Method, device and system for scheduling computing resources

Abstract

The present disclosure provides a scheduling method, device and system for computing power resources, which relate to the field of computing power networks. The method includes: receiving a resource scheduling request from a user, the resource scheduling request includes reference information, the reference information includes user demand information, the demand information includes computing power demand and network demand, the network demand includes first network information of a required first path, the first path is a path between the user device of the user and a computing power node that provides computing power resources; and determining a target computing power resource that meets the computing power demand and the network demand based on the reference information.

Description

Technical Field

[0001] The present disclosure relates to the field of computing power networks, and in particular to a method, device and system for scheduling computing power resources. Background Art

[0002] The implementation of computing power network technology requires the scheduling of various computing power resources. The current computing power network implementation solutions mainly include distributed solutions and centralized solutions.

[0003] The distributed solution publishes computing resources as routing information to the network by extending routing protocols such as the Border Gateway Protocol (BGP) and the Interior Gateway Protocol (IGP); the centralized solution mainly schedules computing resources based on a centralized controller. Summary of the invention

[0004] The inventors have noticed that, regardless of whether it is a distributed solution or a centralized solution, there is a problem that users cannot obtain computing resources in a timely manner or cannot obtain computing resources at all.

[0005] The inventors have conducted further analysis on the above issues:

[0006] On the one hand, currently when users apply for computing resources, they only put forward their demand for computing resources but do not put forward their demand for the network. As a result, the computing resources can meet the user's computing resource needs, but due to problems with the network corresponding to the computing resources, the user cannot obtain the computing resources in a timely manner or cannot obtain them at all.

[0007] For example, in a distributed solution, due to the large amount of computing power resource data, when it is published to the network as routing information, it is easy to cause excessive network load, and users may not be able to obtain computing power resources in a timely manner or at all.

[0008] On the other hand, in a centralized solution, the processing capacity of the computing network is limited by the capabilities of the centralized controller. That is to say, once a centralized controller fails, the corresponding computing network may be paralyzed and data may be lost, making it impossible to obtain computing resource information.

[0009] In order to solve the above problems, the embodiments of the present disclosure propose the following solutions.

[0010] According to one aspect of an embodiment of the present disclosure, a method for scheduling computing power resources is provided, comprising: receiving a resource scheduling request from a user, the resource scheduling request comprising reference information, the reference information comprising demand information of the user, the demand information comprising computing power demand and network demand, the network demand comprising first network information of a required first path, the first path being a path between a user device of the user and a computing power node providing the computing power resources; and determining, based on the reference information, a target computing power resource that meets the computing power demand and the network demand.

[0011] In some embodiments, the target computing power resource is determined from multiple computing power resources according to a first correspondence table, the first correspondence table including the identifiers of the multiple computing power resources and second network information of a second path between the user device and a computing power node corresponding to each of the multiple computing power resources, the identifier of each of the multiple computing power resources including an identification number and identification information, the identification information indicating the computing power situation corresponding to each of the multiple computing power resources.

[0012] In some embodiments, a first subpath between the user device and the user-side access device; a second subpath between the user-side access device and the computing power resource side access device; and a third subpath between the computing power node corresponding to each computing power resource in the multiple computing power resources and the computing power resource side access device.

[0013] In some embodiments, the resource scheduling request carries second network information of the first subpath.

[0014] In some embodiments, the identification information includes a computing power price, and determining the target computing power resource from multiple computing power resources according to the first correspondence table includes: determining, according to the first correspondence table, multiple alternative computing power resources that meet the computing power requirements and the network requirements from the multiple computing power resources; when the total number of the resource scheduling requests received is less than a predetermined number, sorting the multiple alternative computing power resources according to the capacity and computing power price of the multiple alternative computing power resources to select the alternative computing power resource with the largest capacity and the lowest computing power price as the target computing power resource.

[0015] In some embodiments, the reference information further includes attribute information of the user, wherein the attribute information includes at least one of user category, user price sensitivity, user priority, and user location.

[0016] In some embodiments, the identification of the target computing power resource and the reference information are stored correspondingly in a second corresponding table; when the total number of the resource scheduling requests that have been received is not less than the predetermined number, the target computing power resources of historical users similar to the user are selected from the second corresponding table according to the attribute information as the target computing power resources of the user.

[0017] In some embodiments, multiple target computing power resources of historical users similar to the user are selected from the multiple alternative computing power resources; according to the capacity and computing power price of the multiple target computing power resources, the multiple target computing power resources are sorted to select the target computing power resource with the largest capacity and the lowest computing power price as the target computing power resource of the user.

[0018] In some embodiments, an identification application request for each of the multiple computing power resources is received, the identification application request includes first computing power information, the first computing power information of each computing power resource includes a computing power node corresponding to each computing power resource; based on the identification generation information, an identification is generated for each of the multiple computing power resources, the identification generation information includes the first computing power information.

[0019] In some embodiments, the first computing power information also includes the physical machine where each computing power resource among the multiple computing power resources is located and the container in the physical machine that provides the computing power resource, or the physical machine where the computing power resource is located and the virtual machine in the physical machine that provides the computing power resource.

[0020] In some embodiments, the identification generation information also includes time information, and the time information includes a current moment and a preset moment; generating an identification for each of the multiple computing power resources according to the identification generation information includes: generating the identification number based on the difference between the current moment and the preset moment.

[0021] In some embodiments, a binary first serial number is generated based on the time information; a binary second serial number is generated based on the first computing power information, and the identification number includes the first serial number and the second serial number.

