Generative experiments for application deployment in 5G networks

By using neural network models to generate experiments in 5G networks, the logical-physical mapping of applications was identified, solving the challenge of network edge application location and achieving more efficient resource allocation and layout optimization.

CN117280677BActive Publication Date: 2026-11-17INTERNATIONAL BUSINESS MACHINE CORPORATION
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202280032557.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Priority Date
2021-05-20
Filing Date
2022-04-07
Publication Date
2026-11-17
Estimated Expiration
2042-04-07

AI Technical Summary

Technical Problem

Determining the optimal placement of applications or services in 5G networks presents challenges, especially in the selection of applications at the network edge. Manual trial and error leads to low resource efficiency, and the effectiveness of placing novel applications is unknown.

Method used

Experiments are generated using a neural network model. By identifying the logical-physical mapping of the application, the layout of the application to the physical nodes is determined. The layout selection is optimized using a cost function, and the optimal combination of experiments is found through reinforcement learning.

Benefits of technology

It improves the accuracy and efficiency of application deployment, can quickly identify candidate layout scenarios, reduce resource waste caused by manual trial and error, and optimize network resource allocation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117280677B_ABST
    Figure CN117280677B_ABST
Patent Text Reader

Abstract

A computer-implemented method for determining the placement of an application added to a network. The method includes determining a logical-physical mapping for the application; identifying subgraphs within the logical-physical mapping; locating applications stored in a directory of applications previously used on the network; the stored applications including one of the subgraphs within the logical-physical mapping; generating experiments using these subgraphs by a computing device running a neural network model; the experiments including deriving physical node mappings for the subgraphs; and determining a cost function associated with the placement of the application added to the network to one or more physical nodes in the derived physical node mapping.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This disclosure generally relates to network systems, and more specifically, to generative experiments for application deployment in 5G networks. Background Technology

[0002] Cloud computing networks face numerous challenges in determining the optimal placement of applications or services. With the increasing use of 5G standards and mobile edge computing (MEC) servers deployed at the network edge, the options available for placing applications or services in some cloud networks can include millions of choices. Operating in 5G means operating at higher frequencies. Higher frequencies can result in higher propagation losses due to many factors, including reflection, refraction, and diffraction, which can significantly impact the optimal placement of application components on the network.

[0003] Beyond the physical elements that can impact service elements on the network, there are additional challenges in finding the optimal placement for novel applications to the network. While some approaches may be able to address how the application or its microservices were previously placed, novel applications can have entirely different effects when placed on the network. Manual trial and error often leads to resource inefficiencies, which can have a significant impact on the entire network. Systems managing deployments need to improve how they determine application deployment. Summary of the Invention

[0004] According to embodiments of this disclosure, a method for determining the placement of an application added to a network is provided. The method includes determining a logical-physical mapping of the application added to the network; identifying one or more subgraphs in the logical-physical mapping of the application added to the network; locating an application previously stored in an application catalog used in the network; the stored application including one of the one or more subgraphs in the logical-physical mapping of the application added to the network; generating an experiment using the subgraphs in the one or more subgraphs by a computing device running a neural network model; the experiment including deriving a physical node mapping for one or more of the one or more subgraphs; and determining the layout of the application added to the network to one or more physical nodes in the derived physical node mapping.

[0005] In one embodiment, the method includes selecting the layout of an application to be added to the network to one or more physical nodes in a derived physical node map, based on determining the most cost-effective cost function associated with the layout. Understandably, this feature discloses the practical application of the method. This feature offers the benefit of improving the cost efficiency of the computational techniques involved.

[0006] According to embodiments of this disclosure, a computer program product is provided for determining the potential impact of a software upgrade on a computing device. The computer program product includes one or more computer-readable storage media and program instructions commonly stored on the one or more computer-readable storage media. The program instructions include determining a logical-physical mapping of an application to be added to a network; identifying one or more subgraphs in the logical-physical mapping of the application to be added to the network; locating applications stored in a catalog of applications previously used in the network; the stored applications including one of one or more subgraphs in the logical-physical mapping of the application to be added to the network; generating an experiment using the subgraphs in the one or more subgraphs by a computing device running a neural network model; the experiment including deriving a physical node mapping for one or more of the one or more subgraphs; and determining the layout of the application to be added to the network to one or more physical nodes in the derived physical node mapping.

