Cloud product data processing method, system and device and storage medium

Abstract

The embodiment of the invention provides a cloud product data processing method, system and device and a storage medium. In the embodiment of the invention, the cloud product demand is effectively combined with the resource supply capability of the bottom layer of a cloud service manufacturer, and the cloud product demand is analyzed into K candidate resource deployment schemes; according to N resource supply capabilities provided by a cloud service manufacturer, resource arrangement and resource loss cost information estimation processing are carried out on K candidate resource deployment schemes, on one hand, required cloud products can be quickly provided for users by means of the resource supply capabilities of the bottom layer of the cloud service manufacturer, and resource supply flexibility is realized, and on the other hand, resource supply efficiency is improved. The resource loss cost of the resource solution is considered, so that the resource supply cost performance can be improved.

Description

Technical Field

[0001] The present application relates to the field of cloud computing technology, and in particular to a cloud product data processing method, system, device and storage medium. Background Art

[0002] With the continuous development of cloud computing technology, cloud service providers are providing more and more cloud products with more and more specifications and types, and the availability zones and geographical areas covered by various cloud products are becoming more and more extensive. Users can purchase or release cloud products of the required type and specifications on demand, such as a cloud server with 2 core virtual processors (vCPU) and 4G memory, or a cloud database with 1 core vCPU and 2G memory, etc., without having to maintain various hardware resources, which not only reduces management costs, but also enjoys reliable, stable and scalable performance advantages.

[0003] As the demand for cloud products continues to increase, cloud service providers have perfect supply and deployment plans for various cloud products to meet user needs. However, even so, it is inevitable that there will be a sudden user demand that cannot be met, that is, cloud service providers cannot provide the cloud products required by users in a short period of time. Therefore, how to quickly provide users with the required cloud products on the basis of improving the elasticity of resource supply and the cost-effectiveness of resource supply has become a problem that needs to be solved urgently. Summary of the invention

[0004] Multiple aspects of the present application provide a cloud product data processing method, system, device and storage medium, which are used to quickly provide users with the required cloud products on the basis of improving the elasticity of resource supply and the cost-effectiveness of resource supply.

[0005] The embodiment of the present application provides a cloud product data processing system, including: a demand management module, used to obtain cloud product demand information, and parse the cloud product demand information into at least one candidate resource deployment solution;

[0006] A resource orchestration module, configured to perform resource orchestration and resource loss cost information estimation processing on the at least one candidate resource deployment solution according to at least one resource supply capability provided by the cloud service provider, so as to obtain a target resource solution whose resource loss cost information meets the requirements and its corresponding target resource deployment solution;

[0007] An execution delivery module is used to execute resource processing tasks in the available zone corresponding to the target resource deployment solution according to the target resource solution, so as to obtain the cloud product corresponding to the cloud product demand information.

[0008] The present application also provides a cloud product data processing method, including:

[0009] Obtaining cloud product demand information, and parsing the cloud product demand information into at least one candidate resource deployment solution;

[0010] Perform resource orchestration and resource loss cost information estimation processing on the at least one candidate resource deployment solution according to at least one resource supply capability provided by the cloud service provider, so as to obtain a target resource solution whose resource loss cost information meets the requirements and its corresponding target resource deployment solution;

[0011] According to the target resource solution, a resource processing task is executed in an available zone corresponding to the target resource deployment solution to obtain a cloud product corresponding to the cloud product demand information.

[0012] The present application also provides a cloud product data processing method, including:

[0013] Acquire a candidate resource deployment scheme and at least one resource supply capability, wherein the candidate resource deployment scheme includes at least resource information to be deployed and available zone information to be deployed;

[0014] According to the at least one resource supply capability, estimate the unit resource loss cost information when the at least one resource supply capability provides the required resources for any candidate resource deployment solution;

[0015] A resource solution and its resource loss cost information corresponding to the resource deployment solution are generated according to the unit resource loss cost information corresponding to at least one resource supply capability and the resource information to be deployed.

[0016] An embodiment of the present application also provides an electronic device, comprising: a memory and a processor; the memory is used to store a computer program, and the processor is coupled to the memory and is used to execute the computer program to execute the steps in any method provided in the embodiment of the present application.

[0017] An embodiment of the present application also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the processor is enabled to implement the steps of any method provided in the embodiment of the present application.

[0018] In the embodiment of the present application, the cloud product demand is effectively combined with the underlying resource supply capability of the cloud service provider. The cloud product demand is parsed into at least one candidate resource deployment plan. According to at least one resource supply capability provided by the cloud service provider, resource orchestration and resource loss cost information estimation processing are performed on at least one candidate resource deployment plan. The target resource deployment plan and its corresponding target resource solution whose resource loss cost information meets the requirements are obtained. According to the target resource solution, the resource processing task is executed in the available area corresponding to the target resource deployment plan, and the cloud product corresponding to the cloud product demand can be automatically produced. Among them, with the help of the underlying resource supply capability of the cloud service provider, the required cloud products can be quickly provided to users, and the elasticity of resource supply can be realized. In addition, since the resource loss cost of the resource solution is taken into account, it is conducive to improving the cost performance of resource supply. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0020] Figure 1a A schematic diagram of the structure of a cloud product data processing system provided in an embodiment of the present application;

[0021] Figure 1b An internal implementation structure of a demand management module provided in an embodiment of the present application;

[0022] Figure 1c An internal implementation structure of a resource orchestration module provided in an embodiment of the present application;

[0023] Figure 1d A schematic diagram showing receiving cloud product demand information provided in an embodiment of the present application;

[0024] Figure 1e A schematic diagram of cloud product demand information types and conversions provided in an embodiment of the present application;

[0025] Figure 2a A schematic diagram of the combination of specifications, availability zones, and providers provided in an embodiment of the present application;

[0026] Figure 2b A schematic diagram of the corresponding demand time for resource provision by the Provider provided in an embodiment of the present application;

[0027] Figure 2c A schematic diagram of the process of delivering and executing the target resource solution provided in the embodiment of the present application;

[0028] Figure 2dA state diagram of dynamic priority provided in an embodiment of the present application;

[0029] Figure 2e The overall operation block diagram of the cloud product data processing system provided in the embodiment of the present application;

[0030] Figure 3a A flowchart of a cloud product data processing method provided in an embodiment of the present application;

[0031] Figure 3b A flowchart of another cloud product data processing method provided in an embodiment of the present application;

[0032] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0033] In order to make the purpose, technical solution and advantages of the present application clearer, the technical solution of the present application will be clearly and completely described below in combination with the specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.

[0034] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0035] With the development of cloud computing technology and the increasing demand for cloud products, how to quickly provide users with the required cloud products on the basis of improving the elasticity of resource supply and the cost-effectiveness of resource supply has become a problem that needs to be solved urgently. In an embodiment of the present application, a solution that can quickly and automatically produce cloud products is provided, which effectively combines the demand for cloud products with the underlying resource supply capabilities of cloud service providers, and analyzes the demand for cloud products into at least one candidate resource deployment solution. According to at least one resource supply capability provided by the cloud service provider, resource orchestration and resource loss cost information estimation processing are performed on at least one candidate resource deployment solution, and the target resource deployment solution whose resource loss cost information meets the requirements and its corresponding target resource solution formed by at least one resource supply capability are obtained. According to at least one resource supply capability in the target resource solution, resource processing tasks are executed in the available area corresponding to the target resource deployment solution, and cloud products corresponding to the cloud product demand can be automatically produced. Among them, with the help of the underlying resource supply capabilities of the cloud service provider, the required cloud products can be provided to users quickly and on demand, and the elasticity of resource supply can be realized. In addition, since the resource loss cost of the resource solution is taken into account, it is conducive to improving the cost-effectiveness of resource supply.

[0036] In addition, the solution provided by the embodiment of the present application also helps cloud service providers to effectively manage the underlying resource supply capabilities and low-level available resources, solve the problem of optimal use of resources, and allow the underlying resources to create greater efficiency and value.

[0037] Furthermore, the solution provided by the embodiments of the present application can also be combined with the conventional cloud product supply solution of the cloud service provider. On the basis of the conventional cloud product supply solution, the underlying resource supply capabilities and cloud resources can be discovered and utilized. When the conventional cloud product supply solution cannot meet the user's cloud product needs, cloud products that meet the user's needs can be automatically produced for the user in an efficient and fast manner, thereby improving the user experience and increasing the user's stickiness to the cloud service provider.

[0038] The technical solutions provided by various embodiments of the present application are described in detail below in conjunction with the accompanying drawings.

[0039] Figure 1a This is a schematic diagram of the structure of a cloud product data processing system provided in an embodiment of the present application. Figure 1a As shown, the system 100 includes: a demand management module 10 , a resource orchestration module 20 and an execution delivery module 30 .

[0040] In the embodiments of the present application, the deployment and implementation forms of the demand management module 10, the resource orchestration module 20 and the execution delivery module 30 are not limited. The three modules of the demand management module 10, the resource orchestration module 20 and the execution delivery module 30 can be integrated and deployed on the same physical machine, or they can be dispersedly deployed on multiple physical machines. The physical machine can be a server device such as a conventional server, a server cluster, a cloud server, or various terminal devices such as a laptop computer and a smart phone. In addition, the demand management module 10, the resource orchestration module 20 and the execution delivery module 30 can also be deployed in a virtualization instance, such as a virtual machine (VM), a container, etc., which can be centrally deployed in the same virtualization instance, or they can be dispersedly deployed in multiple virtualization instances, and there is no limitation on this. In the embodiments of the present application, the focus is on the functions achieved by the cooperation between the three modules of the demand management module 10, the resource orchestration module 20 and the execution delivery module 30.

[0041] In an embodiment of the present application, the demand management module 10 serves as the demand management entrance of the system 100 and can receive various cloud product demands. These cloud product demands can come from users of cloud products, operation and maintenance personnel within the cloud service provider, the traditional cloud product supply system of the cloud service provider, or a third-party cloud service provider.

[0042] For example, the system 100 provided in the embodiment of the present application may be open to users of cloud products, and the demand management module 10 provides a first demand provision portal to users of cloud products, and the first demand provision portal may be a web page, an application page, or a command window, etc.; users of cloud products may submit cloud product demand information through the first demand provision portal; the demand management module 10 may receive cloud product demand information submitted by users of cloud products through the first demand submission portal. The cloud product demand information describes the user's demand information for cloud products, including the user's description information of the cloud product, and the description information is used to describe at least one of the type, resource specifications, functions, start time, coverage area, etc. of the cloud product required by the user.

[0043] For another example, the system 100 provided in this embodiment can be open to the operation and maintenance personnel of the cloud service provider. The demand management module 10 provides a second demand provision portal for the operation and maintenance personnel of the cloud service provider. The second demand provision portal can be a web page, an application page, or a command window, etc. The operation and maintenance personnel of the cloud service provider can submit cloud product demand information through the second demand provision portal. The demand management module 10 can receive the cloud product demand information submitted by the operation and maintenance personnel of the cloud service provider through the second demand submission portal. The cloud product demand information describes the demand information of the operation and maintenance personnel for the cloud product, including the description information of the operation and maintenance personnel for the cloud product, and the description information is used to describe at least one of the type, resource specifications, functions, start time, coverage area, etc. of the cloud product required by the operation and maintenance personnel. Among them, the operation and maintenance personnel can generate cloud product demand in various operation and maintenance projects.