[0022] In some embodiments, the identification application request also includes second computing power information; the generation of the identification information based on the identification includes: extracting relevant information matching a preset field from the second computing power information, wherein the second computing power information includes at least one of the following: the cloud where each of the multiple computing power resources is located, the computing power type of each of the multiple computing power resources, the computing power form of each of the multiple computing power resources, the computing power unit of each of the multiple computing power resources, the capacity of each of the multiple computing power resources and the time of receiving the identification application request, the price of each of the multiple computing power resources, the capacity of each of the multiple computing power resources indicates the number of computing power units included in each of the multiple computing power resources; and generating an identification in the form of a uniform resource locator as the identification of each of the multiple computing power resources based on the identification number and the relevant information.

[0023] In some embodiments, the identification application request also includes computing power resource network information, and the computing power resource network information includes the second network information of the third sub-path.

[0024] In some embodiments, the second network information of the third sub-path includes the price of the third sub-path, and the identification information includes the computing power price, the computing power price is determined according to the price of the third sub-path, and the price of the third sub-path is determined according to the bandwidth of the third sub-path and the connection method of the third sub-path.

[0025] In some embodiments, the first network information and the second network information include at least one of latency, bandwidth, and packet loss rate.

[0026] In some embodiments, the identification number of the target computing power resource is sent to the user device so that the user device obtains the Internet Protocol IP address of the target computing power resource from the computing power resolution platform based on the identification number.

[0027] In some embodiments, before the identification number of the target computing power resource is sent to the user device, it is confirmed whether the path between the computing power node of the target computing power resource and the user device is available.

[0028] In some embodiments, when the identifier of any one of the multiple computing power resources is lost, the identifier of the any one computing power resource is obtained from the computing power resolution platform.

[0029] According to another aspect of an embodiment of the present disclosure, a scheduling device for computing power resources is provided, including: a receiving module, configured to receive a resource scheduling request from a user, the resource scheduling request including reference information, the reference information including demand information of the user, the demand information including computing power demand and network demand, the computing power demand including the type and capacity of computing power resources required, the network demand including first network information of a required first path, the first path being a path between a user device of the user and a computing power node providing the computing power resources; a determining module, configured to determine a target computing power resource that meets the computing power demand and the network demand based on the reference information.

[0030] According to another aspect of an embodiment of the present disclosure, a scheduling device for computing resources is provided, comprising: a memory; and a processor coupled to the memory, configured to execute the method described in any one of the above embodiments based on instructions stored in the memory.

[0031] According to another aspect of the embodiments of the present disclosure, a scheduling system for computing power resources is provided, comprising: a scheduling device for computing power resources as described in any one of the above embodiments, configured to determine the target computing power resource from multiple computing power resources according to a first correspondence table, the first correspondence table comprising identifiers of the multiple computing power resources and second network information of a second path between the user device and a computing power node corresponding to each of the multiple computing power resources, the identifier of each of the multiple computing power resources comprising an identification number and identification information, the identification information indicating the computing power situation corresponding to each of the multiple computing power resources; sending the identification number of the target computing power resource to the user device; a computing power analysis platform, configured to provide the IP address of the target computing power resource to the user device based on the identification number.

[0032] In some embodiments, when the identifier of any one of the multiple computing power resources is lost, the computing power analysis platform provides the identifier of the any one computing power resource.

[0033] According to another aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided, comprising computer program instructions, wherein when the computer program instructions are executed by a processor, the method described in any one of the above embodiments is implemented.

[0034] According to another aspect of the embodiments of the present disclosure, a computer program product is provided, including a computer program, wherein when the computer program is executed by a processor, the method described in any one of the above embodiments is implemented.

[0035] In the disclosed embodiment, the resource scheduling request issued includes the computing power demand and network demand for computing power resources, and computing power resources that meet both the computing power demand and the network demand can be scheduled, thereby reducing the occurrence of situations where users are unable to obtain computing power resources in a timely manner or are unable to obtain computing power resources due to problems with the network corresponding to the computing power resources.

[0036] The technical solution of the present disclosure is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0038] Figure 1 is a flowchart of a method for scheduling computing resources according to some embodiments of the present disclosure;

[0039] Figure 2 is a flowchart of a method for scheduling computing resources according to other embodiments of the present disclosure;

[0040] Figure 3 is a structural schematic diagram of a scheduling device for computing resources according to some embodiments of the present disclosure;

[0041] Figure 4 is a schematic diagram of the structure of a scheduling device for computing resources according to other embodiments of the present disclosure;

[0042] Figure 5 It is a structural diagram of a scheduling system for computing resources according to some embodiments of the present disclosure. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present disclosure to clearly and completely describe the technical solutions in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present disclosure.

[0044] Unless specifically stated otherwise, the relative arrangement of components and steps, the numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present disclosure.

[0045] At the same time, it should be understood that for the convenience of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship.

[0046] Technologies, methods, and apparatus known to ordinary technicians in the relevant field may not be discussed in detail, but where appropriate, such technologies, methods, and apparatus should be considered part of the authorization specification.

[0047] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0048] It should be noted that like reference numerals and letters refer to similar items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0049] In addition, in the description of the present disclosure, the terms "first", "second", "third", etc. are used for descriptive purposes only and should not be understood as indicating or implying relative importance and order. Similarly, although operations are depicted in a particular order in the accompanying drawings, this should not be understood as requiring that such operations be performed in the particular order shown or in a sequential order, or requiring that all illustrated operations be performed to achieve the desired results. In some cases, multitasking and parallel processing can be advantageous.

