技术领域technical field
本发明涉及互联网技术领域，特别是涉及一种云计算测试资源分配的确定方法及装置。The invention relates to the technical field of the Internet, in particular to a method and device for determining allocation of cloud computing test resources.
背景技术Background technique
云计算测试，是指基于云计算技术，将包含被测对象的基础设施抽象，集中存储于资源池中，通过资源分配算法，从资源池中划分出一部分资源，组装成平台承载测试应用，从而实现按需、定制化地为测试用户提供测试服务。Cloud computing test refers to the abstraction of the infrastructure containing the tested object based on cloud computing technology, and centralized storage in the resource pool. Through the resource allocation algorithm, a part of the resources are divided from the resource pool and assembled into a platform to carry the test application, so that Realize on-demand and customized testing services for test users.
云计算测试的基本框架的示意图如图1所示，自底向上依次为基础设施层(包括基础设施和资源池)、平台层和应用层。基础设施包括：被测装备、执行测试的测试工具、存储测试数据的服务器和连接各个测试实体的网络。通过资源抽象，基础设施的信息被抽象为计算资源、存储资源和网络资源，集中存储在资源池当中。利用资源分配算法，从资源池中划分出一部分资源，组装成平台。一个平台，由云主机、虚拟网络链路、虚拟网络节点，虚拟存储空间共同组成。平台中的每一个实体都是由计算资源、存储资源和空间资源共同组成。平台与平台之间是相互隔离、互不影响的。每一个平台都能够作为执行一个测试的环境。多个平台所承载的不同测试，共同组成了应用层。在云计算测试体系中，所有资源的使用都在管理层的监督和控制之下。The schematic diagram of the basic framework of cloud computing testing is shown in Figure 1, which is the infrastructure layer (including infrastructure and resource pools), platform layer and application layer from bottom to top. The infrastructure includes: the equipment under test, the test tool to perform the test, the server to store the test data and the network connecting the various test entities. Through resource abstraction, infrastructure information is abstracted into computing resources, storage resources, and network resources, which are centrally stored in resource pools. Use the resource allocation algorithm to divide some resources from the resource pool and assemble them into platforms. A platform consists of cloud hosts, virtual network links, virtual network nodes, and virtual storage space. Each entity in the platform is composed of computing resources, storage resources and space resources. Platforms are isolated from each other and do not affect each other. Each platform can serve as an environment for executing a test. Different tests carried by multiple platforms together form the application layer. In a cloud computing testing system, the use of all resources is under the supervision and control of management.
虽然云计算测试能够使测试变得像使用自来水或者电一样，方便地完成测试，而不必关注测试设备或配置测试环境，但是云计算测试的资源是要受限于基础设施的总量。因此，当大量测试用例需要同时运行或者执行高并发测试用例时，如何分配有限的资源，以最小的成本获得最大的收益，就成为亟待解决的问题。Although cloud computing testing can make testing as convenient as using tap water or electricity, without having to pay attention to testing equipment or configuring the testing environment, the resources of cloud computing testing are limited by the total amount of infrastructure. Therefore, when a large number of test cases need to run at the same time or execute high-concurrency test cases, how to allocate limited resources and obtain maximum benefits at minimum cost has become an urgent problem to be solved.
现有的云计算测试资源分配的确定方法，只考虑云计算中虚拟网络资源的分配，但是在实际的云计算测试中，网络资源分配只是资源分配的一部分，计算资源和存储资源也在云计算测试中发挥重要作用，因此也对分配结果有影响，但现有的方法并没有将这些因素考虑在内，从而影响了最终确定的云计算测试资源分配结果的准确性和有效性。The existing methods for determining resource allocation for cloud computing tests only consider the allocation of virtual network resources in cloud computing, but in actual cloud computing tests, network resource allocation is only a part of resource allocation, and computing resources and storage resources are also included in cloud computing. The test plays an important role, so it also has an impact on the allocation results, but the existing methods do not take these factors into account, thus affecting the accuracy and effectiveness of the final cloud computing test resource allocation results.
发明内容Contents of the invention
本发明提供一种云计算测试资源分配的确定方法及装置，用以解决现有技术的如下问题：现有云计算测试资源分配的确定方法，仅考虑了虚拟的网络资源分配，并未考虑其他类型的资源分配，影响了最终确定的云计算测试资源分配结果的准确性和有效性。The present invention provides a method and device for determining the allocation of cloud computing test resources to solve the following problems in the prior art: the existing determination method for cloud computing test resource allocation only considers virtual network resource allocation and does not consider other The type of resource allocation affects the accuracy and effectiveness of the final cloud computing test resource allocation results.
为解决上述技术问题，本发明提供一种云计算测试资源分配的确定方法及装置，方法包括：根据云计算测试资源的最大费效比建立分配所述云计算测试资源的初始数学模型；根据所述初始数学模型建立分配所述云计算测试资源的扩展数学模型；按照预定输入数据求解所述扩展数学模型，以确定所述云计算测试资源的分配结果。In order to solve the above technical problems, the present invention provides a method and device for determining the allocation of cloud computing test resources. The method includes: establishing an initial mathematical model for allocating the cloud computing test resources according to the maximum cost-effectiveness ratio of the cloud computing test resources; The initial mathematical model establishes an extended mathematical model for allocating the cloud computing test resources; solving the extended mathematical model according to predetermined input data to determine the allocation result of the cloud computing test resources.
