of the second time window is greater than the length of the first time window.

2. The cloud service resource monitoring system according to claim 1, characterized in that: The monitoring data includes basic resource information data, operation status data and abnormal event data of the cloud resources.

3. The cloud service resource monitoring system according to claim 2 is characterized in that: The blockchain is also used to: Storing service level agreement data sent by the cloud resource management layer, wherein the service level agreement data represents a service agreement corresponding to the cloud service; In response to the calling instruction of the cloud service evaluation layer, the service level agreement data of the inter-cloud service in the calling instruction is sent to the cloud service evaluation layer.

4. The cloud service resource monitoring system according to claim 1, characterized in that: The monitoring data management layer also includes a time series database, and the monitoring data management layer is specifically used for: The aggregated monitoring data is stored in the time series database, and a plurality of copies of the aggregated monitoring data are generated; The cloud service evaluation layer is specifically used to, when the second data summary is different from the first data summary, read the replica data in the time series database and calculate a third data summary of the replica data; In a case where the third data summary is identical to the first data summary, a reputation score of the cloud service is determined according to the copy data, and the cloud service is supervised according to the reputation score.

5. The cloud service resource monitoring system according to claim 3 is characterized in that: The cloud service evaluation layer further includes an evaluation module, which is used to: Determine a service quality indicator score for any of the cloud services based on the monitoring data; Determine a security score of any of the cloud services based on the abnormal event data and the operating status data; Determining a performance score of any of the cloud services based on the service level agreement data; A weighted sum of the service quality index score, the security score and the fulfillment score is used as the reputation score of any of the cloud services.

6. A cloud service resource supervision method, characterized in that: include: Determine the cloud resources corresponding to each cloud service through the cloud resource management layer, and perform data monitoring on each of the cloud resources; Acquire monitoring data of a plurality of cloud resources from the cloud resource management layer through the monitoring data management layer, and generate and store a first data summary of the monitoring data; Retrieving the monitoring data stored in the monitoring data management layer through the cloud service evaluation layer and generating a second data summary of the monitoring data, and when the second data summary is the same as the first data summary, determining the reputation score of the cloud service according to the monitoring data, and supervising the cloud service according to the reputation score; The step of obtaining the monitoring data of a plurality of cloud resources from the cloud resource management layer through the monitoring data management layer, and generating and storing a first data summary of the monitoring data specifically includes: Acquire monitoring data of a plurality of cloud resources from the cloud resource management layer, aggregate the monitoring data of each cloud resource according to the type of the monitoring data, generate a first data summary of the aggregated monitoring data, and store the first data summary in the blockchain; The step of generating a first data summary of the aggregated monitoring data and storing the first data summary in the blockchain specifically includes: Grouping the aggregated monitoring data according to a first time window, calculating a first data summary of each group of monitoring data and storing the summary in the blockchain; The step of retrieving the monitoring data stored in the monitoring data management layer through the cloud service evaluation layer and generating a second data summary of the monitoring data specifically includes: The monitoring data in a second time window is acquired, and after the monitoring data in the second time window is grouped based on the first time window, a second data summary of each group of monitoring data is calculated, wherein the length of the second time window is greater than the length of the first time window.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the cloud service resource supervision method as claimed in claim 6 is implemented.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the cloud service resource supervision method as claimed in claim 6 is implemented.

Images