[0050] Figure 1 It is a flowchart of a method for scheduling computing resources according to some embodiments of the present disclosure.

[0051] In step 102, a resource scheduling request of a user is received. Here, the resource scheduling request includes reference information.

[0052] The reference information includes user demand information, which includes computing power requirements and network requirements. The network requirements include first network information of a required first path, where the first path is a path between the user's user device and a computing power node that provides computing power resources.

[0053] In some embodiments, the first network information includes at least one of latency, bandwidth, and packet loss rate.

[0054] In some embodiments, the computing power requirement includes the type and capacity of computing power resources required.

[0055] In some embodiments, the computing power types of computing resources include central processing unit computing power, graphics processing unit computing power, and field programmable gate array computing power.

[0056] In some embodiments, the capacity of the computing power resource may indicate the number of computing power units included in the computing power resource. For example, the computing power units of the computing power resource include single-precision floating point numbers (FP32).

[0057] In some embodiments, the first path may include a path between a user device and a user-side access device, a path between a user-side access device and a computing power resource-side access device, and a path between a computing power node providing computing power resources and the computing power resource-side access device.

[0058] In some embodiments, the reference information further includes attribute information of the user, and the attribute information includes at least one of user category, user price sensitivity, user priority, and user location. Here, the user category may be, for example, an individual user, an enterprise user, and a government user; and the user price sensitivity may indicate whether the user is sensitive to the price of computing resources.

[0059] In some embodiments, the user priority may indicate whether the user is a regular user, a member user, or a super member user, and the user may improve the user priority by paying.

[0060] As some implementation methods, when a low user priority user and a high user priority user need the same computing power resource at the same time, the high user priority user can use the computing power resource preferentially.

[0061] As some further implementation methods, high user priority users may apply for computing resources that are only available to high user priority users.

[0062] In step 104, target computing resources that meet computing power requirements and network requirements are determined based on the reference information.

[0063] When the reference information only includes the user's demand information, the target computing power resources are determined based on the user's demand information; when the reference information includes the user's demand information and attribute information, the target computing power resources are determined based on the user's demand information and attribute information.

[0064] In some embodiments, information related to the target computing resources may be sent to the user, and the user may use the information to obtain the target computing resources.

[0065] In this way, the resource scheduling request issued includes the computing power demand and network demand for computing resources, and then computing power resources that meet both computing power demand and network demand can be scheduled, thereby reducing the occurrence of situations where users are unable to obtain computing power resources in a timely manner or are unable to obtain computing power resources due to problems with the network corresponding to the computing power resources.

[0066] In some embodiments, the target computing resource can be determined from multiple computing resources according to the first corresponding table. Here, the first corresponding table includes the identifiers of multiple computing resources and the second network information of the second path between the user device and the computing node corresponding to each of the multiple computing resources, and the identifiers of the multiple computing resources include an identification number and identification information, and the identification information indicates the computing power corresponding to each of the multiple computing resources.

[0067] In some embodiments, the second network information includes at least one of latency, bandwidth, and packet loss rate.

[0068] In this way, the first corresponding table stores the identifiers of different computing resources and the network information corresponding to the paths between the user device and different computing nodes, thereby improving the accuracy of the determined target computing resources and meeting the user's network requirements.

[0069] In some embodiments, the second path includes a first sub-path between the user device and the user-side access device, a second sub-path between the user-side access device and the computing resource-side access device, and a third sub-path from a computing node corresponding to each computing resource in the plurality of computing resources to the computing resource-side access device. Here, the first path may be one of the one or more second paths.

[0070] In this way, by selecting the target computing power resources based on the network information of the three sub-paths in the second path, it can be ensured that the determined target computing power resources meet the network requirements proposed by the user.

[0071] As some implementation methods, the resource scheduling request carries the second network information of the first sub-path. In other words, the user reports the second network information of the first sub-path when applying for computing resources.

[0072] As some other implementation methods, when a user applies for computing resources, the network device (eg, router) on the first sub-path also reports the second network information of the first sub-path.

[0073] In this way, the target computing power resources can be determined more accurately according to the second network information reported by the user. In addition, there is no need to obtain the network information of the first sub-path, which speeds up the determination of the target computing power resources, so that computing power resources can be scheduled for users more quickly.

[0074] The following describes an implementation method for determining a target computing resource from multiple computing resources in conjunction with different embodiments.

[0075] In some embodiments, the identification information includes a computing power price, and the target computing power resources can be determined according to the following method.

[0076] First, according to the first correspondence table, multiple candidate computing resources that meet the computing power requirements and network requirements are determined from multiple computing power resources.

[0077] Then, when the total number of resource scheduling requests received is less than the predetermined number, multiple alternative computing power resources are sorted according to their capacities and computing power prices to select the alternative computing power resource with the largest capacity and the lowest computing power price as the target computing power resource.

[0078] For example, if the predetermined number is 50, then when the total number of resource scheduling requests received (i.e., resource scheduling requests received in history) is less than 50, the computing resources among the multiple candidate computing resources are prioritized. The larger the computing resource capacity (i.e., the number of computing units included in the computing resource), the lower the computing price, and the higher the ranking. In this way, the computing resource with the highest ranking is the target computing resource.

[0079] In this way, by considering the capacity and price of computing resources, users can be provided with more reliable and cost-effective target computing resources.