[0044] For another example, the system 100 provided in this embodiment can be connected with the traditional cloud product supply system of the cloud service provider, specifically, the demand management module 10 is connected with the traditional cloud product supply system. Among them, the traditional cloud product supply system provides various ready-made cloud products for users, and users can purchase the required cloud products through the cloud product supply system. In the case of user purchase failure, the traditional cloud product supply system will collect the user's purchase failure information, and convert the user's purchase information of the cloud product to the cloud product demand information and provide it to the demand management module 10. The demand management module 10 can receive the cloud product demand information provided by the traditional cloud product supply system when the user fails to purchase the cloud product, so that the system 100 can automatically produce the cloud product required by the user. This case is combined with the conventional cloud product supply solution of the cloud service provider, and the underlying resource supply capabilities and cloud resources are discovered and utilized on the basis of the conventional cloud product supply solution. In the case that the conventional cloud product supply solution cannot meet the user's cloud product demand, the cloud product that meets the user's demand is automatically produced for the user in an efficient and fast manner, thereby improving the user experience and increasing the user's stickiness to the cloud service provider.

[0045] For another example, the cloud service provider to which the system 100 provided in this embodiment belongs can cooperate with a third-party cloud service provider to jointly provide various cloud products for users. Based on this, the system 100 provided in this embodiment can be connected with the relevant systems or devices of the third-party cloud service provider, specifically, the demand management module 10 is connected with the relevant systems or devices of the third-party cloud service provider to receive the cloud product demand information provided by the third-party cloud service provider. Optionally, the third-party cloud service provider can provide the user's cloud product demand information to the demand management module 10 when it is unable to provide the required cloud product to the user; or, the third-party cloud service provider provides the cloud product demand of a specific user to the demand management module 10, and the cloud product demand of a specific user can be determined by the collaboration between the two parties. This case is combined with a third-party cloud service provider, which is conducive to enhancing the cooperation between the two parties, and can also automatically produce cloud products that meet the user's needs for the user in an efficient and fast manner, improve the user experience, and increase the user's stickiness to the cloud service provider.

[0046] In the embodiments of the present application, the format of cloud product requirements is not limited. It can be standardized cloud product requirement information or non-standardized cloud product requirement information, and non-standardized cloud product requirement information can be converted into standardized cloud product requirement information. Among them, standardized cloud product requirement information refers to cloud product requirement information with clear resource requirements. For example, cloud product requirement information with clear requirements for virtualized resources, such as an ECS that requires 100 core vCPUs, can also be cloud product requirement information with clear requirements for physical resources, such as an ECS that requires 2 physical servers to carry. Correspondingly, non-standardized cloud product requirement information refers to cloud product requirement information with less clear resource requirements, such as cloud product requirement information from operation and maintenance personnel, such as requiring an ECS, but not limiting the resource specifications of the ECS. Among them, converting non-standardized cloud product requirement information into standardized cloud product requirement information mainly refers to the process of specifying resource specifications for non-standardized cloud product resource requirement information, such as clarifying the number of vCPU cores required for the ECS for cloud product requirement information that requires an ECS, such as 10 core vCPUs.

[0047] In an embodiment of the present application, the demand management module 10 can parse the cloud product demand information into K candidate resource deployment schemes, where K is a positive integer. Each candidate resource deployment scheme refers to a resource deployment scheme corresponding to the cloud product demand information, and the resource deployment scheme is used to describe which resources are deployed in which availability zones, and the cloud product corresponding to the cloud product demand information can be obtained. Cloud products are products that use cloud resources and can provide certain cloud services, such as ECS, load balancing (SLB) services, cloud databases, object storage services, video on demand services, domain name services, log service systems, SMS services, etc. In this case, the same cloud product demand information corresponds to multiple resource deployment schemes, and it is necessary to select one resource deployment scheme from them, and perform resource deployment according to the selected resource deployment scheme to obtain the cloud product. For the sake of ease of description and distinction, each resource deployment scheme parsed by the demand management module 10 is referred to as a candidate resource deployment scheme, and the resource deployment scheme subsequently selected from the multiple candidate resource deployment schemes is referred to as a target resource deployment scheme. Each candidate resource deployment scheme includes at least the resource information to be deployed and the availability zone information to be deployed. The resource information to be deployed is used to describe the resource information required when the candidate resource deployment scheme is adopted to implement the cloud product, including the resource type, resource specifications, etc. The availability zone information to be deployed is used to describe the information of the availability zone where the cloud product is located when the candidate resource deployment scheme is adopted to implement the cloud product. Availability zones refer to physical data centers with independent power and network in the same geographical area, and fault isolation can be achieved between availability zones.

[0048] In the embodiment of the present application, the internal implementation structure of the demand management module 10 is not limited. In an optional embodiment, Figure 1b As shown, the demand management module 10 includes: a demand entry module 11, a demand conversion unit 12 and a demand parsing unit 13.

[0049] Among them, the demand entry unit 11 is used to receive standardized cloud product demand information and / or non-standardized cloud product demand information. The demand conversion unit 12 is used to convert non-standardized cloud product demand information into standardized cloud product demand information. For standardized cloud product demand information and non-standardized cloud product requirements, and the conversion process between the two, please refer to the description above and will not be repeated here. The demand parsing unit 13 is interconnected with the demand entry unit 11 and the demand conversion unit 12, and is used to generate at least one candidate resource deployment plan according to the clear resource requirements corresponding to the standardized cloud product demand information provided by the demand entry unit 11 or the demand conversion unit 12. For the convenience of description, in the following, at least one candidate resource deployment plan is recorded as K candidate resource deployment plans, and K is a positive integer.

[0050] In an optional embodiment, the demand analysis unit 13 is specifically used to: determine X1 types of resource information and Y1 types of availability zone information based on the clear resource requirements corresponding to the standardized cloud product demand information; generate K types of candidate resource deployment plans based on the X1 types of resource information and Y1 types of availability zone information, where X1 and Y1 are positive integers.

[0051] Further optionally, the demand parsing unit 13 may directly combine X1 types of resource information and Y1 types of available zone information, and use the combination of each type of resource information and each type of available zone information as a candidate resource deployment scheme, thereby obtaining K types of candidate resource deployment information. Alternatively, the demand parsing unit 13 may filter X1 types of resource information and Y1 types of available zone information respectively according to the attributes of the resource information and the attributes of the available zone information to obtain X2 types of resource information and Y2 types of available zone information, where X2 and Y2 are positive integers, and X2≤X1, Y2≤Y1; for example, resource information whose resource specifications obviously do not meet the requirements (for example, resource information whose resource specifications exceed the upper limit or resource information whose resource specifications are less than the lower limit) may be filtered out, and available zone information that obviously does not meet the requirements (for example, the geographical location is too remote or the available zone specifications are too small or too large) may also be filtered out; then, the X2 types of resource information and the Y2 types of available zone information are combined to obtain P types of candidate resource deployment schemes, where P is a positive integer and P≥K; further, based on the M types of resource supply capabilities and / or resource capability information of the available zones, the P types of candidate resource deployment schemes are filtered to obtain K types of candidate resource deployment schemes, where M is a positive integer and M≤N. Among them, based on the M resource supply capabilities and / or resource capacity information of the available zones, the P candidate resource deployment solutions are filtered, mainly from the resource supply capabilities and / or resource capacity information of the available zones, the candidate resource deployment solutions that exceed the resource supply capabilities, exceed the resource capacity information of the available zones, or the resource function capabilities do not match the resource capacity information of the available zones are filtered out. For example, assuming that a candidate resource deployment solution includes resource information with a resource specification of 10,000 core vCPUs and available zone A, and the resource capacity upper limit of available zone A is to provide 8,000 core vCPUs, then the candidate resource deployment solution is obviously not feasible and should be filtered out.

[0052] After obtaining K candidate resource deployment solutions, the demand management module 10 provides the K candidate resource deployment solutions to the resource scheduling module 20. The resource scheduling module 20 performs resource scheduling and resource loss cost information estimation processing on the K candidate resource deployment solutions with the help of at least one resource supply capability provided by the cloud service provider, and obtains the target resource deployment solution and its corresponding target resource solution whose resource loss cost information meets the requirements. Similarly, for the convenience of description, at least one resource supply capability is hereinafter referred to as N resource supply capabilities, where N is a positive integer.

[0053] In this embodiment, resource orchestration mainly refers to the process of finding resource solutions for various candidate resource deployment solutions with the help of the underlying resource supply capabilities of the cloud service provider. Resource supply capabilities are any underlying capabilities provided by the cloud service provider that can provide idle resources. They can also be understood as ways to provide idle resources, such as instance migration capabilities (migrated instances will release the resources originally occupied), purchasing capabilities for physical servers (purchased physical servers can provide new resources), turnover inventory capabilities, mixed operation, variable configuration capabilities, temporary resource sales capabilities, etc. Among them, the turnover inventory capability refers to the ability to temporarily schedule resources in the resource library that are below the water level for emergency use. Mixed operation refers to the ability to temporarily generate resource specifications that are not originally supported on a physical machine and allow resources of this specification to be provided within a certain range. Variable configuration capability refers to the ability to take idle physical machines offline, reconfigure the underlying virtualization templates, and deploy them online to provide new specifications of resources to the outside world. The temporary resource selling capability can also be called SPOT (literally translated as temporary substitute) capability. This capability allows some resources outside the safety period to be temporarily sold to other users. After a set time, such as 1 hour, if no protection period is set, the resource may be recycled at any time. Therefore, the cost of using such resources is relatively low.

[0054] Among them, the resource solution of the candidate resource deployment scheme refers to a solution that provides the required resource information to be deployed for the candidate resource deployment scheme, and the resource solution includes at least one resource supply capability, each of the at least one resource supply capability is responsible for providing at least part of the resource information to be deployed in the candidate resource deployment scheme, and the sum of at least part of the resource information provided by each of the at least one resource supply capability is greater than or equal to the resource information to be deployed in the candidate resource deployment scheme. In other words, the resource information to be deployed in a candidate resource deployment scheme can be provided by one or more resource supply capabilities provided by the cloud service provider, and the one or more resource supply capabilities responsible for providing resource information for the candidate resource deployment scheme and the resource information each is responsible for providing form the resource solution of the candidate resource deployment scheme. For example: Assuming that the resource information (i.e., the resource information to be deployed) required by a candidate resource deployment scheme is 100 cores of vCPU, the resource solution of the candidate resource deployment scheme obtained after resource orchestration includes: using the instance migration capability to obtain 60 cores of vCPU, using the mixed operation to obtain 20 cores of vCPU, and using the variable configuration capability to obtain 20 cores of vCPU.