可选的，根据云计算测试资源的最大费效比建立分配所述云计算测试资源的初始数学模型，包括：根据所述云计算测试资源的最大费效比建立所述初始数学模型的目标函数，其中，所述目标函数表示分配给进行云计算测试的第i个用户的云计算测试资源所产生的费用的最大值；根据所述目标函数建立所述初始数学模型的约束条件。Optionally, establishing an initial mathematical model for allocating the cloud computing test resources according to the maximum cost-effectiveness ratio of the cloud computing test resources includes: establishing an objective function of the initial mathematical model according to the maximum cost-effectiveness ratio of the cloud computing test resources , wherein the objective function represents the maximum value of the cost generated by the cloud computing test resources assigned to the i-th user performing the cloud computing test; the constraints of the initial mathematical model are established according to the objective function.
可选的，根据所述初始数学模型建立分配所述云计算测试资源的扩展数学模型，包括：确定所述扩展数学模型的目标函数，其中，所述扩展数学模型的目标函数与所述初始数学模型的目标函数相同；将所述初始数学模型的约束条件中的不等式按照预定数学方法转化为等式，以确定所述扩展数学模型的约束条件。Optionally, establishing an extended mathematical model for allocating the cloud computing test resources according to the initial mathematical model includes: determining an objective function of the extended mathematical model, wherein the objective function of the extended mathematical model is the same as the initial mathematical model The objective functions of the models are the same; the inequalities in the constraints of the initial mathematical model are transformed into equations according to a predetermined mathematical method to determine the constraints of the extended mathematical model.
可选的，按照预定输入数据对所述扩展数学模型进行求解，以确定所述云计算测试资源的分配结果，包括：将所述预定输入数据输入至所述扩展数学模型中，并采用拉格朗日乘子法对输入所述预定输入数据的所述扩展数学模型进行求解，以确定所述云计算测试资源的分配结果。Optionally, solving the extended mathematical model according to predetermined input data to determine the allocation result of the cloud computing test resource includes: inputting the predetermined input data into the extended mathematical model, and using Lager The Langerian multiplier method solves the extended mathematical model input with the predetermined input data, so as to determine the allocation result of the cloud computing test resources.
可选的，所述云计算测试资源包括：网络资源、计算资源和存储资源。Optionally, the cloud computing test resources include: network resources, computing resources and storage resources.
本发明提供的云计算测试资源分配的确定方法及装置，方法首先根据云计算测试资源的最大费效比建立分配云计算测试资源的初始数学模型，再根据初始数学模型建立可求解的分配云计算测试资源的扩展数学模型，最后按照预定输入的已知量数据求解上述扩展数学模型，以确定云计算测试资源的分配结果。该方法可以根据实际需要，确定云计算测试中分配给用户的各种资源的具体数值，并可以确定在合理进行资源分配的基础上可以得到的最大收益，解决了现有技术的如下问题：现有云计算测试资源分配的确定方法，仅考虑了虚拟的网络资源分配，并未考虑其他类型的资源分配，影响最终确定的云计算测试资源分配结果的准确性和有效性。The method and device for determining the allocation of cloud computing test resources provided by the present invention, the method first establishes an initial mathematical model for allocating cloud computing test resources according to the maximum cost-effectiveness ratio of cloud computing test resources, and then establishes a solvable allocation cloud computing model according to the initial mathematical model The extended mathematical model of test resources, and finally solve the above extended mathematical model according to the predetermined input known quantity data, so as to determine the allocation result of cloud computing test resources. This method can determine the specific values of various resources allocated to users in the cloud computing test according to actual needs, and can determine the maximum benefit that can be obtained on the basis of reasonable resource allocation, and solves the following problems in the prior art: There is a method for determining the allocation of cloud computing test resources, which only considers virtual network resource allocation and does not consider other types of resource allocation, which affects the accuracy and effectiveness of the final cloud computing test resource allocation results.
附图说明Description of drawings
图1是本发明背景技术中云计算测试的基本框架的示意图；Fig. 1 is the schematic diagram of the basic frame of cloud computing test in the background technology of the present invention;
图2是本发明第一实施例中云计算测试资源分配的确定方法的流程图；Fig. 2 is a flowchart of a method for determining cloud computing test resource allocation in the first embodiment of the present invention;
图3是本发明第二实施例中云计算测试资源分配的确定装置的结构示意图；3 is a schematic structural diagram of a device for determining cloud computing test resource allocation in a second embodiment of the present invention;
图4是本发明第三实施例中云计算测试资源分配的确定方法的流程图；4 is a flowchart of a method for determining cloud computing test resource allocation in the third embodiment of the present invention;
图5是本发明第三实施例中表示云计算测试资源所形成的测试平台结构示意图；Fig. 5 is a schematic diagram showing the structure of a test platform formed by cloud computing test resources in the third embodiment of the present invention;
图6是本发明第三实施例中云计算测试资源分配的确定方法的另一个流程图。Fig. 6 is another flow chart of the method for determining cloud computing test resource allocation in the third embodiment of the present invention.