[0080] In some other embodiments, the reference information also includes attribute information of the user. In this case, the target computing power resource can be determined according to the following method.

[0081] First, according to the first correspondence table, multiple candidate computing resources that meet the computing power requirements and network requirements are determined from multiple computing power resources.

[0082] Then, it is determined whether the total number of resource scheduling requests that have been received is less than a predetermined number.

[0083] In the case where the total number of resource scheduling requests that have been received is less than the predetermined number, in a similar manner to the above, the multiple candidate computing resources are sorted according to their capacities and computing prices to select the candidate computing resource with the largest capacity and the lowest computing price as the target computing resource. After the target computing resource is determined, the identifier of the target computing resource and the reference information are stored in the second corresponding table.

[0084] When the total number of resource scheduling requests that have been received is not less than a predetermined number, the target computing power resources of historical users similar to the user are selected from the second corresponding table as the target computing power resources according to the attribute information.

[0085] In some implementations, if a historical user has the same price sensitivity, user priority, and user category as the current user, the historical user may be considered a similar user.

[0086] In other implementations, if a historical user has two or more of the same price sensitivity, user priority, and user category as the current user, the historical user may be considered a similar user, so that the determined similar users are more accurate.

[0087] For example, if the reserved number is 50, then when the total number of resource scheduling requests that have been received (i.e., resource scheduling requests received historically) is greater than or equal to 50, based on the current user's price sensitivity, user priority and user category, the target computing power resource allocation information of historical users in the second corresponding table is queried, similar users are matched, and the historical computing power resource allocation information of similar users is obtained.

[0088] A computing power resource having a computing power identification number consistent with that in the historical computing power resource allocation of a similar user is selected from a plurality of candidate computing power resources.

[0089] Prioritize the computing resources with the same computing power identification number as those in the historical computing power resource allocation of similar users. The larger the computing power resource capacity and the lower the computing power price, the higher the ranking. In this way, the computing power resource with the highest ranking is the target computing power resource.

[0090] In this way, when there are many resource scheduling requests received in the past, there is no need to sort all the computing resources according to the capacity and price of the computing resources. Instead, it is only necessary to sort the computing resources with the same computing identification number as the historical computing resource allocation of similar users, thereby speeding up the determination of the target computing resources. In addition, by considering the capacity and price of the target computing resources of similar users corresponding to the user, more reliable and cost-effective target computing resources can be provided to the user.

[0091] Combine the following Figure 2 This section describes how to allocate computing resources.

[0092] Figure 2 It is a flowchart of a method for scheduling computing resources according to other embodiments of the present disclosure.

[0093] In step 202, an identification application request of each computing resource in a plurality of computing resources is received. Here, the identification application request includes first computing information, and the first computing information of each computing resource includes a computing node corresponding to each computing resource.

[0094] For example, a node may be a physical carrier such as a data center, an availability zone, a resource pool, etc. In some embodiments, the identification application request may be sent by the node where the computing resource is located.

[0095] In step 204, an identifier is generated for each computing power resource in the plurality of computing power resources according to the identifier generation information. Here, the identifier generation information includes the first computing power information.

[0096] For example, the identifier of the computing power resource may include an identification number, or the identification number may be converted into an identifier based on a predetermined rule.

[0097] As some implementations, the identification number may include specific characters in the name of the node where the computing power resource is located, such as a number, etc. For example, when the name of the node where the computing power resource is located is "shenzhen1234", the first computing power information includes "shenzhen1234", then the identification number may be 1234.

[0098] As other implementations, the name of the node where the computing power resource is located can be converted into an identification number based on a predetermined rule.

[0099] It can be understood that no matter which method is used, as long as the identification number and the node where the computing power resource is located are uniquely matched, that is, the identification number is different when the node is different.

[0100] The identification generation information may also include other information, which will be described below in conjunction with other embodiments.

[0101] In this way, the computing resource identification application request carries the node where the computing resource is located, and the computing resource identification is generated by using the node where the computing resource is located, thereby realizing the identification of the computing resource. Since the node where each computing resource is located is unique, the generated computing resource identification can uniquely correspond to a computing resource. In this way, according to the computing resource identification, at least the computing resources located at different nodes can be distinguished.

[0102] In some embodiments, the first computing power information includes the node where the computing power resource is located and the physical machine where the computing power resource is located. The physical machine may be, for example, a server. In this case, an identifier of the computing power resource may be generated based on the node where the computing power resource is located and the physical machine where the computing power resource is located. The identifier of the computing power resource generated in this manner can not only distinguish computing power resources located at different nodes, but also distinguish computing power resources of different physical machines located at the same node, thereby achieving a more fine-grained distinction of computing power resources.

[0103] In some embodiments, the first computing power information includes the node where the computing power resource is located, the physical machine where the computing power resource is located, and the container or virtual machine in the physical machine that provides the computing power resource. Here, the container or virtual machine can also be referred to as a computing power resource virtual unit.

[0104] In this way, the identifier of computing power resources generated in this way can not only distinguish the computing power resources located in different nodes and the computing power resources of different physical machines located on the same node, but also distinguish the computing power resources of containers or virtual machines that provide computing power resources in the same physical machine on the same node, thereby achieving a more fine-grained distinction of computing power resources.

[0105] In some embodiments, the identification generation information includes the first computing power information and time information. Here, the time information includes the current time. At this time, an identification number can be generated based on the current time and the first computing power information. As some implementations, the identification number may include a number corresponding to the current time. For example, if the name of the node where the computing power resource is located is "shenzhen1234" and the current time is 14:51, the identification number may be 1451 1234.