[0007] According to one embodiment, the program instructions further include: repeating the following actions for a subgraph with the next lowest matching score: identifying the subgraph with the lowest matching score; generating the experiment; and determining the cost function until the most cost-effective layout for the application. This combination of features allows the subject technique to evaluate different candidates for previously assigned applications to find potentially best comparable layout scenarios. This feature also allows neural models to improve their ability to efficiently locate positions on the network for application layout.

[0008] According to an embodiment of the present invention, a computer server is provided. The computer server includes: a network connection; one or more computer-readable storage media; a processor coupled to the network connection and coupled to the one or more computer-readable storage media; and a computer program product including program instructions commonly stored on the one or more computer-readable storage media. The program instructions include determining a logical-physical mapping for an application to be added to the network; identifying one or more subgraphs in the logical-physical mapping of the application to be added to the network; locating applications stored in a directory of applications previously used in the network; the stored applications including one of the one or more subgraphs in the logical-physical mapping of the application to be added to the network; generating an experiment by running a neural network model on the computer server; the experiment using one of the one or more subgraphs; the experiment including deriving a physical node mapping for one or more of the one or more subgraphs; and determining the layout of the application to be added to the network to one or more physical nodes in the derived physical node mapping.

[0009] According to one embodiment, the program instructions for the computer server further include: determining a local maximum of a cost function based on user-input budget constraints for the steps of running the generative experiment. This feature allows the user to control the resources consumed in running the experiment, which might otherwise take an impractical amount of time to run across millions of possible combinations available for evaluation.

[0010] Using the methods and systems discussed herein, it will be recognized that application deployment improves with greater accuracy in the application's layout within the network. Examples are able to identify candidate layout scenarios more quickly through experimentation. The experiments run by examples are more effective by locating the next optimal experiment in a catalog that may contain subgraphs of microservices similar to the microservices in the application being laid out. Any constraints in the application can be considered during experimentation. Therefore, the neural network model running the experiments can learn from previous scenarios and determine cost-effective layouts.

[0011] The techniques described in this article can be implemented in several ways. Exemplary implementations are provided below with reference to the accompanying figures. Attached Figure Description

[0012] The accompanying drawings are illustrative embodiments. They do not show all embodiments. Other embodiments may be used alternatively or as an alternative. Details that may be obvious or unnecessary may be omitted to save space or for more efficient illustration. Some embodiments may be practiced with additional components or steps and / or without all components or steps shown. When the same number appears in different drawings, it refers to the same or similar parts or steps.

[0013] Figure 1 This is a block diagram of an architecture for application deployment in a 5G network according to an implementation method.

[0014] Figure 2 It is a block diagram mapping the logical diagram of a fog application's microservices according to some implementation methods.

[0015] Figure 3 It is a block diagram of the physical graph mapping of microservices mapped to fog nodes according to the implementation method.

[0016] Figure 4 This is a flowchart of a method for determining the optimal logical-physical mapping for a microservice application, according to an implementation method.

[0017] Figure 5 This is a flowchart of a method for determining the layout selection of microservice applications in a network by performing neural network-based generative experiments according to the implementation method.

[0018] Figure 6 It is a functional block diagram of a computer hardware platform with a specific configuration that can communicate with various networked components.

[0019] Figure 7 A cloud computing environment according to an illustrative implementation is described.

[0020] Figure 8 An abstract model layer according to an illustrative implementation is described. Detailed Implementation

[0021] Overview

[0022] In the following detailed description, numerous specific details are illustrated by way of example to provide a thorough understanding of the relevant teachings. However, it should be clear that these teachings can be practiced without such details. In other cases, well-known methods, processes, components, and / or circuits have been described at a relatively high level without detail to avoid unnecessarily obscuring aspects of these teachings.

[0023] This disclosure generally relates to systems and methods for determining how to deploy applications in a network. In exemplary embodiments, aspects of the subject matter can be used in networks based on 5G standards. Some of these networks may be cloud-based, as described more fully below. Typically, embodiments can be practiced in the fields of computers and computer networks.

[0024] In the following disclosure, embodiments may use neural networks to generate experiments that determine the logical mapping of an application's location on a network. Locations may include network edges, for example, in fog applications. As will be understood, various aspects of this subject matter offer improvements to computing techniques. While some descriptions involve running experiments, it should be understood that some of the described experimental procedures are run using neural network techniques that potentially handle millions of possible combinations (which may be impossible to perform manually). Furthermore, there are features that identify the closest subgraph in historical data, which are used unconventionally to generate the next optimal experiment. A notable consequence is that features of this technique enable more efficient deployment of applications and / or services in cloud computing networks, thereby improving the state of the computing subject matter.