具体实施方式Detailed ways
为了解决现有技术的如下问题：现有云计算测试资源分配的确定方法，仅考虑了虚拟的网络资源分配，并未考虑其他类型的资源分配，影响最终确定的云计算测试资源分配结果的准确性和有效性。本实施例提供了一种云计算测试资源分配的确定方法，该方法的流程图如图2所示，包括步骤S202至S206：In order to solve the following problems in the prior art: the existing method for determining the allocation of cloud computing test resources only considers virtual network resource allocation, and does not consider other types of resource allocation, which affects the accuracy of the final cloud computing test resource allocation results. sex and effectiveness. This embodiment provides a method for determining the allocation of cloud computing test resources. The flow chart of the method is shown in FIG. 2 , including steps S202 to S206:
S202，根据云计算测试资源的最大费效比建立分配云计算测试资源的初始数学模型。S202. Establish an initial mathematical model for allocating cloud computing test resources according to the maximum cost-effectiveness ratio of cloud computing test resources.
在本实施例中，考虑的是作为提供服务的一方，将云计算测试资源提供给用户从而获得收益的问题，因此在具体实现时，要根据最大费效比建立对云计算测试进行资源分配的数学模型，即如何实现投入尽可能少的云计算测试资源给用户，获得尽可能大的收益。In this embodiment, the problem of providing cloud computing test resources to users as a service provider is considered to obtain benefits. Therefore, in the actual implementation, it is necessary to establish a resource allocation method for cloud computing tests according to the maximum cost-effectiveness ratio. Mathematical model, that is, how to invest as little cloud computing testing resources as possible to users and get as much benefit as possible.
S204，根据初始数学模型建立分配云计算测试资源的扩展数学模型。S204. Establish an extended mathematical model for allocating cloud computing test resources according to the initial mathematical model.
这一步骤的目的是将上述的初始数学模型进行改进，转变为可以求解的数学模型，为确定云计算测试资源的分配奠定基础。The purpose of this step is to improve the above-mentioned initial mathematical model and transform it into a mathematical model that can be solved, so as to lay a foundation for determining the allocation of cloud computing test resources.
S206，按照预定输入数据求解扩展数学模型，以确定云计算测试资源的分配结果。S206, solving the extended mathematical model according to the predetermined input data, so as to determine the allocation result of the cloud computing test resources.
在得到可求解的扩展数学模型之后，就可以采用特定的数学方法对其进行求解，并且，根据云计算测试资源分配建立的数学模型，有一些源自云计算测试平台的已知量，要将这些已知量输入到扩展数学模型中，再进行求解。After obtaining the solvable extended mathematical model, specific mathematical methods can be used to solve it, and the mathematical model established according to the allocation of cloud computing test resources has some known quantities from the cloud computing test platform. These known quantities are input into the extended mathematical model and then solved.
本实施例提供的云计算测试资源分配的确定方法，首先根据云计算测试资源的最大费效比建立分配云计算测试资源的初始数学模型，再根据初始数学模型建立可求解的分配云计算测试资源的扩展数学模型，最后按照预定输入的已知量数据求解上述扩展数学模型，以确定云计算测试资源的分配结果。该方法可以根据实际需要，确定云计算测试中分配给用户的各种资源的具体数值，并可以确定在合理进行资源分配的基础上可以得到的最大收益，解决了现有技术的如下问题：现有云计算测试资源分配的确定方法，仅考虑了虚拟的网络资源分配，并未考虑其他类型的资源分配，影响最终确定的云计算测试资源分配结果的准确性和有效性。The method for determining the allocation of cloud computing test resources provided in this embodiment first establishes an initial mathematical model for allocating cloud computing test resources according to the maximum cost-effectiveness ratio of cloud computing test resources, and then establishes a solvable allocation of cloud computing test resources according to the initial mathematical model Finally, solve the above-mentioned extended mathematical model according to the predetermined input known quantity data to determine the allocation result of cloud computing test resources. This method can determine the specific values of various resources allocated to users in the cloud computing test according to actual needs, and can determine the maximum benefit that can be obtained on the basis of reasonable resource allocation, and solves the following problems in the prior art: There is a method for determining the allocation of cloud computing test resources, which only considers virtual network resource allocation and does not consider other types of resource allocation, which affects the accuracy and effectiveness of the final cloud computing test resource allocation results.