[0106] In this way, by utilizing the difference between different current moments to identify the computing power resources, the distinctiveness of the computing power resource identification is increased, thereby improving the uniqueness of the computing power resource identification.

[0107] In some embodiments, the time information also includes a preset time. In this case, an identification number is generated based on the difference between the current time and the preset time. As some implementations, the identification number may include a number corresponding to the difference between the current time and the preset time. For example, the name of the node where the computing power resource is located is "shenzhen1234", the current time is 14:51, and the preset time is 14:50, then the identification number can be 0001 1234. The preset time can also be greater than the current time, for example, if the current time is 14:51 and the preset time is 14:52, then the identification number can also be 0001 1234.

[0108] It should be understood that, in the case where the identification number includes a number corresponding to the difference between the current moment and the preset moment, the smaller the absolute value of the difference between the current moment and the preset moment is, the better, so that the generated corresponding identification number can start from the smallest possible value, thereby reducing the waste of identification numbers. For example, when the current moment is 14:51 and the preset moment is 00:00, the corresponding identification number starts from 1451, and the identification numbers from 0000 to 1450 are wasted; for another example, when the current moment is 14:51 and the preset moment is 14:51, the corresponding identification number starts from 0000, so no identification number is wasted.

[0109] In this way, by using the difference between the current time and the preset time to identify the computing power resources, not only the distinctiveness of the computing power resource identification is increased, but also the waste of identification numbers is reduced.

[0110] Some embodiments of generating identification numbers based on identification generation information are introduced below.

[0111] In some embodiments, a binary first serial number is generated based on time information; a binary second serial number is generated based on first computing power information, and the identification number includes the first serial number and the second serial number.

[0112] For example, the first computing power information includes the current time as 00:00, then the binary first serial number corresponding to the current time is 0000; the binary second serial number of the node corresponding to the first received identification application request is 0001; in this case, the identification number can be expressed as 0000 0001, that is, the identification number indicates that a computing power resource comes from the node corresponding to the first received identification application request, and the identification number is generated at 00:00.

[0113] For another example, the first computing power information includes the current time as 00:00, then the binary first serial number corresponding to the current time is 0000; the binary second serial number of the second physical machine of the node that receives the identification application request is 0010; in this case, the identification number can be expressed as 0000 0010, that is, the identification number indicates that a computing power resource comes from the second physical machine of the node that receives the identification application request, and the identification number is generated at 00:00.

[0114] In this way, when the number of nodes and physical machines is large, more identification numbers can be represented by using binary numbers with fewer digits, thereby increasing the available number of computing resource identifiers.

[0115] In some embodiments, the first serial number and the second serial number may be generated by a snowflake algorithm.

[0116] In some embodiments, the identification application request also includes second computing power information, and the second computing power information includes at least one of the following: the cloud where the computing power resource is located, the computing power type of the computing power resource, the computing power form of the computing power resource, the computing power unit of the computing power resource, the capacity included in the computing power resource, the time when the identification application request is received, and the price of the computing power resource. Here, the computing power form indicates that the computing power resource is directly provided by the server where the computing power resource corresponding to the identification number is located, or the computing power resource is provided by bare metal based on a physical machine, or the computing power resource is provided by a virtual machine or container based on a physical machine.

[0117] For example, the cloud where the computing power resource is located includes the name of the cloud. For example, the computing power unit of the computing power resource includes a single-precision floating point number (FP32).

[0118] In this case, relevant information matching the preset field can be extracted from the second computing power information; and based on the identification number and the relevant information, an identifier in the form of a uniform resource locator is generated as an identifier of the computing power resource.

[0119] In this way, by generating an identifier in the form of a uniform resource locator through an identification number and related information as an identifier of the computing power resource, the uniqueness of the computing power resource identifier can be further improved.

[0120] In some embodiments, the preset fields can be expanded according to actual needs. In this way, the uniform resource locator can be expanded by expanding the preset fields, so that the computing power resource identification method has scalability, is easier to distinguish and can meet future expansion needs for computing power identification.

[0121] In some embodiments, a user may click on a uniform resource locator to access the corresponding computing resource.

[0122] In this way, users can quickly access the corresponding computing resources by clicking, thus saving users' time.

[0123] In some embodiments, a computing resource that applies for a computing resource identifier is subjected to trusted authentication. If the computing resource fails the trusted authentication, the computing resource is not assigned an identifier.

[0124] In this way, identification application requests for computing resources that have not undergone trusted authentication can be rejected, thereby improving the security of computing resource identification.

[0125] For example, an identifier in the form of a uniform resource locator is generated as the identifier of the computing power resource, which can be expressed as URL:1123 / eCloudN / Tianyi Cloud / VM1 / CPU / FP32 / 5 / az1 / s1 / Beijing / 2021-01-01 09:00:00 / 1000.

[0126] From the logo, it can be clearly and easily read that this is a computing power resource with a computing power identification number of 1123, named eCloudN, which is the virtual machine computing power VM1 of Tianyi Cloud. It is the basic CPU computing power and is located on the server s1 of the computing power resource node az1. It is equivalent to 5 FP32 units of computing power. The geographical location corresponding to this computing power resource is in Beijing. The time when the application request was received was 2021-01-01 09:00:00. The price of 1 FP32 unit of computing power of this computing power resource is 1000.

[0127] In some embodiments, the identification application request may further include computing power resource network information. Here, the computing power resource network information may include the second network information of the third subpath.