[0025] Example Architecture

[0026] Figure 1 An exemplary architecture 100 for deploying applications in a network is shown. Architecture 100 includes a network 106 that allows different computing devices 102(1) to 102(N) to communicate with each other, and other elements connected to the network 106, such as an update data source 112, a deployment mapping server 116, and a cloud 120.

[0027] Network 106 may be, but is not limited to, a local area network (“LAN”), a virtual private network (“VPN”), a cellular network, the Internet, or a combination thereof. For example, network 106 may include a mobile network communicatively coupled to a private network, sometimes referred to as an intranet providing various ancillary services, such as communication with different app stores, libraries, and the Internet. Network 106 allows deployment engine 110 to communicate with application data source 112, computing devices 102(1) through 102(N), and cloud 120 to provide data processing, the deployment engine 110 being a software program running on deployment mapping server 116. Application data source 112 may provide data about applications and / or microservices, which will be processed under one or more of the techniques described herein. In one embodiment, application data is processed for deployment on cloud 120.

[0028] For purposes discussed later, several user devices are shown in the figures to represent some examples of computing devices that may be data sources analyzed according to the selected task. Aspects of symbol sequence data (e.g., 103(1) and 103(N)) can communicate via network 106 with deployment engine 110 of deployment mapping server 116. Today, user devices typically take the form of portable handheld devices, smartphones, tablet computers, personal digital assistants (PDAs), and smartwatches, although user devices can be implemented in other form factors, including consumer and business electronic devices.

[0029] For example, a computing device (e.g., 102(N)) may send a request 103(N) to the deployment engine 110 to identify where the layout of the application and / or microservice should be located in the cloud 120.

[0030] Although the application data source 112 and deployment engine 110 are shown as being on different platforms by way of example, it should be understood that in various embodiments, the application data source 112 and deployment mapping server 116 may be combined. In other embodiments, these computing platforms may be implemented by virtual computing devices in the form of virtual machines or software containers hosted in the cloud 120, thereby providing a resilient architecture for processing and storage.

[0031] See now Figure 2 and Figure 3 It shows how to map microservices of a fog application to fog nodes. Figure 2 The logic diagram 200 of the microservices 210-250 of the fog application is shown. Figure 3A physical graph 300 is shown, mapping microservices 210-250 to fog nodes 310-340. Microservices 210-250 may represent different computational services (e.g., 210 may represent a service that captures an image, 220 may represent a service that performs a discrete cosine transform on the image, 230 and 240 may represent services that perform feature finding, and 250 may represent a service that combines the results of services 210-240). When mapped to the physical graph 300, service nodes and logical edges in the logical graph 200 may be mapped to physical nodes in the physical graph 300. Edges may be mapped to single edges or multi-hub edges. While 220 and 230 have direct edges in the logical graph 200, they may have indirect edges in the physical graph 300. As is understandable, regarding physical resources and logical graphs, various aspects of this technology correlate computational actions and communications within the logical graph with the corresponding physical graph. Several layout constraints of nodes must be considered during mapping. For example, for security purposes, anti-association conditions may prevent microservices 210 and 220 from running on the same machine. Another constraint may relate to the power consumption or memory requirements of the microservices. Considering the numerous potential locations and variables involved, the possible combinations of application layouts can be millions. The optimal node layout can differ when users have different application compositions (e.g., applications running differently from previous application sets).

[0032] When microservice applications are to run at the 5G edge, this technology experiments to find the optimal mapping. The specific microservice architecture being run may have been previously invisible, and therefore its optimal placement is unknown. For example, the application may have new components. (Return to reference) Figure 2 Users may have already composed microservices 220, 230, and 240, but now microservices 230, 240, and 250 are being composed, resulting in different sets of nodes being composed. Since there may be an infinite number of combinations available, it's not surprising that the set of services might not match the set seen in the past for a given application. However, this technique utilizes the set of nodes that may have been seen in the past within the application.

[0033] For example, application S may be novel, but its subgraphs may be covered by different applications already studied in the catalog. Subgraphs could be, for example, Figure 2 The combination of microservices 210 and 220 shown, or Figure 2 The combinations shown are 210, 230, and 250.

[0034] In one aspect of this technology, it is proposed to employ a reinforcement learning-like approach to determine the next best experiment that can be performed. "Next best" refers to finding an experiment that includes a combination of subgraphs from directory S that best matches one or more subgraphs in the placed application. Experiments on the directory can be constructed using new applications, allowing the neural network to predict the performance of previously unseen microservice applications within its network.