[0055] In this embodiment, resource loss cost information refers to the relevant resource loss cost information involved in the process of finding resource solutions for various candidate resource deployment solutions, and these resource loss cost information at least include the resource loss cost information generated by the resource solutions corresponding to the various candidate resource deployment solutions. In this embodiment, based on the resource loss cost information, a target resource solution whose resource loss cost information meets the requirements is selected and the candidate resource deployment solution corresponding to the target resource solution is obtained as the target resource deployment solution. In the embodiment of the present application, the implementation method of selecting the target resource solution whose resource loss cost information meets the requirements is not limited, and can be flexibly implemented according to the application scenario. For example, in one application scenario, a resource solution with the smallest resource loss cost information can be selected as the target resource solution; in another application scenario, a resource solution with resource loss cost information less than a set cost threshold can be selected as the target resource solution; in another application scenario, a resource solution with resource loss cost information within a set cost range can be selected as the target resource solution.

[0056] It is explained here that the process of selecting the target resource solution is also the process of selecting the target resource deployment plan. The target resource solution can be selected, and the target resource deployment plan is then determined by the target resource solution. Alternatively, the target resource deployment plan can be selected, and the target resource solution can be determined by the target resource deployment plan. Both description methods are supported by the embodiments of the present application.

[0057] After obtaining the target resource deployment scheme and its corresponding target resource solution, the resource orchestration module 20 can provide the target resource deployment scheme and its corresponding target resource solution to the execution delivery module 30. The execution delivery module 30 can execute resource processing tasks in the available zone corresponding to the available zone information to be deployed in the target resource deployment scheme according to the target resource solution, so as to obtain the cloud product corresponding to the cloud product demand information. The resource processing task here mainly refers to the process of reserving corresponding resource information according to the various resource supply capabilities in the target resource solution. In the case where all the resource information to be deployed in the target resource deployment scheme is successfully reserved, these successfully reserved resource information can be assembled into cloud products corresponding to the cloud resource demand. For example, a 60-core vCPU reserved using instance migration capability, a 20-core vCPU reserved using mixed operation, and a 20-core vCPU reserved using variable configuration capability can be assembled into an ECS with 100 core vCPUs.

[0058] In the embodiments of the present application, by leveraging the underlying resource supply capabilities of the cloud service provider, the required cloud products can be provided to users quickly and on demand, thereby achieving elastic resource supply; in addition, since the resource loss cost of the resource solution is taken into consideration, it is beneficial to improve the cost-effectiveness of resource supply.

[0059] In the embodiment of the present application, the internal implementation structure of the resource scheduling module 20 is not limited, nor is the implementation method of the resource encoding module 20 performing resource scheduling and resource loss cost information estimation processing on K candidate resource deployment schemes according to N resource supply capabilities. Figure 1c This is an internal implementation structure of the resource arrangement module 20 provided in the embodiment of the present application. Figure 1c The structure shown in FIG. 1 is a detailed description of a detailed implementation method in which the resource encoding module 20 performs resource arrangement and resource loss cost information estimation processing on K candidate resource deployment schemes according to N resource supply capabilities.

[0060] like Figure 1c As shown, the resource scheduling module 20 includes: a first resource scheduling unit 21, a second resource scheduling unit 22 and a solution selection unit 23.

[0061] The first resource scheduling unit 21 is used to estimate, for any candidate resource deployment scheme, unit resource loss cost information when N types of resource supply capabilities provide the required resources for any candidate resource deployment scheme based on the resource information to be deployed and the available zone information to be deployed in any candidate resource deployment scheme.

[0062] The second resource scheduling unit 22 is used to generate a resource solution and its resource loss cost information corresponding to any candidate resource deployment scheme based on the unit resource loss cost information when the N types of resource supply capabilities provide the required resources for any of the candidate resource deployment schemes and the resource information to be deployed in any of the candidate resource deployment schemes.

[0063] The solution selection unit 23 is used to select a target resource deployment solution and its corresponding target resource solution whose resource loss cost information meets the requirements according to the resource loss cost information of the resource solution corresponding to each of the K candidate resource deployment solutions.

[0064] In detail, for each candidate resource deployment scheme, the first resource scheduling unit 21 needs to use N types of resource supply capabilities to provide the required resources for the candidate resource deployment scheme respectively, and estimate the unit resource loss cost information generated when each resource supply capability provides the required resources for the candidate resource deployment scheme. Among them, the unit resource loss cost information refers to the resource cost consumed by a resource supply capability to provide the required unit resources for a candidate resource deployment scheme. In the embodiment of the present application, the unit or specification of the unit resource is not limited, and it can be determined according to the resource type. For example, if the resource is a vCPU, the unit resource can be a single vCPU. If the resource is a memory resource, the unit resource can be a 1G memory resource, and there is no limitation on this.

[0065] For the convenience of description and distinction, any candidate resource deployment scheme A is taken as an example for description. For any candidate resource deployment scheme A, the first resource scheduling unit 21 estimates the unit resource loss cost information when N types of resource supply capabilities respectively provide the required resources for any candidate resource deployment scheme A based on the resource information to be deployed and the available area information to be deployed in any candidate resource deployment scheme A, and provides the N unit resource loss cost information corresponding to the N types of resource supply capabilities to the second resource scheduling unit 22.

[0066] In an optional embodiment, the first resource scheduling unit 21 estimates the unit resource loss cost information when N types of resource supply capabilities provide the required resources for any candidate resource deployment scheme A, including: for any type of resource supply capability, obtaining the target resource information and target time information consumed when any type of resource supply capability provides the required resources for any candidate resource deployment scheme A; generating the target resource loss cost information when any type of resource supply capability provides the required resources for any candidate resource deployment scheme A based on the target resource information and the target time information; and calculating the unit resource loss cost information when any type of resource supply capability provides the required resources for any candidate resource deployment scheme A based on the target resource loss cost information and the resource information to be deployed in any candidate resource deployment scheme A.

[0067] Among them, for any resource supply capability, the amount of resources consumed by it to provide the required resources for any candidate resource deployment scheme A is greater than or equal to the amount of resources required by any candidate resource deployment scheme A (the amount of resources is the amount of resources in the resource information to be deployed in any candidate resource deployment scheme A), and it needs to prepare resources to provide the required resources for any candidate resource deployment scheme A, which takes a certain amount of time. In an embodiment of the present application, the resource loss cost information generated by any resource supply capability providing the required resources for any candidate resource deployment scheme A is measured by the target resource information and target time information consumed when any resource supply capability provides the required resources for any candidate resource deployment scheme A. For the convenience of distinction and description, the resource loss cost information generated when any resource supply capability provides the required resources for any candidate resource deployment scheme A is referred to as target resource loss cost information.

[0068] In the embodiment of the present application, the first resource scheduling unit 21 is not limited to the way in which any resource supply capability consumes target resource information and target time information when providing the required resources for any candidate resource deployment solution A. An optional implementation is given below, which is as follows:

[0069] After the cloud product demand information is parsed into K candidate resource deployment schemes, the K candidate resource deployment schemes can wait to be scheduled. There is no limitation on how to schedule the K resource deployment schemes. They can be randomly scheduled, or they can be scheduled in order from small to large according to the identifiers of the candidate resource deployment schemes, or they can be scheduled in order from small to large according to the resource information to be deployed in the candidate resource deployment schemes, and so on. In this way, any candidate resource deployment scheme A will have a scheduling processing start time. In addition, any candidate resource deployment scheme A can also include the expected resource use start time. Based on this, the first resource scheduling unit 21 can determine the resource waiting time of any resource supply capability according to the scheduling processing start time of any candidate resource deployment scheme A, the resource processing time of any resource supply capability, and the resource use start time in any candidate resource deployment scheme A. The target time information includes the resource processing time and the resource waiting time. Among them, the resource processing time of any resource supply capability can be determined according to its supply capability and the amount of resources to be deployed in any candidate resource deployment scheme A. In addition, the first resource scheduling unit 21 can obtain the effective resource information and resource fragment information generated when any resource supply capability provides the required resources for any candidate resource deployment plan A based on the resource information to be deployed in any candidate resource deployment plan A, and the target resource information includes effective resource information and resource fragment information. Effective resource information refers to the minimum resource information required to ensure the resources required by any candidate resource deployment plan A, and the amount of resources in the effective resource information is greater than or equal to the amount of resources to be deployed in any candidate resource deployment plan A; resource fragment information refers to resource fragments that have to be generated in order to generate effective resource information. For example, 300 vCPU cores can be opened on a physical machine. For a scenario that requires 1000 vCPU cores, 4 physical machines can be opened with a total of 1200 vCPU cores. 1000 vCPU cores are actually required, and the other 200 vCPU cores are resource fragments.

[0070] Assume that any candidate resource deployment plan A requires 1000 vCPU cores. Take the use of mixed-open capability to provide the required resources for any candidate resource deployment plan A as an example. Assume that the candidate resource deployment plan A starts to be used after 7 days. The process of using mixed-open capability to provide the required resources for any candidate resource deployment plan A takes 5 minutes. There is a 2% risk. The execution success rate is expected to be 95%, and the effective resource information is 1200 vCPU cores and 100 vCPU cores. Resource fragments.

[0071] After obtaining the target resource information and target time information consumed when any resource supply capability provides the required resources for any candidate resource deployment solution A, the first resource scheduling unit 21 can generate the target resource loss cost information when any resource supply capability provides the required resources for any candidate resource deployment solution A according to the target resource information and the target time information. Optionally, the first resource scheduling unit 21 can perform certain numerical processing on the target resource information and the target time information, such as normalization, and after normalization, perform weighted summation of the normalized target resource information and the target time information using a set resource cost parameter, etc., to obtain the target resource loss cost information when any resource supply capability provides the required resources for any candidate resource deployment solution A.

[0072] In another optional embodiment, the risks of each resource supply capability when providing the required resources for any candidate resource deployment solution A, the resource information that must be additionally provided to ensure the satisfaction rate, and the filtering conditions that should be met can also be considered, so as to more accurately estimate the target resource loss cost information when any resource supply capability provides the required resources for any candidate resource deployment solution A. Based on this, the first resource scheduling unit is also used to: obtain at least one of the risk probability corresponding to any resource supply capability providing the required resources for any candidate resource deployment solution A, the proportion of additional resources to ensure the satisfaction rate, and / or the filtering conditions.

[0073] Further optionally, the first resource scheduling unit 21 obtains the corresponding risk probability when any resource supply capability provides the required resources for any candidate resource deployment scheme A, including: generating the risk probability according to the impact information on other resource objects and / or other users when any resource supply capability provides the required resources for any candidate resource deployment scheme A. Taking the instance migration capability as an example, in order to provide the required resources for any candidate resource deployment scheme A, instance migration is required, wherein the migrated instance, the original physical machine and the target physical machine where the migrated instance is located, and the user to whom the migrated instance belongs, etc. will be affected by the migration operation, which poses a certain risk to the system, and the risk probability can be determined according to the impact information on other resource objects and / or other users when any resource supply capability provides the required resources for any candidate resource deployment scheme A.

[0074] Further optionally, the first resource scheduling unit 21 obtains the proportion of additional resources to ensure the satisfaction rate when any resource supply capability provides the required resources for any candidate resource deployment scheme A, including: generating the execution success rate when any resource supply capability provides the required resources for any candidate resource deployment scheme A according to the historical deployment data of the available zone corresponding to the available zone information to be deployed in any candidate resource deployment scheme A; and calculating the proportion of additional resources to ensure the satisfaction rate according to the user's expected satisfaction rate and execution success rate. The execution success rate here refers to the probability that any resource supply capability can successfully provide the required resources for any candidate resource deployment scheme A, which can be obtained by statistical analysis based on the historical deployment data of the available zone corresponding to the available zone information to be deployed in any candidate resource deployment scheme A.