[0128] As some implementation methods, when the computing power resource application is identified, the network device (such as a router) on the third sub-path also reports the second network information of the third sub-path.

[0129] In this way, there is no need to obtain additional network information of the third sub-path, which speeds up the determination of target computing resources.

[0130] In some embodiments, the second network information of the third subpath includes the price of the third subpath, and the computing power price can be determined according to the price of the third subpath. Here, the price of the third subpath can be determined according to the bandwidth of the third subpath and the connection mode of the third subpath. Here, the connection mode of the third subpath can be a wired connection or a wireless connection, and the wired connection can be, for example, an optical fiber connection, and the wireless connection can be, for example, a fifth generation mobile communication (5G) connection.

[0131] In this way, by combining the bandwidth of the path and the connection method of the path, the price of computing power resources can be measured, making the measured computing power price more reasonable, thereby making the determined target computing power resources more accurate.

[0132] In some embodiments, the second network information of the third sub-path may further include reliability and security of the path. Here, the reliability of the path may be best effort transmission or deterministic transmission; the security may indicate whether the path is securely encrypted, and the secure encryption method may be symmetric encryption (e.g., AES) or asymmetric encryption (e.g., RSA).

[0133] In this way, users can further put forward requirements for the reliability and security of the path, and then further fully consider user needs when selecting target computing resources, thereby further improving user experience.

[0134] In some embodiments, the identification number of the target computing power resource is sent to the user device so that the user device obtains the IP address of the target computing power resource from the computing power resolution platform based on the identification number.

[0135] In some embodiments, the specific access port number opened by the target computing power resource can also be obtained from the computing power analysis platform.

[0136] In this way, only the identification number of the computing power resource is sent instead of the entire computing power resource information, so it is not easy to cause excessive network load when it is published to the network, so that users can obtain the identification number of the target computing power resource in time, and thus obtain the target computing power resource in time from the computing power analysis platform.

[0137] In some embodiments, before the identification number of the target computing power resource is sent to the user device, it is confirmed whether the path between the computing power node of the target computing power resource and the user device is available.

[0138] In some embodiments, a path being unavailable may mean that the network path is saturated or the network path is interrupted.

[0139] In this way, by confirming whether the path is available, it is possible to avoid the situation where the user cannot receive the identification number of the target computing resource, thereby ensuring that the user can receive the identification number of the target computing resource.

[0140] In some embodiments, when the identifier of any one of the multiple computing power resources is lost, the identifier of any one of the multiple computing power resources can be obtained from the computing power resolution platform.

[0141] In this way, the computing power analysis platform also serves as a backup database, that is, in the event that the computing power resource data is lost, the lost computing power resource data can be obtained from the computing power analysis platform.

[0142] Each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device embodiment, since it basically corresponds to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0143] Figure 3 It is a structural diagram of a scheduling device for computing resources according to some embodiments of the present disclosure.

[0144] like Figure 3 As shown, the scheduling device for computing power resources includes a receiving module 301, which is configured to receive a resource scheduling request from a user, the resource scheduling request includes reference information, the reference information includes user demand information, the demand information includes computing power demand and network demand, the computing power demand includes the type and capacity of computing power resources required, the network demand includes first network information of a required first path, the first path is the path between the user device of the user and the computing power node providing the computing power resources; the determination module 302 is configured to determine the target computing power resources that meet the computing power demand and the network demand based on the reference information.

[0145] It should be understood that the computing resource scheduling device may also include other modules to perform other corresponding operations described above.

[0146] Figure 4 It is a structural diagram of a scheduling device for computing resources according to some embodiments of the present disclosure.

[0147] like Figure 4 As shown, the scheduling device 400 for computing resources includes a memory 401 and a processor 402 coupled to the memory 401 , and the processor 402 is configured to execute the method of any one of the aforementioned embodiments based on the instructions stored in the memory 401 .

[0148] The memory 401 may include, for example, a system memory, a fixed non-volatile storage medium, etc. The system memory may store, for example, an operating system, an application program, a boot loader, and other programs.

[0149] The computing resource scheduling device 400 may also include an input / output interface 403, a network interface 404, a storage interface 405, etc. These interfaces 403, 404, 405, and the memory 401 and the processor 402 may be connected, for example, via a bus 406. The input / output interface 403 provides a connection interface for input / output devices such as a display, a mouse, a keyboard, and a touch screen. The network interface 404 provides a connection interface for various networked devices. The storage interface 405 provides a connection interface for external storage devices such as SD cards and USB flash drives.

[0150] like Figure 5 As shown, the embodiment of the present disclosure also provides a scheduling system 500 for computing power resources, including a scheduling device 501 for computing power resources of any one of the above embodiments, configured to send an identification number of a target computing power resource to a user device; and a computing power resolution platform 502, configured to provide the IP address of the target computing power resource to the user device based on the identification number.

[0151] In some embodiments, there is two-way data communication between the computing power resource scheduling device 501 and the computing power analysis platform 502. After the computing power resource scheduling device 501 generates a computing power resource identifier for a computing power resource, the generated computing power resource identifier is sent to the computing power analysis platform 502. In the event that the identifier of any one of the multiple computing power resources is lost, the computing power analysis platform 502 provides the identifier of any one of the multiple computing power resources to the computing power resource scheduling device 501.

[0152] As some implementation methods, the computing power analysis platform 502 can be integrated with the computing power resource scheduling device 501.