此外，本实施中还对建立分配云计算测试资源的初始数学模型进行了具体限定，可以按照数学规划的方法建立上述初始模型，即建立初始模型中的目标函数，本实施例中其具体表示的是分配给进行云计算测试的第i个用户的云计算测试资源所产生的费用的最大值。在确定目标函数后，还要根据目标函数建立初始数学模型的约束条件，以保证建立的初始数学模型符合云计算测试资源分配的实际情况。In addition, in this implementation, the establishment of the initial mathematical model for allocating cloud computing test resources is also specifically limited. The above-mentioned initial model can be established according to the method of mathematical programming, that is, the objective function in the initial model is established. In this embodiment, it is specifically expressed is the maximum cost of cloud computing test resources allocated to the i-th user for cloud computing testing. After determining the objective function, the constraints of the initial mathematical model should also be established according to the objective function to ensure that the established initial mathematical model conforms to the actual situation of cloud computing test resource allocation.
进一步，为建立可求解的数学模型，还要建立分配云计算测试资源的扩展数学模型，建立扩展数学模型也要遵循数学规划的方法，具体是：确定扩展数学模型的目标函数，在本实施例中中，扩展数学模型的目标函数与上述初始数学模型的目标函数相同；之后，将初始数学模型的约束条件中的不等式按照预定数学方法转化为等式，以确定扩展数学模型的约束条件。这样的建模方式是为进一步求解奠定可实现的数学基础。Further, in order to establish a solvable mathematical model, an extended mathematical model for allocating cloud computing test resources must also be established, and the establishment of an extended mathematical model must also follow the method of mathematical programming, specifically: determine the objective function of the extended mathematical model, in this embodiment In the middle, the objective function of the extended mathematical model is the same as the objective function of the above-mentioned initial mathematical model; after that, the inequalities in the constraints of the initial mathematical model are converted into equations according to a predetermined mathematical method to determine the constraints of the extended mathematical model. Such a modeling method is to lay a realistic mathematical foundation for further solving.
在获得扩展数学模型之后，可以将表示云计算测试资源分配时间情况的预定输入数据输入至扩展数学模型中，并采用拉格朗日乘子法对扩展数学模型进行求解，由于本实施例中的扩展数学模型由目标函数和相应的约束条件组成，采用拉格朗日乘子法是求解该类型数学模型的常用方法，求解扩展数学模型就可以确定云计算测试资源的分配结果。其中，本实施例中的云计算测试资源包括：网络资源、计算资源和存储资源。After the extended mathematical model is obtained, the predetermined input data representing the allocation time of the cloud computing test resources can be input into the extended mathematical model, and the extended mathematical model is solved by using the Lagrange multiplier method, due to the The extended mathematical model is composed of the objective function and the corresponding constraints. Using the Lagrange multiplier method is a common method to solve this type of mathematical model. Solving the extended mathematical model can determine the allocation results of cloud computing test resources. Wherein, the cloud computing test resources in this embodiment include: network resources, computing resources and storage resources.
本发明第二实施例提供了一种云计算测试资源分配的确定装置，该装置的结构示意图如图3所示，包括：第一建立模块10，用于根据云计算测试资源的最大费效比建立分配云计算测试资源的初始数学模型；第二建立模块20，与第一建立模块10耦合，用于根据初始数学模型建立分配云计算测试资源的扩展数学模型；确定模块30，与第二建立模块20耦合，用于按照预定输入数据求解扩展数学模型，以确定云计算测试资源的分配结果。The second embodiment of the present invention provides a device for determining the allocation of cloud computing test resources. The structural diagram of the device is shown in FIG. Set up the initial mathematical model of allocating cloud computing test resources; The second building module 20, coupled with the first building module 10, is used to set up the extended mathematical model of allocating cloud computing test resources according to the initial mathematical model; Determination module 30, establishes with the second The module 20 is coupled to solve the extended mathematical model according to the predetermined input data, so as to determine the allocation result of the cloud computing test resources.
在本实施例中，实际上考虑的是作为提供云计算测试服务的一方，将云计算测试资源提供给用户进而获得收益的问题，因此在具体实现时，第一建立模块可以根据最大费效比建立对云计算测试进行资源分配的数学模型，即如何实现投入尽可能少的云计算测试资源给用户，获得尽可能大的收益。In this embodiment, what is actually considered is that as a party providing cloud computing testing services, the problem of providing cloud computing testing resources to users to obtain benefits, so in specific implementation, the first building module can be based on the maximum cost-effectiveness ratio Establish a mathematical model for cloud computing test resource allocation, that is, how to invest as little cloud computing test resources as possible to users and obtain as much benefit as possible.
进一步，为了将上述的初始数学模型进行改进，转变为可以求解的数学模型，确定装置中的第二建立模块可以根据上述初始数学模型建立分配云计算测试资源的扩展数学模型，为确定云计算测试资源的分配奠定基础。Further, in order to improve the above-mentioned initial mathematical model and transform it into a mathematical model that can be solved, the second building module in the determination device can establish an extended mathematical model for allocating cloud computing test resources according to the above-mentioned initial mathematical model, in order to determine the cloud computing test. The basis for the allocation of resources.