[0035] Example Method

[0036] Now for reference Figure 4 This illustrates a method 400 for determining the optimal logical-physical mapping for a microservice application, according to an exemplary implementation. To implement method 400, the exemplary setup may assume working with a set of microservice applications S (e.g., from open-source or custom VNF / CNF chains, etc.) stored in a database. Method 400 can be implemented using a 5G testbed and workload generator that allows running microservice applications s (410) in a directory S (450) on a selected physical infrastructure 420 and a given logical-physical mapping. Given microservice applications s (410) that may not yet be in the set S of directory 450 and a target physical infrastructure (420) (which may not be visible), method 400 determines the optimal logical-physical mapping for the microservice application (410). When the microservice application s (410) is novel, its subgraph may be covered by different applications in directory S.

[0037] The processor can recommend the next best experiment (430) to run on the test bench to improve the logical-physical mapping of microservice applications (410). See below. Figure 5 Examples of generating experiments under this topic are discussed in more detail. Previous application key performance indicators (KPIs) measurements of a subset of the application's logical-physical mapping can be extracted from the S set across different physical infrastructures. Previous performance measurements can be stored in a database accessible to the system running the experiment. In some embodiments, the processor can iterate this process 440 with more experiments aimed at improving the logical-physical mapping. In some embodiments, if it is determined that the microservice application (410) is novel, the microservice application (410) can be added to the catalog 450.

[0038] Now for reference Figure 5 This illustrates a method 500 for performing neural network-based generative experiments to determine the placement selection of microservice applications in a network, according to an exemplary embodiment. For microservice nodes (nodes in a logical graph), method 500 learns neural embeddings based on the description of the microservice. For infrastructure nodes (nodes in a physical graph), method 500 learns neural embeddings based on canonical learning. The following elements can be trained from historical application KPIs:

[0039] MLP (Multilayer Perceptron) n (l n ,p n Logical node l n With physical node p n Matching score between; n and p n The embedding vector representations derived from their descriptions, and the MLP is a learned vector representation. n and p n A simple feedforward neural network for matching (between 0 and 1). The higher the matching score, the better. n With p n The better the match between them, and therefore the better the match for logical nodes l n Mapped to physical node p n That's good.

[0040] MLP F e (l n1 ,p n1 ,l n2 ,p n2 Edge connectivity (p) n1 On the l n1 ) and (p n2 On the l n2 The matching score between two nodes; extending the matching score between two nodes to the matching score of the link / edge. Note that p n1 Can be with p n2 Same or different. Similar to the above, each logical / physical node is represented as an embedding vector derived from its description, and ML learns to connect to p. n2 On the l n2 p n1 On the l n1 The match is between 0 and 1.

[0041] GCNN (Graphical Convolutional Neural Network) sg (l sg ,p sg Mapping): Matching scores for logical-physical subgraph mappings; extending the matching scores to subgraphs where two or more logical nodes are mapped to one or more physical nodes. Since MLPs may fail to capture graph structure, GCNNs (Graph Convolutional Neural Networks) can be used to capture the subgraph structure in both logical and physical graphs, as well as the mapping structure between them.

[0042] F g (l g ,p g Using the above cost function in l g China to pg Determine the placement of each node. Given matching functions at the levels of nodes, links, and subgraphs, it is possible to identify matches between nodes, links, and subgraphs in the application and previously measured values, and to estimate overall application KPIs.

[0043] Starting with a feasible logical-physical mapping 510, the method can identify 520 sub-graphs s with the lowest matching scores. g (Subgraphs include nodes and edges). Matching can be applied in the order of subgraph matching, then link matching, and then node matching. In some embodiments, the process can begin with the largest matching subgraph to evaluate the total score. If no large subgraph match is found, the process may attempt to find matching links. If no links are found, the process may review node-level matching scores. A 530s interval can be retrieved. g The derived logical subgraph in the table is sg(l). Applications in the catalog 540 with the subgraph closest to sg(l) can be identified. When determining the closest matching subgraph, both the logical and physical graphs can be directed graphs, and a match can be defined on both subgraphs as a score of nodes / links shared by both subgraphs. Using the identified closest matching subgraph, experiments 550 deriving the physical node mapping of sg(l) can be generated. In some embodiments, the method can re-evaluate 560 the KPIs used for the logical-physical subgraph mapping. 570 The process for determining the placement of Fg can be rerun based on the updated cost function. Some costs come from Fn, Fe, and Fsg, while others come directly from the lookup table of the experiment. The method can be repeated or looped from step 520 as needed. The constraints of the loop can be based on the budget provided to the system running the experiment. For example, the time budget can be based on the time spent performing multiple runs. The computational cost budget can be based on how much computational power is required to perform multiple runs.