[0075] Based on the above, the first resource scheduling unit 21 generates an optional implementation method of target resource loss cost information when any resource supply capability provides the required resources for any candidate resource deployment plan A according to the target resource information and the target time information, including: generating target resource loss cost information according to the risk probability, the proportion of additional resources to ensure the satisfaction rate and / or at least one of the filtering conditions, as well as the effective resource information, resource fragmentation information, resource processing time and resource waiting time.

[0076] Further optionally, the first resource scheduling unit 21 generates target resource loss cost information according to the risk probability, the proportion of additional resources to ensure the satisfaction rate and the filtering conditions, as well as the effective resource information, resource fragmentation information, resource processing time and resource waiting time, including: generating resource processing cost information according to the effective resource information, resource fragmentation information and resource processing time; respectively correcting the resource processing cost information according to the risk probability and the proportion of additional resources to ensure the satisfaction rate to obtain the first intermediate state cost information and the second intermediate state cost information; generating resource waiting cost information according to the resource information to be deployed and the resource waiting time in any candidate resource deployment scheme A; generating target resource loss cost information according to the first intermediate state cost information, the second intermediate state cost information, the resource waiting cost information and the filtering conditions. In this embodiment, the implementation method of generating resource processing cost information according to effective resource information, resource fragmentation information and resource processing time is not limited, nor is the implementation method of correcting the resource processing cost information according to the risk probability and the proportion of additional resources to ensure the satisfaction rate to obtain the first intermediate state cost information and the second intermediate state cost information is not limited. Please refer to the examples in the subsequent scenario embodiments.

[0077] Further optionally, the first resource orchestration unit 21 generates target resource loss cost information based on the first intermediate state cost information, the second intermediate state cost information, the resource waiting cost information and the filtering condition, including: taking a weighted sum of the first intermediate state cost information, the second intermediate state cost information and the resource waiting cost information to obtain initial resource loss cost information; when any resource supply capability meets the filtering condition, using the initial resource loss cost information as the target resource loss cost information; when any resource supply capability does not meet the filtering condition, using the preset resource loss cost information as the target resource loss cost information.

[0078] Further optionally, the system 100 may also include a confidence module, which is used to calculate the confidence of any resource supply capability based on the resource information actually consumed when any resource supply capability provides the required resources for any candidate resource deployment solution and the resource information to be deployed in any candidate resource deployment solution; and adjust the weight information required for calculating the resource loss cost information corresponding to any resource supply capability based on the confidence. Due to the existence of risk probability and the need to ensure the satisfaction rate and other requirements, the resource information actually consumed when any resource supply capability provides the required resources for any candidate resource deployment solution is greater than the resource information to be deployed in any candidate resource deployment solution, including at least the effective resource information, resource fragmentation information, and resource information consumed additionally to ensure the satisfaction rate generated by any resource supply capability.

[0079] After obtaining the target resource loss cost information when any resource supply capability provides the required resources for any candidate resource deployment scheme A, the first resource scheduling unit 21 can calculate the unit resource loss cost information when any resource supply capability provides the required resources for any candidate resource deployment scheme A according to the target resource loss cost information and the resource information to be deployed in any candidate resource deployment scheme A. For example, by dividing the target resource loss cost information by the amount of resources to be deployed in any candidate resource deployment scheme A, the unit resource loss cost information when any resource supply capability provides the required resources for any candidate resource deployment scheme A can be obtained.

[0080] After obtaining the unit resource loss cost information when any resource supply capability provides the required resources for any candidate resource deployment solution A, the first resource scheduling unit 21 may provide the unit resource loss cost information when any resource supply capability provides the required resources for any candidate resource deployment solution A to the second resource scheduling unit 22. After obtaining the unit resource loss cost information when N types of resource supply capabilities respectively provide the required resources for the candidate resource deployment solution A, the second resource scheduling unit 22 may generate a resource solution corresponding to the candidate resource deployment solution A and its resource loss cost information according to the unit resource loss cost information when the N types of resource supply capabilities respectively provide the required resources for the candidate resource deployment solution A and the resource information to be deployed in the candidate resource deployment solution A.

[0081] Specifically, the second resource orchestration unit 22 is specifically used to: determine at least one resource supply capability required to meet the resource information to be deployed in any candidate resource deployment scheme A and at least part of the resource information that it is responsible for providing, as the resource solution corresponding to any candidate resource deployment scheme A, based on the resource information to be deployed in any candidate resource deployment scheme A and the upper limit resource information that each of the N resource supply capabilities can provide; and further, generate the resource loss cost information of the resource solution corresponding to any candidate resource deployment scheme A based on the unit resource loss cost information corresponding to at least one resource supply capability and at least part of the resource information that it is responsible for providing. For example, assuming that the amount of resources to be deployed in any candidate resource deployment plan A is 100 vCPU cores, among the N resource supply capabilities, the instance migration capability can provide up to 60 vCPU cores, the mixed operation can provide up to 20 vCPU cores, the variable configuration capability can provide up to 20 vCPU cores, the turnover inventory capability can provide up to 30 vCPU cores, and the procurement capability can provide up to 40 vCPU cores. In view of this, the instance migration capability, mixed operation and variable configuration capability can be selected. The instance migration capability is responsible for providing 60 vCPU cores, the mixed operation is responsible for providing 20 vCPU cores, and the variable configuration capability is responsible for providing 20 vCPU cores, and the resource solution for any candidate resource deployment plan A is obtained. Furthermore, the resource loss cost information is obtained based on the unit resource loss cost information corresponding to the instance migration capability and the 60-core vCPU it is responsible for providing; the resource loss cost information is obtained based on the unit resource loss cost information corresponding to the mixed operation and the 20-core vCPU it is responsible for providing; the resource loss cost information is obtained based on the unit resource loss cost information corresponding to the variable configuration capability and the 20-core vCPU it is responsible for providing. The sum of the three parts of resource loss cost information is the resource loss cost information of the resource solution of any candidate resource deployment plan A.

[0082] After obtaining the resource solution and its resource loss cost information corresponding to any candidate resource deployment scheme A, the second resource scheduling unit 22 provides the resource solution and its resource loss cost information corresponding to the candidate resource deployment scheme A to the scheme selection unit 23. After receiving the resource loss cost information of the resource solution corresponding to each of the K candidate resource deployment schemes provided by the second resource scheduling unit 22, the scheme selection unit 23 can select the target resource deployment scheme and its corresponding target resource solution whose resource loss cost information meets the requirements according to the resource loss cost information of the resource solution corresponding to each of the K candidate resource deployment schemes. The detailed implementation method for selecting the target resource deployment scheme and its corresponding target resource solution can be found in the description above, which will not be repeated here.

[0083] After obtaining the target resource deployment scheme and its corresponding target resource solution, the scheme selection unit 23 may submit the target resource deployment scheme and its corresponding target resource solution to the execution and delivery module 30. The execution and delivery module 30 may execute the resource processing task in the available zone corresponding to the available zone information to be deployed in the target resource deployment scheme according to the target resource solution, so as to obtain the cloud product corresponding to the cloud product demand information.

[0084] In this embodiment, the execution delivery module 30 can manage various target resource solutions through the ready queue. Whenever a target resource solution is received, the target resource solution is added to the ready queue, and each resource solution in the ready queue is scheduled according to the priority of the resource solution, wherein the priority of the target resource solution is determined according to the resource scenario to which the cloud product demand information belongs. Further, when the target resource solution is scheduled, the resource processing task is executed in the available zone corresponding to the available zone information to be deployed in the target resource deployment plan according to the target resource solution, so as to obtain the cloud product corresponding to the cloud product demand information.

[0085] In this embodiment, in order to prevent a resource solution in the ready queue from being preempted for scheduling opportunities and not being processed due to its low priority, the execution interaction module 30 adopts a dynamic priority solution, that is, each resource solution will have an original priority, which is determined according to the resource scenario to which the cloud product demand information belongs, and the priority will be dynamically adjusted as the waiting time for scheduling increases, so that resource solutions with longer waiting times can be scheduled. Based on this, the execution interaction module 30 is also used in the process of scheduling each resource solution in the ready queue, and is also used to monitor the waiting time of each resource solution in the ready queue, and when a first resource solution whose waiting time exceeds the corresponding waiting time threshold is detected, the priority of the first resource solution is increased. The waiting time threshold is dynamically determined according to the priority of the first resource solution before adjustment, and will change with the dynamic change of the priority of the first resource solution. The higher the priority, the smaller the waiting time threshold. Assume that the priority range of a resource solution is 1-99, and the smaller the number, the higher the priority; assume that the initial priority of a resource solution is 60 based on the resource scenario, and the waiting time threshold corresponding to the priority 60 is 30 minutes. If the resource solution is still not scheduled after waiting for 30 minutes, the priority of the resource solution can be increased, for example, to 20, and the waiting time threshold corresponding to the priority 20 is 15 minutes. The resource solution needs to be scheduled within the next 15 minutes, otherwise its priority will continue to increase until the resource solution is scheduled. Among them, the resource solution with a priority of 1 will be scheduled first.

[0086] Further, in the present embodiment, when the execution delivery module 30 is scheduled to the target resource solution, taking into account that the target resource solution contains at least one resource supply capability, the at least one resource supply capability may be optionally scheduled in order according to the priority of the at least one resource supply capability. Optionally, the priority of the at least one resource supply capability may be determined based on the unit resource loss cost information corresponding to the at least one resource supply capability. For example, the smaller the unit resource loss cost information, the higher the priority of the corresponding resource supply capability. In this case, the execution delivery module 30 may give priority to scheduling resource supply capabilities with smaller unit resource loss cost information, and control the scheduled resource supply capabilities to reserve resources for at least part of the resource information they are responsible for providing; when the resource reservation results corresponding to the at least one resource supply capability are all successful, the resources reserved by the at least one resource supply capability are assembled into cloud products corresponding to the cloud product demand information.

[0087] Furthermore, in the process of sequentially scheduling each resource supply capability, if the scheduled resource supply capability fails to perform resource reservation, the execution delivery module 30 can calculate the additional resource amount according to the execution success rate of the failed resource supply capability and the execution success rate of the next resource supply capability; when scheduling the next resource supply capability, control the next resource supply capability to perform resource reservation according to at least part of the resource information and the additional resource amount that the next resource supply capability is responsible for providing. For the convenience of description, this scheduling process can be called reduced scheduling.

[0088] Further optionally, when the demand management module 10 receives the cloud product demand information, it can also add a scenario tag to the cloud product demand information. Different scenario tags correspond to different resource scenarios, allowing the scheduling priority, execution topology, downgrade strategy, delivery strategy, resource reservation strategy, retry strategy, orchestration strategy, approval strategy, callback strategy, etc. to be pre-configured for each resource scenario. For example, the delivery strategy can constrain the maximum number of deliveries of the target resource solution, for example, if the delivery is unsuccessful once, it is allowed to be delivered again. The resource reservation strategy is used to constrain the behavior of resource reservation for each resource supply capability in the target resource solution, such as limiting the maximum number of repeated resource reservations. The downgrade strategy is used to constrain the scheduling method of each resource supply capability in the target resource solution (see the reduced execution part in the following embodiment), etc. The retry strategy refers to a strategy that repeats the target resource solution output step until the target resource solution location that meets the cloud product requirements is obtained when the output target resource solution does not meet the cloud product requirements. The callback strategy refers to a strategy that allows the execution of the target resource solution to be viewed through a callback function after the target resource solution is delivered for execution. The orchestration strategy refers to the strategy for how to orchestrate resources. Optionally, the orchestration strategy can produce Figure 2a The combination of resource supply capabilities and resource deployment solutions shown, etc.