[0153] In some embodiments, the computing power resource scheduling system 500 has two-way data communication with the data communication network 520 to receive the computing power resource identification application request and the resource scheduling request of the user device, and send the computing power resource identification to the computing power resource and send the computing power resource scheduling related information (such as the identification number of the target computing power resource) to the user device.

[0154] In some embodiments, there is two-way data communication between the user device 510 and the data communication network 520 to send resource scheduling requests and receive information related to computing resource scheduling.

[0155] In some embodiments, the computing power nodes 531 - 533 have two-way data communication with the data communication network 520 to send identification application requests and receive identification of computing power resources generated by the computing power resource scheduling device 501 .

[0156] In some embodiments, there is a network device (eg, a router) between the user equipment 510 and the data communication network 520 , and the network device can collect and report information about the network path between the user equipment 510 and the data communication network 520 .

[0157] In some embodiments, there is a network device (e.g., a router) between any one or more of the computing power nodes 531-533 and the data communication network 520, and the network device can collect and report information about the network path between any one or more of the computing power nodes 531-533 and the data communication network 520.

[0158] In some embodiments, the data communication network 520 may be one or a combination of the Internet, a metropolitan area network, and a local area network.

[0159] In some embodiments, the data communication network 520 may be accessed via wireless and / or wired connections.

[0160] In some embodiments, when the identification of any one of the multiple computing resources is lost, the computing power analysis platform in the scheduling system of the above computing resources provides the identification of any one of the multiple computing resources.

[0161] The embodiments of the present disclosure further provide a computer-readable storage medium, comprising computer program instructions, which implement the method of any one of the above embodiments when executed by a processor.

[0162] The embodiments of the present disclosure further provide a computer program product, including a computer program, which implements the method of any one of the above embodiments when executed by a processor.

[0163] So far, various embodiments of the present disclosure have been described in detail. In order to avoid obscuring the concept of the present disclosure, some details known in the art are not described. Based on the above description, those skilled in the art can fully understand how to implement the technical solution disclosed here.

[0164] Those skilled in the art will appreciate that the embodiments of the present disclosure may be provided as methods, systems, or computer program products. Therefore, the present disclosure may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present disclosure may take the form of a computer program product implemented on one or more computer-usable non-transient storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0165] The present disclosure is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present disclosure. It should be understood that the functions specified in one or more processes in the flowchart and / or one or more blocks in the block diagram can be implemented 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 generate instructions for implementing the functions specified in the flowchart. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0166] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0167] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0168] Although some specific embodiments of the present disclosure have been described in detail by way of examples, it should be understood by those skilled in the art that the above examples are for illustration only and are not intended to limit the scope of the present disclosure. It should be understood by those skilled in the art that the above embodiments may be modified or some technical features may be replaced by equivalents without departing from the scope and spirit of the present disclosure. The scope of the present disclosure is defined by the appended claims.

Claims

1. A method for scheduling computing resources, comprising: Receive a resource scheduling request from a user, the resource scheduling request including reference information, the reference information including demand information of the user and attribute information of the user, the demand information including computing power demand and network demand, the network demand including first network information of a required first path, the first path being a path between a user device of the user and a computing power node providing the computing power resources, and the attribute information including at least one of user category, user price sensitivity, user priority, and user location; Determine, according to the reference information, a target computing power resource that meets the computing power demand and the network demand, including: determining, according to a first corresponding table, a plurality of candidate computing power resources that meet the computing power demand and the network demand from a plurality of computing power resources; in a case where the total number of the resource scheduling requests that have been received is less than a predetermined number, sorting the plurality of candidate computing power resources according to the capacity and computing power price of the plurality of candidate computing power resources to select the candidate computing power resource with the largest capacity and the lowest computing power price as the target computing power resource; storing the identifier of the target computing power resource and the reference information in a second corresponding table in correspondence; in a case where the total number of the resource scheduling requests that have been received is not less than the predetermined number, selecting, according to the attribute information, a target computing power resource of a historical user similar to the user from the second corresponding table as the target computing power resource of the user; The first correspondence table includes the identifiers of the multiple computing resources and the second network information of the first subpath of the second path between the user equipment and the computing node corresponding to each computing resource, the second network information of the second subpath of the second path, and the second network information of the third subpath of the second path. The identifier of each computing resource includes an identification number and identification information. The identification information includes the computing price. The identification information indicates the computing situation. Among them, the first sub-path represents the path between the user device and the user-side access device, the second sub-path represents the path between the user-side access device and the computing power resource side access device, and the third sub-path represents the path between the computing power node corresponding to each computing power resource in the multiple computing power resources and the computing power resource side access device.

2. The method according to claim 1, wherein: The resource scheduling request carries the second network information of the first sub-path.

3. The method according to claim 1 or 2, wherein: The selecting, according to the attribute information, from the second correspondence table the target computing power resources of historical users similar to the user as the target computing power resources of the user comprises: Selecting a plurality of target computing resources of historical users similar to the user from the plurality of candidate computing resources; According to the capacities and computing power prices of the multiple target computing power resources, the multiple target computing power resources are sorted to select the target computing power resource with the largest capacity and the lowest computing power price as the target computing power resource of the user.

4. The method according to claim 1, further comprising: Receive an identification application request for each computing power resource among the multiple computing power resources, where the identification application request includes first computing power information, and the first computing power information of each computing power resource includes a computing power node corresponding to each computing power resource; An identifier is generated for each computing power resource in the multiple computing power resources according to the identifier generation information, where the identifier generation information includes the first computing power information.