[0044] Example computer platform

[0045] As discussed above, the interpretable modeling functionality disclosed in this subject matter can be implemented using one or more computing devices connected for data communication via wireless or wired communication, such as... Figure 1 As shown. Figure 6 This is a functional block diagram of a computer hardware platform capable of communicating with various networked components such as training input data sources and the cloud. Specifically, Figure 6 This shows what can be used to implement servers (such as...) Figure 1 The network or host computer platform 600 of the deployment mapping server 116).

[0046] The computer platform 600 may include a central processing unit (CPU) 604, a hard disk drive (HDD) 606, random access memory (RAM) and / or read-only memory (ROM) 608, a keyboard 610, a mouse 612, a display 614, and a communication interface 616 connected to a system bus 602.

[0047] In one embodiment, HDD 606 includes the ability to store programs (such as deployment engine 640) capable of performing various processes in the manner described herein. Generally, deployment engine 640 can be configured to analyze the anticipated stability of a computing device after a software upgrade, according to the embodiments described above. Deployment engine 640 may have various modules configured to perform different functions. In some embodiments, deployment engine 640 may include submodules. For example, application analyzer 642, physical logic graph analyzer 644, microservice analyzer 646, generation experiment engine 648, and service placement module 650.

[0048] In another embodiment, computer platform 500 may represent an end-user computer (e.g., computing devices 102(1) to 102(N)). In this context, different end-user computer platforms 600 may include a mixed configuration of hardware components and software packages.

[0049] Example cloud platform

[0050] As discussed above, features related to analyzing the impact of software upgrades on computing devices may include Cloud120 (see [link to Cloud120]). Figure 1 It should be understood that while this disclosure includes a detailed description of cloud computing, the implementation of the teachings cited herein is not limited to cloud computing environments. Rather, embodiments of this disclosure can be implemented in conjunction with any other type of computing environment now known or developed hereafter.

[0051] Cloud computing is a service delivery model that enables convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, network bandwidth, servers, processing, memory, storage, applications, virtual machines, and services), which can be rapidly provisioned and released with minimal management effort or interaction with the service provider. This cloud model may include at least five features, at least three service models, and at least four deployment models.

[0052] The features are as follows:

[0053] On-demand self-service: Cloud consumers can unilaterally and automatically provide computing power, such as server time and network storage, as needed, without requiring human interaction with the service provider.

[0054] Extensive network access: Capabilities are available through networks and accessed via standard mechanisms that facilitate the use of heterogeneous thin client or thick client platforms (e.g., mobile phones, laptops, and PDAs).

[0055] Resource pooling: A provider's computing resources are pooled to serve multiple consumers using a multi-tenant model, where different physical and virtual resources are dynamically assigned and reassigned as needed. There is a sense of location independence because consumers typically do not have control or knowledge of the exact location of the resources provided, but may be able to specify the location at a higher level of abstraction (e.g., country, state, or data center).

[0056] Rapid flexibility: The ability to provide capacity quickly and flexibly, automatically scaling down and up rapidly in some situations to scale up rapidly. For consumers, the available supply capacity often appears unlimited and can be purchased in any quantity at any time.

[0057] Measurable services: Cloud systems automatically control and optimize resource usage by leveraging metering capabilities at a level of abstraction appropriate to the service type (e.g., storage, processing, bandwidth, and active user accounts). Resource usage can be monitored, controlled, and reported, providing transparency to both service providers and consumers.

[0058] The service model is as follows:

[0059] Software as a Service (SaaS): This provides consumers with the ability to use the provider's applications running on cloud infrastructure. Applications can be accessed from different client devices via thin client interfaces such as web browsers (e.g., web-based email). Consumers do not manage or control the underlying cloud infrastructure, including the network, servers, operating system, storage, or even individual application capabilities, with possible exceptions such as limited user-specific application configuration settings.

[0060] Platform as a Service (PaaS): This provides consumers with the ability to deploy applications created or acquired by the consumer using programming languages ​​and tools supported by the provider onto cloud infrastructure. Consumers do not manage or control the underlying cloud infrastructure, including networks, servers, operating systems, or storage, but they have control over the deployed applications and the configuration of any application hosting environment.