[0089] For ease of understanding, the processing process of the above cloud product data processing system is described in detail below in combination with specific application scenarios.

[0090] like Figure 1d-Figure 1eAs shown, when a user or operation and maintenance personnel has a cloud product demand, they can trigger a cloud product demand operation containing standardized cloud product demand information and / or non-standardized cloud product demand information to the cloud product data processing system. The demand entry unit receives the standardized cloud product demand information and / or non-standardized cloud product demand information and transmits it to the demand conversion unit. The demand conversion unit converts the non-standardized cloud product demand information into standardized cloud product demand information with clear resource requirements, and transmits the standardized cloud product demand information to the demand parsing unit. The demand parsing unit determines X1 types of resource information and Y1 types of availability zone information according to the clear resource requirements corresponding to the standardized cloud product demand information; further, the X1 types of resource information and Y1 types of availability zone information are filtered to obtain X2 types of resource information and Y2 types of availability zone information, and the X2 types of resource information and Y2 types of availability zone information are combined to obtain P types of candidate resource deployment schemes; further, among the P types of resource candidate deployment schemes, there may be a situation where at least one resource supply capability does not match the resource capability of the availability zone. Based on this, the P types of candidate resource deployment schemes can be filtered based on the resource capability information of at least one Provider and / or availability zone to obtain K types of candidate resource deployment schemes.

[0091] The generation process of candidate resource deployment solutions is as follows: Figure 2a As shown, the resource information includes at least specification 1, specification 2, ... specification X, and the availability zone information includes at least availability zone 1, availability zone 2, ... availability zone Y. Each resource information in the X types of resource information can be matched with each availability zone in the Y types of availability zones, and X*Y (P) candidate resource deployment solutions are obtained. Each candidate resource deployment solution can be represented by "<Specification X, Availability Zone Y>". Further, assuming that the memory required by the cloud resource of specification 1 is greater than the memory that can be provided by availability zone 1, that is, "<Specification 1, Availability Zone 1>" belongs to the mismatched resource mismatched resource deployment solution, so the resource deployment solutions whose at least one resource supply capability does not match the resource capability of the availability zone can be filtered to obtain the final K candidate resource deployment solutions.

[0092] Furthermore, the cloud service provider provides N types of resource supply capabilities. In this scenario embodiment, the resource supply capability is recorded as Provider, for example Figure 2a In the resource providers Provider1, Provider2, ..., ProviderN, each candidate resource deployment scheme in the K candidate resource deployment schemes can be combined with each Provider in the N Providers in pairs to obtain K*N solutions based on a single Provider.

[0093] The following is the resource loss cost information for the Provider-based solution:

[0094] Time cost indication:

[0095] like Figure 2b As shown, for any Provider, the scheduling process starts at time T according to any candidate resource deployment solution. dis.star 、The resource processing time of any Provider h loss and the resource usage start time T in any candidate resource deployment solution use.star , determine the resource waiting time h of any Provider waste The resource waiting time of any Provider is h waste , the following formula can be used but is not limited to:

[0096] h waste =T use.star -T dis.star -h loss

[0097] Since the target time information includes the resource processing time h loss and resource waiting time h waste , so the target time information can be used but not limited to the following formula:

[0098] h loss+ h waste

[0099] Resource cost diagram:

[0100] According to the resource information to be deployed in any candidate resource deployment scheme, obtain the effective resource information occupied by any Provider when providing the required resources for any candidate resource deployment scheme. block And the generated resource fragmentation information n chip Since the target resource information includes effective resource information and resource fragment information, that is, the target resource information, the following formula may be used but is not limited to:

[0101] n block +n chip

[0102] According to the effective resource information n block , Resource fragment information n chip , Resource processing time h loss And the duration of resource fragmentation h chip , generate resource processing cost information C loss , resource processing cost information C loss Including resource fragmentation loss C chip and resource processing loss C block Among them, it can be assumed that the resource processing time h loss And the duration of resource fragmentation hchip same.

[0103] Among them, resource fragment loss C chip , the following formula can be used but is not limited to:

[0104] C chip =n chip *h chip *w free

[0105] Among them, resource processing loss C block , the following formula can be used but is not limited to:

[0106] C block =n block *h loss *w free

[0107] Among them, resource processing cost information C loss , the following formula can be used but is not limited to:

[0108] C loss =C chip +C block

[0109] Optionally, some Providers have certain additional costs. For Providers with additional costs, some cost factors e, e∈[1,+∞], can be multiplied, such as the separate cost of relocation and the separate cost of procurement. The higher the actual supply chain cost, the higher e, and the higher the final resource loss.

[0110] Risk cost illustration:

[0111] Generate a risk probability P based on the impact information on other resource objects such as physical machines, virtual machines, and / or other users when any Provider provides the required resources for any candidate resource deployment solution. risk , where the risk probability and the probability of risk occurrence p risk It is related to the loss multiple value caused by the risk. The loss multiple value can be 100 times, but is not limited to this. Risk probability P risk , the following formula can be used but is not limited to:

[0112] P risk =100*p risk

[0113] Satisfaction rate cost diagram:

[0114] According to the historical deployment data of the available zone corresponding to the available zone information to be deployed in any candidate resource deployment plan, generate the execution success rate p when any Provider provides the required resources for any candidate resource deployment plan sucess , and according to the user's expected satisfaction rate p fill and execution success rate p sucess , calculated as the proportion of additional resources to ensure satisfaction rate P extra . To ensure the satisfaction rate, the proportion of additional resources is P extra , the following formula can be used but is not limited to:

[0115] P extra =p fill / p sucess

[0116] Filter conditions:

[0117] Get the filter condition Filter(rule) corresponding to any provider providing the required resources for any candidate resource deployment solution. The filter condition can be, for example, the impact of the solution on the sales rate, the impact on online turnover, whether the solution can support hot migration, whether the solution has a low historical success rate in the system's internal market, etc. These filter conditions determine whether the solution evaluation cost is meaningful. The filter condition Filter(rule) can be, but is not limited to, the following formula:

[0118]

[0119] The above filter condition Filter (rule) can be 1 when the value is satisfied, and 0 when the value is not satisfied. However, the above values ​​are only for illustrative purposes and do not limit the embodiments of the present application.

[0120] Target resource loss cost information:

[0121] According to the risk probability and the proportion of additional resources to ensure the satisfaction rate, the resource processing cost information is modified to obtain the first intermediate state cost information C″1 and the second intermediate state cost information C″2.

[0122] The first intermediate state cost information C″1 may be expressed by, but not limited to, the following formula:

[0123] C″1=C loss *P extra

[0124] The second intermediate state cost information C″2 may be expressed by, but not limited to, the following formula:

[0125] C″2=C loss *P risk

[0126] According to the resource information to be deployed and the resource waiting time in any candidate resource deployment scheme, the resource waiting cost information C″3 is generated. The resource waiting cost information C″3 can be expressed by, but not limited to, the following formula:

[0127] C″3=C free

[0128] Further, a weighted sum is taken for the first intermediate state cost information, the second intermediate state cost information and the resource waiting cost information to obtain the initial resource loss cost information C″4; when any Provider meets the filtering conditions, the initial resource loss cost information is used as the target resource loss cost information C″; when any Provider does not meet the filtering conditions, the preset resource loss cost information is used as the target resource loss cost information.

[0129] Assuming that the weights of the first intermediate state cost information, the second intermediate state cost information and the resource waiting cost information initial resource loss cost information are W1, W2 and W3 respectively, the initial resource loss cost information C″4 can be adopted but not limited to the following formula:

[0130] C″4=W1*C″1+W 2* C″2+W 3* C″3

[0131] The target resource loss cost information may be, but is not limited to, the following formula:

[0132] C″=C″4*Filter(rule)

[0133] That is, when any Provider meets the filtering condition (for example, the value is 1), the initial resource loss cost information is used as the target resource loss cost information; when any Provider does not meet the filtering condition (for example, the value is 0), the preset resource loss cost information is used as the target resource loss cost information.

[0134] Unit resource loss cost information:

[0135] After obtaining the target resource cost information, the unit resource loss cost information Score when any provider provides the required resources for any candidate resource deployment solution is calculated based on the target resource loss cost information and the resource information to be deployed in any candidate resource deployment solution, mainly referring to the amount of resources to be deployed n. The unit resource power consumption cost information Score can be adopted but not limited to the following formula:

[0136]

[0137] Wherein, Score∈(0,k], k represents the normalized value range, which can be determined according to application requirements and is a constant; normalization(C″) represents the normalization processing of C″.

[0138] Resource solutions for each candidate resource deployment solution:

[0139] Based on the above unit resource loss cost information, we can obtain the resource solutions and resource loss cost information corresponding to the K candidate resource deployment solutions. j The unit loss information can be expressed as Score j , responsible for providing at least some of the resource information for n j , the total score of the resource solution of the kth candidate resource deployment solution Value k , the following formula can be used but is not limited to:

[0140]

[0141] in, M is the number of Providers corresponding to the kth candidate resource deployment solution.

[0142] According to the resource loss cost information of the resource solutions corresponding to the K candidate resource deployment solutions, the resource solution with the smallest resource loss cost information (corresponding to the highest score) can be selected as the target resource solution Value final , the following formula can be used but is not limited to:

[0143] Value final =Max(Value k )

[0144] The following are examples of resource loss cost information generated by several providers:

[0145] Example 1: Taking turnover inventory as an example, a candidate resource deployment plan requires a total of 1000 vCPUs. Scheduling starts 7 days later. The turnover inventory generates 1000 vCPUs. There is a 2% chance of risk during the processing. The total processing time is 1 minute, and the execution success rate is expected to be 99%. Then the loss C loss =W4*C chip +W5*C block , assuming that the influence of the post-weight W1…W is not considered n =1, and the solution does not break any filtering conditions, then:

[0146] C loss =0

[0147] C free =1000*(7*24)*W free

[0148] C″=(168000W free )*Filter(rule)

[0149]

[0150] Example 2: Taking SPOT capability as an example, a candidate resource deployment plan requires a total of 1000 vCPUs. The scheduling starts 7 days later. SPOT capability generates 1000 vCPUs. There is a 5% chance of risk during the processing. The total processing time is 1 minute, and the execution success rate is expected to be 99%. Then the loss C loss =W4*C chip +W5*C block , assuming that the influence of the post-weight W1…W is not considered n =1, and the solution does not break any filtering conditions, then:

[0151] C loss =1000*(1 / 60)*W free =17W free

[0152] C free =1000*(7*24-1 / 60)*W free =167983W free

[0153]

[0154] C″=176500W free

[0155]

[0156] It should be noted that the confidence of any Provider can be calculated based on the resource information actually consumed when any Provider provides the required resources for any candidate resource deployment solution and the resource information to be deployed in any candidate resource deployment solution; the weight information required for calculating the resource loss cost information corresponding to any Provider is adjusted according to the confidence.