5. The method according to claim 4, wherein: The first computing power information also includes the physical machine where each computing power resource among the multiple computing power resources is located and the container in the physical machine that provides the computing power resource, or the physical machine where the computing power resource is located and the virtual machine in the physical machine that provides the computing power resource.

6. The method according to claim 4, wherein: The identification generation information also includes time information, and the time information includes the current time and the preset time; Generating an identifier for each of the plurality of computing resources according to the identifier generation information includes: The identification number is generated based on the difference between the current time and the preset time.

7. The method according to claim 6, wherein: Generating an identifier for each of the plurality of computing resources according to the identifier generation information includes: Generate a first binary serial number according to the time information; A binary second serial number is generated according to the first computing power information, and the identification number includes the first serial number and the second serial number.

8. The method according to claim 4, wherein: The identification application request also includes second computing power information; The generating of the identifier for each of the plurality of computing resources according to the identifier generating information comprises: Extract relevant information matching a preset field from the second computing power information, wherein the second computing power information includes at least one of the following: the cloud where each of the multiple computing power resources is located, the computing power type of each of the multiple computing power resources, the computing power form of each of the multiple computing power resources, the computing power unit of each of the multiple computing power resources, the capacity of each of the multiple computing power resources and the time when the identification application request is received, and the price of each of the multiple computing power resources, and the capacity of each of the multiple computing power resources indicates the number of computing power units included in each of the multiple computing power resources; and According to the identification number and the related information, an identifier in the form of a uniform resource locator is generated as an identifier of each computing power resource in the multiple computing power resources.

9. The method according to claim 4, wherein: The identification application request also includes computing power resource network information, and the computing power resource network information includes the second network information of the third sub-path.

10. The method according to claim 9, wherein: The second network information of the third sub-path includes the price of the third sub-path, and the identification information includes the computing power price, the computing power price is determined according to the price of the third sub-path, and the price of the third sub-path is determined according to the bandwidth of the third sub-path and the connection method of the third sub-path.

11. The method according to claim 1 or 2, wherein: The first network information and the second network information include at least one of latency, bandwidth and packet loss rate.

12. The method according to claim 1 or 2, further comprising: The identification number of the target computing power resource is sent to the user device so that the user device obtains the Internet Protocol IP address of the target computing power resource from the computing power resolution platform based on the identification number.

13. The method according to claim 12, further comprising: Before the identification number of the target computing power resource is sent to the user equipment, it is confirmed whether the path between the computing power node of the target computing power resource and the user equipment is available.

14. The method according to claim 12, wherein: In the event that the identifier of any one of the multiple computing power resources is lost, the identifier of the any one computing power resource is obtained from the computing power analysis platform.

15. A computing resource scheduling device, comprising: a receiving module configured to receive a resource scheduling request from a user, the resource scheduling request including reference information, the reference information including demand information of the user and attribute information of the user, the demand information including computing power demand and network demand, the computing power demand including the computing power type and capacity of the required computing power resources, the network demand including first network information of a required first path, the first path being a path between a user device of the user and a computing power node providing the computing power resources, and the attribute information including at least one of user category, user price sensitivity, user priority, and user location; A determination module is configured to determine, according to the reference information, a target computing power resource that meets the computing power demand and the network demand, including: determining, according to a first corresponding table, a plurality of candidate computing power resources that meet the computing power demand and the network demand from a plurality of computing power resources; in a case where the total number of the resource scheduling requests that have been received is less than a predetermined number, sorting the plurality of candidate computing power resources according to the capacity and computing power price of the plurality of candidate computing power resources to select the candidate computing power resource with the largest capacity and the lowest computing power price as the target computing power resource; storing the identifier of the target computing power resource and the reference information in a second corresponding table in correspondence; in a case where the total number of the resource scheduling requests that have been received is not less than the predetermined number, selecting, according to the attribute information, a target computing power resource of a historical user similar to the user from the second corresponding table as the target computing power resource of the user; The first correspondence table includes the identifiers of the multiple computing resources and the second network information of the first subpath of the second path between the user equipment and the computing node corresponding to each computing resource, the second network information of the second subpath of the second path, and the second network information of the third subpath of the second path. The identifier of each computing resource includes an identification number and identification information. The identification information includes the computing price. The identification information indicates the computing situation. Among them, the first sub-path represents the path between the user device and the user-side access device, the second sub-path represents the path between the user-side access device and the computing power resource side access device, and the third sub-path represents the path between the computing power node corresponding to each computing power resource in the multiple computing power resources and the computing power resource side access device.

16. A computing resource scheduling device, comprising: Memory; as well as A processor coupled to the memory, configured to execute the method of any one of claims 1-14 based on instructions stored in the memory.

17. A computing resource scheduling system, comprising: The computing power resource scheduling device of claim 15 or 16, configured to send the identification number of the target computing power resource to the user equipment; The computing power analysis platform is configured to provide the IP address of the target computing power resource to the user device based on the identification number.

18. The computing resource scheduling system according to claim 17, wherein: In the event that the identifier of any one of the multiple computing power resources is lost, the computing power analysis platform provides the identifier of the any one computing power resource.

19. A computer-readable storage medium comprising computer program instructions, wherein: When the computer program instructions are executed by a processor, the method according to any one of claims 1 to 14 is implemented.

20. A computer program product comprising a computer program, wherein: When the computer program is executed by a processor, the method according to any one of claims 1 to 14 is implemented.

Images