[0061] Infrastructure as a Service (IaaS): The capabilities offered to consumers are processing, storage, networking, and other basic computing resources that enable consumers to deploy and run arbitrary software, which may include operating systems and applications. Consumers do not manage or control the underlying cloud infrastructure, but rather have control over the operating system, storage, deployed applications, and potentially limited control over selected networking components (e.g., host firewalls).

[0062] The deployment model is as follows:

[0063] Private cloud: A cloud infrastructure that operates solely for an organization. It can be managed by the organization or a third party and can exist on-site or off-site.

[0064] Community cloud: A cloud infrastructure shared by several organizations and supporting a specific community with shared concerns (e.g., tasks, security requirements, policies, and compliance considerations). It can be managed by an organization or a third party and can exist on-site or off-site.

[0065] Public cloud: Makes cloud infrastructure available to the public or large industry groups and is owned by an organization that sells cloud services.

[0066] Hybrid cloud: A cloud infrastructure is a combination of two or more clouds (private, community, or public) that remain a single entity but are bound together by standardized or proprietary technologies that enable data and applications to be ported (e.g., cloud bursting for load balancing between clouds).

[0067] Cloud computing environments are service-oriented, focusing on statelessness, loose coupling, modularity, and semantic interoperability. At the heart of cloud computing is the infrastructure comprising a network of interconnected nodes.

[0068] Now for reference Figure 7 This describes an illustrative cloud computing environment 700. As shown, the cloud computing environment 700 includes one or more cloud computing nodes 710 that can communicate with local computing devices used by cloud consumers, such as, for example, personal digital assistants (PDAs) or cellular phones 754A, desktop computers 754B, laptop computers 754C, and / or automotive computer systems 754N. The nodes 710 can communicate with each other. They can be physically or virtually grouped (not shown) in one or more networks, such as private clouds, community clouds, public clouds, or hybrid clouds, or combinations thereof, as described above. This allows the cloud computing environment 750 to provide infrastructure, platform, and / or software as a service without requiring cloud consumers to maintain resources on their local computing devices. It should be understood that... Figure 7 The types of computing devices 754A-N shown are intended to be illustrative only, and the computing node 810 and cloud computing environment 750 can communicate with any type of computerized device via any type of network and / or network-addressable connection (e.g., using a web browser).

[0069] See now Figure 8 This demonstrates the 750 (cloud computing environment) Figure 7 This provides a set of functional abstractions. It should be understood beforehand. Figure 8The components, layers, and functions shown are intended to be illustrative only, and embodiments of this disclosure are not limited thereto. As described, the following layers and corresponding functions are provided:

[0070] The hardware and software layer 860 includes hardware and software components. Examples of hardware components include: a mainframe 861; a server 862 based on a RISC (Reduced Instruction Set Computer) architecture; a server 863; a blade server 864; a storage device 865; and a network and network components 866. In some embodiments, the software components include network application server software 867 and database software 868.

[0071] The virtualization layer 870 provides an abstraction layer from which the following examples of virtual entities can be provided: virtual server 871; virtual storage 872; virtual network 873, including virtual private network; virtual application and operating system 874; and virtual client 875.

[0072] In one example, management layer 880 may provide the following functionalities: Resource Provisioning 881 Provides dynamic procurement of computing resources and other resources used to perform tasks within the cloud computing environment. Metering and Pricing 882 Provides cost tracking as resources are utilized within the cloud computing environment and bills or invoices for the consumption of these resources. In one example, these resources may include application software licenses. Security Provides authentication for cloud consumers and tasks, as well as protection for data and other resources. User Portal 883 Provides access to the cloud computing environment for consumers and system administrators. Service Level Management 884 Provides allocation and management of cloud computing resources to ensure that required service levels are met. Service Level Agreement (SLA) Planning and Fulfillment 985 Provides pre-scheduling and procurement of cloud computing resources, anticipating future requirements for those resources according to the SLA.

[0073] The workload layer 890 provides examples of functionalities that can be leveraged in a cloud computing environment. Examples of workloads and functionalities that can be provided from this layer include: mapping and navigation 891; software development and lifecycle management 892; virtual classroom education delivery 893; data analytics and processing 894; transaction processing 895; and deployment services 896, as discussed herein.

[0074] in conclusion

[0075] Various embodiments of this teaching have been described for illustrative purposes, but are not intended to be exhaustive or limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope of the described embodiments. The terminology used herein has been chosen to best explain the principles of the embodiments, their practical application, or technical improvements to technologies found in the market, or to enable those skilled in the art to understand the embodiments disclosed herein.