[0157] After the target resource solution is obtained, it can be delivered for execution. Different scenarios may require different resource delivery methods and delivery processes. In the corresponding strategy scenario, the specified execution topology is declared. The target resource solution can be submitted to the general topology process for execution. Figure 2cshown.

[0158] The implementation process includes several major stages:

[0159] Provider execution, calling the Provider processing execution interface;

[0160] Resource delivery: according to the delivery method configured in the scenario, the resources that have been executed are reserved for delivery;

[0161] Customized callback: according to the callback function configured in the scene, execute the customized processing completion callback to facilitate other systems to perceive the resource processing status;

[0162] Global status update, target resource solution lifecycle maintenance and combing.

[0163] Dynamic Priority:

[0164] The resource solution corresponding to the target candidate resource deployment solution is added to the ready queue to wait for scheduling. In this embodiment, each resource solution in the ready queue can be called a task. During the execution process, the waiting time of each task in the ready queue is monitored. When the waiting time exceeds the waiting time threshold, the priority of the task is increased; the waiting time threshold is dynamically determined according to the priority of the task. Among them, the higher the priority, the smaller the waiting time threshold.

[0165] For example, suppose the priority p ranges from 1 to 99. Priority 1 is the highest. When the priority is 1, the task will be directly triggered when the demand arrives; when the priority is -1, the task will not be actually scheduled. The remaining priority levels will determine how long it will take for the task to be executed after it is started. Assuming that the maximum execution time is x minutes (min), the relationship between the value of x and the priority p can be adopted but not limited to the following formula:

[0166] x=ceil(p / 10)*10k

[0167] That is, if the priority is 26 and k is a constant, the task will be scheduled at the latest in 30k minutes. Therefore, for tasks with a priority greater than 1, the resources of the pre-evaluation plan will be preempted, and the actual execution plan may be inconsistent.

[0168]

[0169] It should be noted that if Figure 2b , as shown, the demand time h need =h delay +h loss +h waste Finally, in order to ensure that low-priority tasks are not suspended all the time, there are h waste ≈0, then h need -hloss >h delay Therefore, if all tasks can be scheduled, although the final priority does not meet the requirements, in order for the task to be executed at least once, the final scheduling time can be adopted but not limited to the following formula: Min(h delay ,h need -avg(h loss )). It should also be noted that tasks that exceed the maximum number of retries will be set to -1 and cannot be automatically scheduled but can only be manually scheduled.

[0170] like Figure 2d As shown, in the actual execution process, it can be stipulated that the priority of the tasks in the ready queue increases at a rate of a per hour as their waiting time increases. For example, if the initial priority of a task is 80, but it is still not scheduled after 1 hour, its priority can be increased to 20.

[0171] For the sorting of tasks with the same maximum execution time of x min, the requirements will be aggregated according to the key information in the task. For tasks with key information, the task with the smallest sum of scores of all Providers in the task will be prioritized to achieve the smallest possible cost loss. For tasks with the same priority and consistent parameters after aggregation, the First Come First Service (FCFS) algorithm will be used for processing.

[0172] Execute the downgrade mechanism:

[0173] If the execution node corresponding to the scheduled Provider fails to execute, the additional resource amount n is calculated based on the execution success rate p1 of the failed Provider and the execution success rate p2 of the next Provider. buffer ; and when scheduling the next Provider, controlling the execution node corresponding to the next Provider to reserve resources according to at least part of the resource information and additional resources provided by the next Provider.

[0174] Among them, for the execution success rate of p1, the supply demand n can take certain additional supplementary measures to prevent the problem of non-delivery caused by solution failure. For the remaining providers with excess resources, additional resources are provided to mitigate the impact of failure. The additional provider success rate is p2, and the additional resource quantity is n buffer , the following formula can be used but is not limited to:

[0175] n buffer =n(1-p1) / p2

[0176] To summarize the above process, the entire execution process is as follows: Figure 2e As shown, the following operations are included:

[0177] 1. First, the system has a corresponding demand management module, which needs to receive and manage cloud product demand information, including standardized cloud product demand information and some project-based non-standardized cloud product demand information.

[0178] 2. Convert non-standardized cloud product demand information to obtain standardized cloud product demand information.

[0179] 3. The standardized cloud product demand information is evaluated through the resource orchestration and resource loss cost information mentioned above to produce the corresponding resource solutions, and finally select the executable target resource solutions.

[0180] 4. If the target resource solution does not meet the cloud product requirement information, repeat step 3 until the cloud product requirement information is met.

[0181] 5. After obtaining the target resource solutions that meet the cloud product demand information, sort and execute the target resource solutions according to their priorities.

[0182] 6. Before execution, the target resource solution can be evaluated again, and whether to finally execute the target resource solution can be determined based on the new evaluation results.

[0183] 7. After determining the target resource solution, enter the actual execution phase and ultimately produce the required cloud products.

[0184] The solution provided by the embodiment of the present application can not only quickly and automatically produce resource solutions, but also improve the automation ability of cloud service providers to produce cloud products, and solve and optimize the use of the underlying server resources of cloud service providers, so that the underlying server resources can be more efficient and generate more profits.

[0185] In addition to the above system embodiments, the present application also provides a cloud product data processing method, which can be applied to the above cloud product data processing system, but is not limited thereto. Figure 3a As shown, the method of this embodiment includes:

[0186] 301a. Obtain cloud product demand information, and parse the cloud product demand information into at least one candidate resource deployment solution;

[0187] 302a. Perform resource orchestration and resource loss cost information estimation processing on the at least one candidate resource deployment solution according to at least one resource supply capability provided by the cloud service provider, so as to obtain a target resource solution whose resource loss cost information meets the requirements and its corresponding target resource deployment solution;

[0188] 303a. According to the target resource solution, execute resource processing tasks in the available zone corresponding to the target resource deployment solution to obtain the cloud product corresponding to the cloud product demand information.

[0189] In the following content of this embodiment, at least one resource supply capability is recorded as N resource supply capabilities, and at least one candidate resource deployment scheme is recorded as K candidate resource deployment schemes, where N and K are both positive integers. In this embodiment, the target resource solution includes at least one resource supply capability and at least part of the resource information it is responsible for providing.

[0190] In an optional embodiment, the cloud product demand information includes: standardized cloud product demand information and / or non-standardized cloud product demand information. Accordingly, the above-mentioned parsing of the cloud product demand information into K candidate resource deployment solutions includes: converting the non-standardized cloud product demand information into standardized cloud product demand information, wherein the standardized cloud product demand information refers to cloud product demand information with clear resource requirements; generating K candidate resource deployment solutions according to the clear resource requirements corresponding to the standardized cloud product demand information.

[0191] In an optional embodiment, the above-mentioned generation of K candidate resource deployment schemes based on the clear resource requirements corresponding to the standardized cloud product demand information includes: determining X1 types of resource information and Y1 types of availability zone information based on the clear resource requirements corresponding to the standardized cloud product demand information; generating K candidate resource deployment schemes based on the X1 types of resource information and the Y1 types of availability zone information, where X1 and Y1 are positive integers.

[0192] In an optional embodiment, the above-mentioned generating K kinds of candidate resource deployment schemes according to the X1 kinds of resource information and the Y1 kinds of available zone information includes: filtering the X1 kinds of resource information and the Y1 kinds of available zone information to obtain X2 kinds of resource information and Y2 kinds of available zone information, X2 and Y2 are positive integers, and X2≤X1, Y2≤Y1; combining the X2 kinds of resource information and the Y2 kinds of available zone information to obtain P kinds of candidate resource deployment schemes, P is a positive integer, and P≥K; based on M kinds of resource supply capabilities and / or resource capability information of available zones, filtering the P kinds of candidate resource deployment schemes to obtain K kinds of candidate resource deployment schemes, M is a positive integer, and M≤N.

[0193] In an optional embodiment, each candidate resource deployment scheme includes at least resource information to be deployed and available zone information to be deployed. The above-mentioned resource orchestration and resource loss cost information estimation processing of the K candidate resource deployment schemes according to the N resource supply capabilities provided by the cloud service provider to obtain a target resource solution whose resource loss cost information meets the requirements and its corresponding target resource deployment scheme, including: for any candidate resource deployment scheme, according to the resource information to be deployed and the available zone information to be deployed in any candidate resource deployment scheme, estimating the unit resource loss cost information when the N resource supply capabilities respectively provide the required resources for any candidate resource deployment scheme; generating a resource solution and its resource loss cost information corresponding to any candidate resource deployment scheme according to the unit resource loss cost information when the N resource supply capabilities respectively provide the required resources for any candidate resource deployment scheme and the resource information to be deployed in any candidate resource deployment scheme; selecting a target resource deployment scheme and its corresponding target resource solution whose resource loss cost information meets the requirements according to the resource loss cost information of the resource solutions corresponding to each of the K candidate resource deployment schemes.

[0194] In an optional embodiment, the above-mentioned estimation of the unit resource loss cost information when N types of resource supply capabilities respectively provide the required resources for any of the candidate resource deployment schemes based on the resource information to be deployed and the available zone information to be deployed in any of the candidate resource deployment schemes includes: for any type of resource supply capability, obtaining the target resource information and target time information consumed when any of the resource supply capabilities provides the required resources for any of the candidate resource deployment schemes; generating the target resource loss cost information when any of the resource supply capabilities provides the required resources for any of the candidate resource deployment schemes based on the target resource information and the target time information; calculating the unit resource loss cost information when any of the resource supply capabilities provides the required resources for any of the candidate resource deployment schemes based on the target resource loss cost information and the resource information to be deployed in any of the candidate resource deployment schemes.

[0195] In an optional embodiment, the above-mentioned acquisition of the target resource information and target time information consumed when any one of the resource supply capabilities provides the required resources for any one of the candidate resource deployment schemes includes: determining the resource waiting time of any one of the resource supply capabilities based on the scheduling processing start time of any one of the candidate resource deployment schemes, the resource processing duration of any one of the resource supply capabilities, and the resource usage start time in any one of the candidate resource deployment schemes, and the target time information includes the resource processing duration and the resource waiting time; according to the resource information to be deployed in any one of the candidate resource deployment schemes, acquiring the effective resource information occupied and the resource fragmentation information generated when any one of the resource supply capabilities provides the required resources for any one of the candidate resource deployment schemes, and the target resource information includes the effective resource information and the resource fragmentation information.

[0196] In an optional embodiment, the above method further includes: obtaining the corresponding risk probability when any one of the resource supply capabilities provides the required resources for any one of the candidate resource deployment solutions, the additional resource proportion to ensure the satisfaction rate, and / or at least one of the filtering conditions. Accordingly, based on the target resource information and target time information, the target resource loss cost information is generated when any one of the resource supply capabilities provides the required resources for any one of the candidate resource deployment solutions, including: generating the target resource loss cost information based on the risk probability, the additional resource proportion to ensure the satisfaction rate, and / or at least one of the filtering conditions, as well as the effective resource information, the resource fragmentation information, the resource processing time, and the resource waiting time.