进一步，在得到可求解的扩展数学模型之后，确定模块就可以采用特定的数学方法对其进行求解，并且，根据云计算测试资源分配建立的数学模型，有一些源自云计算测试平台的已知量，确定模块要将这些已知量输入到扩展数学模型中，再进行求解，以确定云计算测试资源的分配结果。Further, after obtaining the solvable extended mathematical model, the determination module can use a specific mathematical method to solve it, and, according to the mathematical model established by the cloud computing test resource allocation, there are some known data from the cloud computing test platform. Quantities, the determination module should input these known quantities into the extended mathematical model, and then solve them to determine the allocation results of cloud computing test resources.
此外，本实施中，第一建立模块具体可以用于：按照数学规划的方法建立上述初始模型中的目标函数，在本实施例中，其具体表示的是分配给进行云计算测试的第i个用户的云计算测试资源所产生的费用的最大值。在确定目标函数后，第一建立模块还要用于根据目标函数建立初始数学模型的约束条件，以保证第一建立模块所建立的初始数学模型符合云计算测试资源分配的实际情况。In addition, in this implementation, the first building module can be specifically used to: establish the objective function in the above initial model according to the method of mathematical programming. The maximum cost incurred by the user's cloud computing test resources. After determining the objective function, the first establishment module is also used to establish the constraints of the initial mathematical model according to the objective function, so as to ensure that the initial mathematical model established by the first establishment module conforms to the actual situation of cloud computing test resource allocation.
进一步，为建立可求解的数学模型，第二建立模块所建立的分配云计算测试资源的扩展数学模型，其也要遵循数学规划的方法，第二建立模块具体用于：确定扩展数学模型的目标函数。在本实施例中，扩展数学模型的目标函数与上述初始数学模型的目标函数相同，表示的是分配给进行云计算测试的第i个用户的云计算测试资源所产生的费用的最大值。之后，第二建立模块将初始数学模型的约束条件中的不等式按照预定数学方法转化为等式，以确定扩展数学模型的约束条件。第二建立模块这样的建模方式是为进一步求解奠定可实现的数学基础。Further, in order to establish a solvable mathematical model, the extended mathematical model for allocating cloud computing test resources established by the second establishment module also follows the method of mathematical programming, and the second establishment module is specifically used to: determine the target of the extended mathematical model function. In this embodiment, the objective function of the extended mathematical model is the same as that of the above initial mathematical model, and represents the maximum cost generated by the cloud computing test resources allocated to the i-th user performing the cloud computing test. Afterwards, the second building module converts the inequalities in the constraints of the initial mathematical model into equations according to a predetermined mathematical method, so as to determine the constraints of the extended mathematical model. The modeling method of the second building block is to lay a realizable mathematical foundation for further solving.
在获得扩展数学模型之后，确定模块可以将表示云计算测试资源分配时间情况的预定输入数据输入至扩展数学模型中，并采用拉格朗日乘子法对扩展数学模型进行求解。由于本实施例中的扩展数学模型由目标函数和相应的约束条件组成，采用拉格朗日乘子法是求解该类型数学模型的常用的有效方法，确定模块求解扩展数学模型就可以确定云计算测试资源的分配结果。本实施例与第一实施例相同，云计算测试资源包括：网络资源、计算资源和存储资源。After obtaining the extended mathematical model, the determination module may input predetermined input data representing the allocation time of cloud computing test resources into the extended mathematical model, and solve the extended mathematical model by using the Lagrange multiplier method. Since the extended mathematical model in this embodiment is composed of an objective function and corresponding constraints, using the Lagrange multiplier method is a common and effective method for solving this type of mathematical model, and determining the module to solve the extended mathematical model can determine the cloud computing Test resource allocation results. This embodiment is the same as the first embodiment, and the cloud computing test resources include: network resources, computing resources and storage resources.
本实施例提供的云计算测试资源分配的确定装置，首先第一建立模块根据云计算测试资源的最大费效比建立分配云计算测试资源的初始数学模型，再由第二建立模块根据初始数学模型建立可求解的分配云计算测试资源的扩展数学模型，最后确定模块按照预定输入的已知量数据求解上述扩展数学模型，以确定云计算测试资源的分配结果。该装置可以根据实际需要，确定云计算测试中分配给用户的各种资源的具体数值，并可以确定在合理进行资源分配的基础上可以得到的最大收益，解决了现有技术的如下问题：现有云计算测试资源分配的确定装置，仅考虑了虚拟的网络资源分配，并未考虑其他类型的资源分配，影响最终确定的云计算测试资源分配结果的准确性和有效性。In the device for determining the allocation of cloud computing test resources provided in this embodiment, firstly, the first building module establishes an initial mathematical model for allocating cloud computing test resources according to the maximum cost-effectiveness ratio of cloud computing test resources, and then the second building module establishes an initial mathematical model according to the initial mathematical model Establish a solvable extended mathematical model for allocating cloud computing test resources, and finally determine the module to solve the above-mentioned extended mathematical model according to the predetermined input known quantity data, so as to determine the allocation result of cloud computing test resources. The device can determine the specific values of various resources allocated to users in the cloud computing test according to actual needs, and can determine the maximum benefit that can be obtained on the basis of reasonable resource allocation, which solves the following problems in the prior art: There is a device for determining the allocation of cloud computing test resources, which only considers virtual network resource allocation and does not consider other types of resource allocation, which affects the accuracy and effectiveness of the final cloud computing test resource allocation results.