[0076] While the content considered to be the best state and / or other instances has been described above, it should be understood that various modifications may be made therein, and the subject matter disclosed herein can be implemented in different forms and instances, and the teachings can be applied to many applications, of which only some have been described herein. The appended claims are intended to claim protection for any and all applications, modifications, and variations falling within the true scope of this teaching.

[0077] The components, steps, features, purposes, benefits, and advantages discussed herein are illustrative only. They, and the discussions associated with them, are not intended to limit the scope of protection. While various advantages have been discussed herein, it will be understood that not all embodiments are necessarily required to include all advantages. Unless otherwise stated, all measurements, values, ratings, positions, amplitudes, sizes, and other specifications set forth in this specification (including in the following claims) are approximate and imprecise. They are intended to have a reasonable range of functionality associated with them and consistent with functionality customary in the art to which they pertain.

[0078] Many other embodiments are also contemplated. These include embodiments with fewer, additional, and / or different components, steps, features, purposes, benefits, and advantages. These also include embodiments in which components and / or steps are arranged and / or ordered differently.

[0079] This document describes aspects of the present disclosure with reference to call flow diagrams and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present disclosure. It should be understood that each step of the flowcharts and / or block diagrams, and combinations of boxes in the call flow diagrams and / or block diagrams, can be implemented by computer-readable program instructions.

[0080] These computer-readable program instructions may be provided to a processor of a computer, special-purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions / actions specified in one or more blocks of a call flow procedure and / or block diagram. These computer-readable program instructions may also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner, such that the computer-readable storage medium storing the instructions comprises an article of manufacture containing instructions that implement aspects of the functions / actions specified in one or more blocks of a call flow procedure and / or block diagram.

[0081] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable apparatus, or other device to produce a computer-implemented process, such that the instructions executed on the computer, other programmable apparatus, or other device perform the functions / actions specified in one or more blocks of a flowchart.

[0082] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. Each block in the calling flow or block diagram may represent a module, segment, or portion of instructions, including one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than indicated in the figures. For example, depending on the functions involved, two consecutively shown blocks may actually execute substantially simultaneously, or these blocks may sometimes execute in reverse order. It should also be noted that each block in the block diagrams and / or calling flow descriptions, and combinations of blocks in the block diagrams and / or calling flow descriptions, can be implemented using a dedicated hardware-based system that performs the specified function or action or executes a combination of dedicated hardware and computer instructions.

[0083] Although the foregoing has been described in conjunction with exemplary embodiments, it should be understood that the term "exemplary" means only as an example, and not the best or optimal. Nothing stated or shown beyond what is immediately stated above is intended or should be construed as causing any contribution or public equivalent of any component, step, feature, object, benefit, advantage, or advantage, whether or not it is stated in the claims.

[0084] It should be understood that the terms and expressions used herein have their general meanings as assigned to their respective corresponding queries and fields of study, unless otherwise specified herein. Relational terms such as "first" and "second" may be used merely to distinguish one entity or action from another, without necessarily requiring or implying any actual such relationship or order between these entities or actions. The terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but may also include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element beginning with "an" or "a" does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes that element.

[0085] An abstract of this disclosure is provided to allow the reader to quickly determine the nature of this technical disclosure. It is submitted under the understanding that it is not intended to interpret or limit the scope or meaning of the claims. Furthermore, as can be seen from the above detailed description, various features are combined in various embodiments for the purpose of simplification. The method of this disclosure should not be construed as reflecting an intention to have more features than expressly recited in each claim of the claimed embodiments. Rather, as reflected in the following claims, the inventive subject matter lies in all features of fewer than a single disclosed embodiment. Therefore, the following claims are hereby incorporated into the detailed description, wherein each claim is independently claimed as a separate subject matter.

Claims

1. A computer implementation method for determining the placement of an application added to a network, comprising: Determine the logical-physical mapping of applications added to the network; Identify one or more subgraphs in the logical-physical mapping of applications added to the network; From the one or more subgraphs, iteratively identify the subgraph with the lowest matching score between the embedding vector of the logical node and the embedding vector of the physical node; For each iteration: Locate the stored application in the directory of applications previously used in the network, wherein the stored application includes one or more subgraphs in the logical-physical mapping of applications added to the network, and the location of the stored application is based on the subgraph with the lowest matching score; and An experiment is generated using one of the one or more subgraphs and executed by a computing device running a neural network model, wherein the experiment includes deriving a physical node mapping of one of the one or more subgraphs; and Based on determining the most cost-effective cost function associated with placement, the application added to the network is deployed to one or more physical nodes in the exported physical node map.