[0197] In an optional embodiment, the risk probability corresponding to the above-mentioned acquisition of any one of the resource supply capabilities to provide the required resources for any of the candidate resource deployment solutions includes:

[0198] generating the risk probability according to information on the impact on other resource objects and / or other users when any one of the resource supply capabilities provides the required resources for any one of the candidate resource deployment solutions;

[0199] The first resource scheduling unit obtains the additional resource proportion corresponding to ensuring the satisfaction rate when any one of the resource supply capabilities provides the required resources for any one of the candidate resource deployment solutions, including:

[0200] According to the historical deployment data of the available zones corresponding to the available zone information to be deployed in any of the candidate resource deployment schemes, generate the execution success rate of any of the resource supply capabilities when providing the required resources for any of the candidate resource deployment schemes; based on the user's expected satisfaction rate and the execution success rate, calculate the proportion of additional resources required to ensure the satisfaction rate.

[0201] In an optional embodiment, the target resource loss cost information is generated according to the risk probability, the proportion of additional resources to ensure the satisfaction rate and the filtering conditions, as well as the effective resource information, the resource fragmentation information, the resource processing time and the resource waiting time, including: generating resource processing cost information according to the effective resource information, the resource fragmentation information and the resource processing time; correcting the resource processing cost information according to the risk probability and the proportion of additional resources to ensure the satisfaction rate, respectively, to obtain the first intermediate state cost information and the second intermediate state cost information; generating resource waiting cost information according to the resource information to be deployed in any candidate resource deployment scheme and the resource waiting time; generating the target resource loss cost information according to the first intermediate state cost information, the second intermediate state cost information, the resource waiting cost information and the filtering conditions.

[0202] In an optional embodiment, the target resource loss cost information is generated according to the first intermediate state cost information, the second intermediate state cost information, the resource waiting cost information and the filtering condition, including: taking a weighted sum of the first intermediate state cost information, the second intermediate state cost information and the resource waiting cost information to obtain initial resource loss cost information; when any one of the resource supply capabilities meets the filtering condition, using the initial resource loss cost information as the target resource loss cost information; when any one of the resource supply capabilities does not meet the filtering condition, using the preset resource loss cost information as the target resource loss cost information.

[0203] In an optional embodiment, the above method also includes: calculating the confidence of any one of the resource supply capabilities based on the resource information actually consumed when any one of the resource supply capabilities provides the required resources for any one of the candidate resource deployment schemes and the resource information to be deployed in any one of the candidate resource deployment schemes; and adjusting the weight information required for calculating the initial resource loss cost information corresponding to any one of the resource supply capabilities based on the confidence.

[0204] In an optional embodiment, the above-mentioned unit resource loss cost information when the N types of resource supply capabilities provide the required resources for any of the candidate resource deployment schemes and the resource information to be deployed in any of the candidate resource deployment schemes are used to generate a resource solution corresponding to any of the candidate resource deployment schemes and its resource loss cost information, including: determining, based on the resource information to be deployed in any of the candidate resource deployment schemes and the upper limit resource information that can be provided by each of the N types of resource supply capabilities, at least one resource supply capability required to meet the resource information to be deployed in any of the candidate resource deployment schemes and at least part of the resource information it is responsible for providing, as the resource solution corresponding to any of the candidate resource deployment schemes; generating resource loss cost information of the resource solution corresponding to any of the candidate resource deployment schemes based on the unit resource loss cost information corresponding to the at least one resource supply capability and at least part of the resource information it is responsible for providing.

[0205] In an optional embodiment, the above-mentioned resource loss cost information of the resource solutions corresponding to each of the K candidate resource deployment solutions is used to select a target resource deployment solution and its corresponding target resource solution whose resource loss cost information meets the requirements, including: adding the target resource solution to a ready queue to wait for scheduling, and scheduling each resource solution in the ready queue according to the priority of the target resource solution, wherein the priority of the target resource solution is determined according to the resource scenario to which the cloud product demand information belongs; when scheduling to the target resource solution, according to the target resource solution, executing a resource processing task in the available zone corresponding to the available zone information to be deployed in the target resource deployment solution, so as to obtain the cloud product corresponding to the cloud product demand information.

[0206] In an optional embodiment, the above method also includes: monitoring the waiting time of each resource solution in the ready queue, and when monitoring a first resource solution whose waiting time exceeds a corresponding waiting time threshold, increasing the priority of the first resource solution; the waiting time threshold is determined based on the priority of the first resource solution before adjustment.

[0207] In an optional embodiment, the above-mentioned executing resource processing tasks in the available zone corresponding to the available zone information to be deployed in the target resource deployment solution according to the target resource solution to obtain the cloud product corresponding to the cloud product demand information includes: scheduling the at least one resource supply capability in sequence according to the unit resource loss cost information of the at least one resource supply capability in the target resource solution; for the scheduled resource supply capability, controlling the scheduled resource supply capability to reserve resources according to at least part of the resource information that the scheduled resource supply capability is responsible for providing; and when the resource reservation results corresponding to the at least one resource supply capability are all successful, assembling the resources reserved by the at least one resource supply capability into the cloud product corresponding to the cloud product demand information.

[0208] In an optional embodiment, the above method also includes: in the event that the scheduled resource supply capability fails to execute resource reservation, calculating the additional resource amount based on the execution success rate of the resource supply capability that failed to execute and the execution success rate of the next resource supply capability; when scheduling the next resource supply capability, controlling the next resource supply capability to reserve resources according to at least part of the resource information that the next resource supply capability is responsible for providing and the additional resource amount.

[0209] Figure 3b A flowchart of another cloud product data processing method provided in an embodiment of the present application. Figure 3b As shown, the method includes:

[0210] 301b. Obtain candidate resource deployment solutions and at least one resource supply capability, wherein the candidate resource deployment solutions at least include resource information to be deployed and available zone information to be deployed;

[0211] 302b. According to the at least one resource supply capability, estimate the unit resource loss cost information when the at least one resource supply capability provides the required resources for any candidate resource deployment solution;

[0212] 303b. Generate a resource solution and its resource loss cost information corresponding to the resource deployment solution according to the unit resource loss cost information corresponding to at least one resource supply capability and the resource information to be deployed.

[0213] In the following content of this embodiment, at least one resource supply capability may also be recorded as N resource supply capabilities. In this embodiment, the candidate resource deployment scheme may be implemented as any candidate resource deployment scheme in the above embodiment. For the relevant explanation of the resource deployment scheme and the resource solution and the detailed implementation of each step in this embodiment, please refer to the above embodiment, which will not be repeated here.

[0214] The detailed implementation and beneficial effects of each step in the method of this embodiment have been described in detail in the aforementioned embodiments and will not be elaborated here.

[0215] It should be noted that in some of the processes described in the above embodiments and the accompanying drawings, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or executed in parallel, and the sequence numbers of the operations, such as 301a, 302a, etc., are only used to distinguish between different operations, and the sequence numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent the order of precedence, and do not limit the "first" and "second" to be different types.

[0216] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Figure 4 As shown, the electronic device includes: a memory 41 and a processor 42.

[0217] The memory 41 is used to store computer programs and can be configured to store various other data to support operations on the electronic device. Examples of such data include instructions for any application or method operating on the electronic device, contact data, phone book data, messages, pictures, videos, etc.

[0218] The processor 42 is coupled to the memory 41 and is used to execute the computer program in the memory 41 to implement the steps in the cloud product data processing method provided in the embodiment of the present application. The detailed description of each step in the cloud product data processing method can be found in the above embodiment and will not be repeated here.

[0219] Further, if Figure 4 As shown, the computing platform also includes: a communication component 43, a display 44, a power component 45, an audio component 46 and other components. Figure 4 Only some components are shown schematically, which does not mean that the electronic device only includes Figure 4 In addition, Figure 4 The components in the dashed box are optional components, not mandatory components, and the specific components depend on the product form of the electronic device. The electronic device of this embodiment can be implemented as a terminal device such as a desktop computer, a laptop computer, a smart phone, or an IOT device, or a server device such as a conventional server, a cloud server, or a server array. If the electronic device of this embodiment is implemented as a terminal device such as a desktop computer, a laptop computer, a smart phone, etc., it can include Figure 4If the electronic device of this embodiment is implemented as a server device such as a conventional server, a cloud server or a server array, it may not include Figure 4 Components within the dashed box.

[0220] Accordingly, an embodiment of the present application further provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the processor is enabled to implement each step in the above method embodiment.

[0221] The above-mentioned memory can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read only memory (EEPROM), erasable programmable read only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0222] The above-mentioned communication component is configured to facilitate wired or wireless communication between the device where the communication component is located and other devices. The device where the communication component is located can access a wireless network based on a communication standard, such as WiFi, 2G, 3G, 4G / LTE, 5G and other mobile communication networks, or a combination thereof. In an exemplary embodiment, the communication component receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be based on Radio Frequency Identification (RFID) technology, Infrared Data Association (IrDA) technology, Ultra Wide Band (UWB) technology, Bluetooth (BT) technology and other technologies.

[0223] The above-mentioned display includes a screen, and the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from a user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundary of a touch or slide action, but also detect the duration and pressure associated with the touch or slide operation.

[0224] The power supply assembly provides power to various components of the device where the power supply assembly is located. The power supply assembly may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to the device where the power supply assembly is located.

[0225] The above-mentioned audio component can be configured to output and / or input audio signals. For example, the audio component includes a microphone (Microphone, MIC), and when the device where the audio component is located is in an operating mode, such as a call mode, a recording mode, and a speech recognition mode, the microphone is configured to receive an external audio signal. The received audio signal can be further stored in a memory or sent via a communication component. In some embodiments, the audio component also includes a speaker for outputting an audio signal.

[0226] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-readable storage media (including but not limited to disk storage, compact disc read-only memory (Compact Disc Read-Only Memory, CD-ROM), optical storage, etc.) containing computer-usable program code.

[0227] The present application is described with reference to the flowchart and / or block diagram of the method, device (system) and computer program product according to the embodiment of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, and the combination of the process and / or box in the flowchart and / or block diagram can be realized by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device for realizing the function specified in one process or multiple processes in the flowchart and / or one box or multiple boxes in the block diagram.

[0228] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device that implements the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0229] These computer program instructions may also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0230] In a typical configuration, a computing device includes one or more processors (Central Processing Unit, CPU), input / output interface, network interface and memory.

[0231] Memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.

[0232] Computer-readable media include permanent and non-permanent, removable and non-removable media that can be used to store information by any method or technology. Information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase-change random access memory (Phase-change Random Access Memory, PRAM), static random access memory (SRAM), dynamic random access memory (Dynamic Random Access Memory, DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory, etc.

[0233] (EEPROM), flash memory or other memory technology, CD-ROM, Digital Video Disc (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory media such as modulated data signals and carrier waves.

[0234] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0235] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A cloud product data processing system, characterized in that: include: A demand management module, used to obtain cloud product demand information and parse the cloud product demand information into at least one candidate resource deployment solution; A resource orchestration module, configured to perform resource orchestration and resource loss cost information estimation processing on the at least one candidate resource deployment solution according to at least one resource supply capability provided by the cloud service provider, so as to obtain a target resource solution whose resource loss cost information meets the requirements and its corresponding target resource deployment solution; The execution delivery module is used to execute resource processing tasks in the available zone corresponding to the target resource deployment solution according to the target resource solution, so as to obtain the cloud product corresponding to the cloud product demand information.