本发明第三实施例提供了一种云计算测试资源分配的确定方法，该方法的流程图如图4所示，包括步骤S402至S408。The third embodiment of the present invention provides a method for determining allocation of cloud computing test resources. The flow chart of the method is shown in FIG. 4 , including steps S402 to S408.
为了叙述方便，如表1所示，本实施例定义了要用到的符号所表示的意义。For the convenience of description, as shown in Table 1, this embodiment defines the meanings of the symbols to be used.
表1符号的意义The meaning of the symbols in Table 1
需要说明的是，在具体实现时，表示分配给第i个用户的云计算测试资源所形成的测试平台结构示意图如图5所示。其中，提供给第i个用户的云计算测试资源包括：云主机xi台，虚拟节点yi台，虚拟链路带宽zi Mbps，网络拓扑结构Ai，在本实施例中，云计算测试中各部分的含义如下：It should be noted that, in actual implementation, a schematic structural diagram of a test platform formed by cloud computing test resources allocated to the i-th user is shown in FIG. 5 . Among them, the cloud computing test resources provided to the i-th user include: cloud host xi platform, virtual node yi platform, virtual link bandwidth zi Mbps, network topology Ai , in this embodiment, cloud computing test The meaning of each part is as follows:
云主机：承载云计算测试业务的实体，由CPU、内存和硬盘按照一定的比例共同组成。Cloud host: The entity that carries the cloud computing test business, which is composed of CPU, memory and hard disk in a certain proportion.
虚拟节点：承载云计算测试业务数据转发任务的实体，由CPU和内存共同组成。Virtual node: An entity that carries cloud computing test business data forwarding tasks, consisting of CPU and memory.
虚拟链路带宽：单条承载云计算测试业务数据链路的容量，由带宽组成。Virtual link bandwidth: the capacity of a single link carrying cloud computing test service data, consisting of bandwidth.
网络拓扑结构：表明云主机和虚拟网络节点之间连接关系的矩阵，是一个(xi+yi)*(xi+yi)的矩阵。Network topology: a matrix indicating the connection relationship between cloud hosts and virtual network nodes, which is a matrix of (xi +yi )*(xi +yi ).
S402，建立表示分配给第i个用户云计算测试资源的原始模型。S402. Establish an original model representing the cloud computing test resources allocated to the i-th user.
本实施例中，上述原始模型相当于第一实施例在的初始数学模型，其中，目标函数表示为：In this embodiment, the above-mentioned original model is equivalent to the initial mathematical model in the first embodiment, wherein the objective function is expressed as:
max[M1*(N1*xi+N4*yi)+M2*(N2*xi+N5*yi)max[M1 *(N1 *xi +N4 *yi )+M2 *(N2 *xi +N5 *yi )
+M3*(N3*xi)+M4*(Ii*zi)] (1)+M3 *(N3 *xi )+M4 *(Ii *zi )] (1)
约束条件为：The constraints are:
N1*xi+N4*yi≤Cai (2)N1 *xi +N4 *yi ≤Cai (2)
N2*xi+N5*yi≤Ca2i (3)N2 *xi +N5 *yi ≤Ca2i (3)
N3*xi≤Ca3i (4)N3 *xi ≤ Ca3i (4)
zi≤Ca4i (5)zi ≤ Ca4i (5)
公式(1)表示求取所有资源的最高费效比，小括号中的4个多项式依次表示CPU、内存、硬盘和带宽的需求量。公式(2)的物理意义为第i个用户对CPU的需求量不大于现有的可用CPU容量。公式(3)的物理意义为内存的需求量不大于现有的可用内存容量。公式(4)的物理意义为硬盘的需求量不大于现有的可用硬盘容量。公式(5)的物理意义为CPU的需求量不大于现有的可用CPU容量。Formula (1) means to find the highest cost-effectiveness ratio of all resources, and the four polynomials in parentheses represent the demand for CPU, memory, hard disk, and bandwidth in turn. The physical meaning of formula (2) is that the i-th user's demand for CPU is not greater than the existing available CPU capacity. The physical meaning of formula (3) is that the memory demand is not greater than the existing available memory capacity. The physical meaning of formula (4) is that the demand of hard disk is not greater than the existing available hard disk capacity. The physical meaning of formula (5) is that the CPU demand is not greater than the existing available CPU capacity.
S404，根据原始模型建立扩展模型。S404. Establish an extended model according to the original model.
为了求解上述原始模型，本实施例中可以引入约束条件变量w1、w2、w3、w4，以将上述原始模型中的不等式化为等式，进而能够得到扩展模型。In order to solve the above original model, constraint variables w1 , w2 , w3 , w4 may be introduced in this embodiment, so as to convert the inequalities in the above original model into equations, and then an extended model can be obtained.