2. The method according to claim 1, further comprising: Iteration: Identify the action of the subgraph with the lowest matching score for the subgraph with the next lowest matching score; The action that generates the experiment; as well as The action of deploying the application to one or more physical nodes continues until the most cost-effective placement of the application is determined.

3. The method of claim 2, further comprising determining the local maximum of the cost function based on user input budget constraints for actions used to run the generation experiment.

4. The method according to claim 1, further comprising: Generate a logic-physics subgraph mapping based on the generated experiments; as well as Re-evaluate the key performance metrics of the logical-physical subgraph mapping.

5. The method according to claim 1, wherein, The experiment considers conditions imposed on the application, including anti-affinity constraints and / or memory conditions associated with the application.

6. A computer program product for determining the placement of an application added to a network, the computer program product comprising: Program instructions, the program instructions including: Determine the logical-physical mapping of applications added to the network; Identify one or more subgraphs in the logical-physical mapping of applications added to the network; From the one or more subgraphs, iteratively identify the subgraph with the lowest matching score between the embedding vector of the logical node and the embedding vector of the physical node; For each iteration: Locate the stored application in the directory of applications previously used in the network, wherein the stored application includes one or more subgraphs in the logical-physical mapping of applications added to the network, and the location of the stored application is based on the subgraph with the lowest matching score; and An experiment is generated using one of the one or more subgraphs and executed by a computing device running a neural network model, wherein the experiment includes deriving a physical node mapping of one of the one or more subgraphs; and Based on determining the most cost-effective cost function associated with placement, the application added to the network is deployed to one or more physical nodes in the exported physical node map.

7. The computer program product according to claim 6, wherein, The program instructions further include: Iteration: Identify the action of the subgraph with the lowest matching score for the subgraph with the next lowest matching score; The actions that generate the experiment; and The action of deploying the application to one or more physical nodes continues until the most cost-effective placement of the application is determined.

8. The computer program product according to claim 7, wherein, The program instructions further include determining the local maximum of the cost function based on user-input budget constraints for the actions used to run the generation experiment.

9. The computer program product according to claim 6, wherein, The program instructions further include: Generate a logic-physics subgraph mapping based on the generated experiments; and Re-evaluate the key performance metrics of the logical-physical subgraph mapping.

10. The computer program product according to claim 6, wherein, The experiment considers conditions imposed on the application, including anti-affinity constraints and / or memory conditions associated with the application.

11. An application deployment computer server, comprising: Network connectivity; One or more computer-readable storage media; A processor, coupled to the network connection and coupled to the one or more computer-readable storage media; as well as A computer program product, comprising program instructions commonly stored on the one or more computer-readable storage media, the program instructions including: Determine the logical-physical mapping of applications added to the network; Identify one or more subgraphs in the logical-physical mapping of applications added to the network; From the one or more subgraphs, iteratively identify the subgraph with the lowest matching score between the embedding vector of the logical node and the embedding vector of the physical node; For each iteration: Locate the stored application in the directory of applications previously used in the network, wherein the stored application includes one or more subgraphs in the logical-physical mapping of applications added to the network, and the location of the stored application is based on the subgraph with the lowest matching score; and An experiment is generated using one of the one or more subgraphs and executed by a computing device running a neural network model, wherein the experiment includes deriving a physical node mapping of one of the one or more subgraphs; and Based on determining the most cost-effective cost function associated with placement, the application added to the network is deployed to one or more physical nodes in the exported physical node map.

12. The computer server according to claim 11, wherein, The program instructions further include: Iteration: Identify the action of the subgraph with the lowest matching score for the subgraph with the next lowest matching score; The actions that generate the experiment; and The action of deploying the application to one or more physical nodes continues until the most cost-effective placement of the application is determined.

13. The computer server according to claim 12, wherein, The program instructions further include determining the local maximum of the cost function based on user-input budget constraints for the actions used to run the generation experiment.

14. The computer server according to claim 11, wherein, The program instructions further include: Generate a logic-physics subgraph mapping based on the generated experiments; and Re-evaluate the key performance metrics of the logical-physical subgraph mapping.

Citation Information

Patent Citations

  • Virtual network mapping method and device based on multi-objective optimization

    CN111385151A

  • Artificial intelligence-powered diagram-to-deployment converter

    US20210075887A1