2. The system according to claim 1, characterized in that The demand management module includes: A demand entry unit, used to receive standardized cloud product demand information and / or non-standardized cloud product demand information; a demand conversion unit, configured to convert the non-standardized cloud product demand information into standardized cloud product demand information, wherein the standardized cloud product demand information refers to cloud product demand information with clear resource requirements; The demand parsing unit is used to generate at least one candidate resource deployment solution according to the clear resource demand corresponding to the standardized cloud product demand information.

3. The system according to claim 1 or 2, characterized in that: The candidate resource deployment scheme includes at least the resource information to be deployed and the available zone information to be deployed, and the resource orchestration module is specifically used to: A first resource scheduling unit is used to estimate, for any candidate resource deployment scheme, unit resource loss cost information when at least one resource supply capability provides required resources for any candidate resource deployment scheme according to resource information to be deployed and available zone information to be deployed in any candidate resource deployment scheme; A second resource scheduling unit is used to generate a resource solution corresponding to any candidate resource deployment scheme and its resource loss cost information according to the unit resource loss cost information when at least one resource supply capability provides the required resources for any candidate resource deployment scheme and the resource information to be deployed in any candidate resource deployment scheme; The solution selection unit is used to select a target resource deployment solution and its corresponding target resource solution whose resource loss cost information meets the requirements according to the resource loss cost information of the resource solution corresponding to each of the at least one candidate resource deployment solutions.

4. The system according to claim 3, characterized in that The first resource scheduling unit is specifically configured to: For any resource supply capability, obtaining target resource information and target time information consumed when any resource supply capability provides required resources for any candidate resource deployment solution; Generate target resource loss cost information when any one of the resource supply capabilities provides required resources for any one of the candidate resource deployment solutions according to the target resource information and the target time information; The unit resource loss cost information when any one of the resource supply capabilities provides the required resources for any one of the candidate resource deployment schemes is calculated according to the target resource loss cost information and the resource information to be deployed in any one of the candidate resource deployment schemes.

5. The system according to claim 3, characterized in that The second resource arrangement unit is specifically configured to: According to the resource information to be deployed in any of the candidate resource deployment schemes and the upper limit resource information that can be provided by each of the at least one resource supply capability, determine at least one resource supply capability required to meet the resource information to be deployed in any of the candidate resource deployment schemes and at least part of the resource information it is responsible for providing, as the resource solution corresponding to any of the candidate resource deployment schemes; Resource loss cost information of a resource solution corresponding to any one of the candidate resource deployment solutions is generated based on the unit resource loss cost information corresponding to the at least one resource supply capability and at least part of the resource information that is responsible for being provided.

6. A cloud product data processing method, characterized in that: include: Parse cloud product demand information into at least one candidate resource deployment solution; Perform resource orchestration and resource loss cost information estimation processing on the at least one candidate resource deployment solution according to at least one resource supply capability provided by the cloud service provider, so as to obtain a target resource solution whose resource loss cost information meets the requirements and its corresponding target resource deployment solution; According to the target resource solution, a resource processing task is executed in an available zone corresponding to the target resource deployment solution to obtain a cloud product corresponding to the cloud product demand information.

7. The method according to claim 6, characterized in that The cloud product demand information includes: standardized cloud product demand information and / or non-standardized cloud product demand information; Parsing the cloud product demand information into at least one candidate resource deployment solution includes: Converting the non-standardized cloud product demand information into standardized cloud product demand information, wherein the standardized cloud product demand information refers to cloud product demand information with clear resource requirements; At least one candidate resource deployment solution is generated according to the clear resource requirements corresponding to the standardized cloud product requirement information.

8. The method according to claim 6, characterized in that The candidate resource deployment scheme includes at least the resource information to be deployed and the available zone information to be deployed. Resource orchestration and resource loss cost information estimation processing are performed on the at least one candidate resource deployment scheme according to at least one resource supply capability provided by the cloud service provider to obtain a target resource solution whose resource loss cost information meets the requirements and its corresponding target resource deployment scheme, including: For any candidate resource deployment scheme, based on the resource information to be deployed and the available zone information to be deployed in any candidate resource deployment scheme, estimate the unit resource loss cost information when at least one resource supply capability provides the required resources for any candidate resource deployment scheme; Generate a resource solution and its resource loss cost information corresponding to any candidate resource deployment scheme according to the unit resource loss cost information when at least one resource supply capability provides the required resources for any candidate resource deployment scheme and the resource information to be deployed in any candidate resource deployment scheme; According to the resource loss cost information of the resource solution corresponding to each of the at least one candidate resource deployment schemes, a target resource deployment scheme whose resource loss cost information meets the requirements and its corresponding target resource solution are selected.

9. The method according to claim 8, characterized in that According to the resource information to be deployed and the available zone information to be deployed in any of the candidate resource deployment schemes, estimating the unit resource loss cost information when at least one resource supply capability provides the required resources for each of the candidate resource deployment schemes, including: For any resource supply capability, obtaining target resource information and target time information consumed when any resource supply capability provides required resources for any candidate resource deployment solution; Generate target resource loss cost information when any one of the resource supply capabilities provides required resources for any one of the candidate resource deployment solutions according to the target resource information and the target time information; The unit resource loss cost information when any one of the resource supply capabilities provides the required resources for any one of the candidate resource deployment schemes is calculated according to the target resource loss cost information and the resource information to be deployed in any one of the candidate resource deployment schemes.

10. The method according to claim 9, characterized in that Obtaining target resource information and target time information consumed when any one of the resource supply capabilities provides the required resources for any one of the candidate resource deployment solutions, including: Determine the resource waiting time of any resource supply capability according to the scheduling processing start time of any candidate resource deployment scheme, the resource processing time of any resource supply capability, and the resource use start time in any candidate resource deployment scheme, wherein the target time information includes the resource processing time and the resource waiting time; According to the resource information to be deployed in any of the candidate resource deployment schemes, obtain the effective resource information occupied and the resource fragment information generated when any of the resource supply capabilities provides the required resources for any of the candidate resource deployment schemes, and the target resource information includes the effective resource information and the resource fragment information.

11. The method according to claim 10, characterized in that Also includes: Obtain at least one of the risk probability corresponding to when any one of the resource supply capabilities provides the required resources for any one of the candidate resource deployment solutions, the proportion of additional resources to ensure the satisfaction rate, and / or the filtering conditions; Accordingly, according to the target resource information and the target time information, target resource loss cost information is generated when any one of the resource supply capabilities provides the required resources for any one of the candidate resource deployment solutions, including: The target resource loss cost information is generated based on the risk probability, the proportion of additional resources to ensure the satisfaction rate and / or at least one of the filtering conditions, as well as the effective resource information, the resource fragmentation information, the resource processing time and the resource waiting time.

12. The method according to claim 11, characterized in that The target resource loss cost information is generated according to the risk probability, the proportion of additional resources to ensure the satisfaction rate and the filtering conditions, as well as the effective resource information, the resource fragmentation information, the resource processing time and the resource waiting time, including: Generate resource processing cost information according to the effective resource information, the resource fragment information and the resource processing time; According to the risk probability and the proportion of additional resources to ensure the satisfaction rate, the resource processing cost information is respectively modified to obtain the first intermediate state cost information and the second intermediate state cost information; Generate resource waiting cost information according to the resource information to be deployed in any candidate resource deployment solution and the resource waiting time; The target resource loss cost information is generated according to the first intermediate state cost information, the second intermediate state cost information, the resource waiting cost information and the filtering condition.

13. The method according to claim 12, characterized in that Generating the target resource loss cost information according to the first intermediate state cost information, the second intermediate state cost information, the resource waiting cost information and the filtering condition includes: Performing a weighted summation on the first intermediate state cost information, the second intermediate state cost information, and the resource waiting cost information to obtain initial resource loss cost information; In the case where any resource supply capability satisfies the filtering condition, taking the initial resource loss cost information as the target resource loss cost information; In the case that any resource supply capability does not satisfy the filtering condition, the preset resource loss cost information is used as the target resource loss cost information.

14. The method according to any one of claims 8 to 13, characterized in that: According to the unit resource loss cost information when at least one resource supply capability provides the required resources for any of the candidate resource deployment schemes and the resource information to be deployed in any of the candidate resource deployment schemes, a resource solution corresponding to any of the candidate resource deployment schemes and its resource loss cost information are generated, including: According to the resource information to be deployed in any of the candidate resource deployment schemes and the upper limit resource information that can be provided by each of the at least one resource supply capability, determine at least one resource supply capability required to meet the resource information to be deployed in any of the candidate resource deployment schemes and at least part of the resource information it is responsible for providing, as the resource solution corresponding to any of the candidate resource deployment schemes; Resource loss cost information of a resource solution corresponding to any one of the candidate resource deployment solutions is generated based on the unit resource loss cost information corresponding to the at least one resource supply capability and at least part of the resource information that is responsible for being provided.

15. The method according to any one of claims 6 to 13, characterized in that: According to the resource loss cost information of the resource solution corresponding to each of the at least one candidate resource deployment schemes, selecting a target resource deployment scheme whose resource loss cost information meets the requirements and its corresponding target resource solution, including: Adding the target resource solution to a ready queue to wait for scheduling, and scheduling each resource solution in the ready queue according to a priority of the target resource solution, wherein the priority of the target resource solution is determined according to a resource scenario to which the cloud product demand information belongs; When the target resource solution is scheduled, the resource processing task is executed in the available zone corresponding to the available zone information to be deployed in the target resource deployment plan according to the target resource solution to obtain the cloud product corresponding to the cloud product demand information.

16. The method according to claim 15, characterized in that Also includes at least one of the following operations: Monitoring the waiting time of each resource solution in the ready queue, and when monitoring a first resource solution whose waiting time exceeds a corresponding waiting time threshold, increasing the priority of the first resource solution; The waiting time threshold is determined according to the priority of the first resource solution before adjustment; In the case that the resource supply capability that is scheduled fails to execute resource reservation, the additional resource amount is calculated according to the execution success rate of the resource supply capability that failed to execute and the execution success rate of the next resource supply capability; When scheduling the next resource supply capability, the next resource supply capability is controlled to reserve resources according to at least part of the resource information and the additional resource amount that the next resource supply capability is responsible for providing.

17. A cloud product data processing method, characterized in that: include: Acquire a candidate resource deployment scheme and at least one resource supply capability, wherein the candidate resource deployment scheme includes at least resource information to be deployed and available zone information to be deployed; According to the at least one resource supply capability, estimate the unit resource loss cost information when the at least one resource supply capability provides the required resources for any candidate resource deployment solution; A resource solution and its resource loss cost information corresponding to the resource deployment solution are generated according to the unit resource loss cost information corresponding to at least one resource supply capability and the resource information to be deployed.

18. An electronic device, characterized in that: include: Memory and processor; The memory is used to store a computer program, and the processor is coupled to the memory and is used to execute the computer program to perform the steps in the method according to any one of claims 6 to 17.

19. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the processor is enabled to implement the steps in the method according to any one of claims 6 to 17.

Images