扩展模型的目标函数表示为：The objective function of the extended model is expressed as:
max[M1*(N1*xi+N4*yi)+M2*(N2*xi+N5*yi)max[M1 *(N1 *xi +N4 *yi )+M2 *(N2 *xi +N5 *yi )
+M3*(N3*xi)+M4*(Ii*zi)] (9)+M3 *(N3 *xi )+M4 *(Ii *zi )] (9)
扩展模型的约束条件表示为：The constraints of the extended model are expressed as:
N1*xi+N4*yi+w1-Ca1i＝0 (10)N1 *xi +N4 *yi +w1 -Ca1i =0 (10)
N2*xi+N5*yi+w2-Ca2i＝0 (11)N2 *xi +N5 *yi +w2 -Ca2i ＝0 (11)
N3*xi+w3-Ca3i＝0 (12)N3 *xi +w3 -Ca3i =0 (12)
zi+w4-Ca4i＝0 (13)zi +w4 -Ca4i ＝0 (13)
N6≤Ca4i (14)N6 ≤ Ca4i (14)
S406，对扩展模型采用拉格朗日乘子法进行求解。S406, solving the extended model by using the Lagrangian multiplier method.
本实施例中，根据扩展模型的目标函数和约束条件形成的拉格朗日函数表示为：In this embodiment, the Lagrangian function formed according to the objective function and constraints of the extended model is expressed as:
F＝M1*(N1*xi+N4*yi)+M2*(N2*xi+N5*yi)F＝M1 *(N1 *xi +N4 *yi )+M2 *(N2 *xi +N5 *yi )
+M3*(N3*xi)+M4*(Ii*zi)-λ1(N1*xi+N4*yi+w1-Ca1i)+M3 *(N3 *xi )+M4 *(Ii *zi )-λ1 (N1 *xi +N4 *yi +w1 -Ca1i )
-λ2(N2*xi+N5*yi+w2-Ca2i)-λ3(N3*xi+w3-Ca3i)-λ2 (N2 *xi +N5 *yi +w2 -Ca2i )-λ3 (N3 *xi +w3 -Ca3i )
-λ4(zi+w4-Ca4i) (18)-λ4 (zi +w4 -Ca4i ) (18)
进一步，根据拉格朗日乘子法解方程：Further, solve the equation according to the Lagrange multiplier method:
通过求解公式(19)，得到N1，N2，N3，N4，N5，N6，λ1，λ2，λ3，λ4，从而确定分配给第i个用户的云计算测试资源。By solving formula (19), N1 , N2 , N3 , N4 , N5 , N6 , λ1 , λ2 , λ3 , λ4 are obtained, so as to determine the cloud computing test assigned to the i-th user resource.
S408，判断计算结果，确定云计算测试资源的分配。S408, judging the calculation result, and determining the allocation of cloud computing test resources.
判断公式(19)的解是否存在。如果解存在，按照求得的解进行对第i个用户进行云计算测试资源分配。本实施例的云计算测试资源分配的确定方法，也可以按照算法的流程图形式，加入循环执行的框图，可以取i＝i+1，再跳转到S402，计算第i+1个用户的云计算资源分配的请求。如果公式(19)的解不存在，则拒绝本次请求，然后取i＝i+1，跳转到S402，执行下一个用户的资源请求。将本实施例中的方法按照算法流程图进行表示，其流程图如图6所示。Determine whether the solution of formula (19) exists. If the solution exists, allocate cloud computing test resources to the i-th user according to the obtained solution. The method for determining the allocation of cloud computing test resources in this embodiment can also add a block diagram of cyclic execution according to the flow chart of the algorithm, and can take i=i+1, and then jump to S402 to calculate the i+1th user. A request for cloud computing resource allocation. If the solution of formula (19) does not exist, reject this request, then take i=i+1, jump to S402, and execute the resource request of the next user. The method in this embodiment is represented according to an algorithm flow chart, and the flow chart is shown in FIG. 6 .
本实施例中，管理层在收到每个用户的云计算测试资源分配请求后，根据最大费效比建立的表示云计算测试资源分配的初始模型，同时确定该模型的计算资源、存储资源和网络资源的约束条件，再对初始模型进行改进得到扩展模型，去除初始模型中的不等式，最后通过拉格朗日乘子法求解扩展模型，得到资源分配的结果。本实施例的云计算测试资源分配的确定方法，建立了用户请求资源与计算资源、存储资源和网络资源的对应关系，解决了现有技术的如下问题：现有云计算测试资源分配的确定装置，仅考虑了虚拟的网络资源分配，并未考虑其他类型的资源分配，影响最终确定的云计算测试资源分配结果的准确性和有效性。In this embodiment, after receiving each user's cloud computing test resource allocation request, the management layer establishes an initial model representing the cloud computing test resource allocation based on the maximum cost-effectiveness ratio, and simultaneously determines the computing resources, storage resources and According to the constraints of network resources, the initial model is improved to obtain the extended model, and the inequalities in the initial model are removed. Finally, the extended model is solved by the Lagrange multiplier method to obtain the result of resource allocation. The method for determining the allocation of cloud computing test resources in this embodiment establishes the corresponding relationship between user request resources and computing resources, storage resources, and network resources, and solves the following problems in the prior art: the existing cloud computing test resource allocation determination device , only virtual network resource allocation is considered, and other types of resource allocation are not considered, which affects the accuracy and effectiveness of the final cloud computing test resource allocation results.
尽管为示例目的，已经公开了本发明的优选实施例，本领域的技术人员将意识到各种改进、增加和取代也是可能的，因此，本发明的范围应当不限于上述实施例。Although preferred embodiments of the present invention have been disclosed for illustrative purposes, those skilled in the art will appreciate that various modifications, additions and substitutions are possible, and therefore, the scope of the present invention should not be limited to the above-described embodiments.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810619745.4ACN109032858A (en) | 2018-06-14 | 2018-06-14 | A kind of determination method and device of cloud computing test resource distribution |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810619745.4ACN109032858A (en) | 2018-06-14 | 2018-06-14 | A kind of determination method and device of cloud computing test resource distribution |
| Publication Number | Publication Date |
|---|---|
| CN109032858Atrue CN109032858A (en) | 2018-12-18 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810619745.4APendingCN109032858A (en) | 2018-06-14 | 2018-06-14 | A kind of determination method and device of cloud computing test resource distribution |
| Country | Link |
|---|---|
| CN (1) | CN109032858A (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114902273A (en)* | 2020-01-09 | 2022-08-12 | 支付宝实验室(新加坡)有限公司 | System and method for optimizing resource allocation using GPU |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011091056A1 (en)* | 2010-01-19 | 2011-07-28 | Servicemesh, Inc. | System and method for a cloud computing abstraction layer |
| CN103957261A (en)* | 2014-05-06 | 2014-07-30 | 湖南体运通信息技术有限公司 | Cloud computing resource distributing method based on energy consumption optimization |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011091056A1 (en)* | 2010-01-19 | 2011-07-28 | Servicemesh, Inc. | System and method for a cloud computing abstraction layer |
| CN103957261A (en)* | 2014-05-06 | 2014-07-30 | 湖南体运通信息技术有限公司 | Cloud computing resource distributing method based on energy consumption optimization |
| Title |
|---|
| 师雪霖 等: "云虚拟机资源分配的效用最大化模型", 《计算机学报》* |
| 罗杰等: "基于效益最大化的云虚拟机资源分配研究", 《计算机应用研究》* |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114902273A (en)* | 2020-01-09 | 2022-08-12 | 支付宝实验室(新加坡)有限公司 | System and method for optimizing resource allocation using GPU |
| CN114902273B (en)* | 2020-01-09 | 2025-07-18 | 支付宝(杭州)信息技术有限公司 | System and method for optimizing resource allocation using a GPU |
| Publication | Publication Date | Title |
|---|---|---|
| CN110851529B (en) | Computing power scheduling method and related equipment | |
| Mohan et al. | Edge-Fog cloud: A distributed cloud for Internet of Things computations | |
| Gan et al. | Genetic simulated annealing algorithm for task scheduling based on cloud computing environment | |
| Cheng et al. | Virtual network embedding through topology-aware node ranking | |
| CN113037877B (en) | Optimization method for time-space data and resource scheduling under cloud edge architecture | |
| WO2020258920A1 (en) | Network slice resource management method and apparatus | |
| CN112003660B (en) | Dimensional measurement method, computing power scheduling method and storage medium of in-network resources | |
| CN103778474A (en) | Resource load capacity prediction method, analysis prediction system and service operation monitoring system | |
| CN110650032A (en) | A method for building a QoS-based application optimization deployment scheme in a multi-cloud environment | |
| CN104270421A (en) | A multi-tenant cloud platform task scheduling method supporting bandwidth guarantee | |
| CN113438678B (en) | A method and device for allocating cloud resources for network slicing | |
| CN115002215B (en) | Resource allocation model training method and resource allocation method for government and enterprises on the cloud | |
| CN112491741B (en) | Virtual network resource allocation method and device and electronic equipment | |
| CN114186750A (en) | Routing planning method and device for hierarchical requirements of power business of smart power grid | |
| CN116149855A (en) | Medium performance resource cost optimization method and system under microservice architecture | |
| Cao et al. | Towards tenant demand-aware bandwidth allocation strategy in cloud datacenter | |
| CN105871618B (en) | More resource regulating methods of erratic demand are coped in cloud scheduler | |
| Guan et al. | Multidimensional resource fragmentation-aware virtual network embedding for IoT applications in MEC networks | |
| CN111611076A (en) | Fair distribution method for mobile edge computing shared resources under task deployment constraint | |
| CN102186182A (en) | Comprehensive network plan management system and method | |
| CN103825963B (en) | Virtual Service moving method | |
| CN114817846A (en) | Dynamic OD matrix estimation method and device under prior time characteristic constraint | |
| CN115016889A (en) | A virtual machine optimization scheduling method for cloud computing | |
| CN109032858A (en) | A kind of determination method and device of cloud computing test resource distribution | |
| Shehu et al. | Fruit fly optimization algorithm for network-aware web service composition in the cloud |
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | Application publication date:20181218 | |
| RJ01 | Rejection of invention patent application after publication |