WO2020017843A1 - 클라우드 플랫폼에서의 클러스터 리소스 할당 및 관리 방법 - Google Patents
클라우드 플랫폼에서의 클러스터 리소스 할당 및 관리 방법 Download PDFInfo
- Publication number
- WO2020017843A1 WO2020017843A1 PCT/KR2019/008697 KR2019008697W WO2020017843A1 WO 2020017843 A1 WO2020017843 A1 WO 2020017843A1 KR 2019008697 W KR2019008697 W KR 2019008697W WO 2020017843 A1 WO2020017843 A1 WO 2020017843A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- cluster
- resource
- cloud
- cloud platform
- application
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5083—Techniques for rebalancing the load in a distributed system
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
- G06F9/5044—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering hardware capabilities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5061—Partitioning or combining of resources
- G06F9/5072—Grid computing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5061—Partitioning or combining of resources
- G06F9/5077—Logical partitioning of resources; Management or configuration of virtualized resources
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2209/00—Indexing scheme relating to G06F9/00
- G06F2209/50—Indexing scheme relating to G06F9/50
- G06F2209/503—Resource availability
Definitions
- the present invention relates to a method for allocating and managing cluster resources in a cloud platform, and more particularly, to a method for allocating and managing cluster resources in a cloud platform that can efficiently allocate and manage cluster resources.
- the cloud is referred to as the 'service provider's server' according to the practice of displaying the computing service provider server in a cloud shape.
- SoaaS Software as a Service
- AWS RDS Google AppEngine
- Google AppEngine application services that are provided on-demand to many users, such as Salesforce.com and Google e-mail.
- IaaS Infrastructure as a Service
- PaaS Platform as a Service
- AWS EC2 AWS EC2.
- the cloud is a private cloud that operates only for one organization depending on the type of introduction and distribution, a public cloud rendered through an open network for public use, and two that remain distinct but tied together. It can also be divided into a hybrid cloud (hybrid cloud) that is a combination of the above clouds.
- an object of the present invention is to provide a cluster resource allocation and management method in a cloud platform capable of efficiently allocating and managing cluster resources.
- Cluster resource allocation and management method in a cloud platform if the cluster resource allocation request is input, the cloud platform system generating service group information; Adding / deleting a user of the service group by the cloud platform system; Selecting a cluster to be allocated to a user of the service group and inputting allocation information; Querying the cluster resource capacity by the cloud platform system; If the resource availability inquiry result resource is available, the cloud platform system registering resource allocation information with the service group and allocating resources to complete cluster resource allocation; When the resource availability inquiry results in a lack of resources, checking, by the cloud platform system, whether a cluster resource is added; If the cluster resource can be added, additionally registering the cluster resource by the cloud platform system; Registering, by the cloud platform system, resource allocation information with the service group and allocating resources to complete cluster resource allocation; And if the cluster resource is not addable, reselecting an available cluster.
- Terminating an application for returning the cluster resources Recovering allocated resources by the cloud platform system; Updating, by the cloud platform system, cluster available resource information; And returning the cluster resource to be completed.
- the cluster resource allocation and management method in the cloud platform according to the present invention has an effect of efficiently allocating and managing cluster resources.
- FIG. 1 is a block diagram of a cloud platform system according to an embodiment of the present invention.
- FIG. 2 briefly illustrates the function of the cloud integrator of FIG. 1.
- FIG. 3 briefly illustrates the function of the service manager of FIG. 1.
- FIG. 4 briefly illustrates the function of the application orchestration unit of FIG. 1.
- FIG. 5 illustrates a framework of application containerization according to one embodiment of the invention.
- FIG. 12 illustrates an architecture of a cloud platform system according to an embodiment of the invention.
- FIG 13 shows the configuration of the cocktail server and its surrounding architecture.
- FIGS. 14 to 16 are diagrams for explaining a multi-cluster provisioning and management function of a cloud platform system according to an embodiment of the present invention.
- FIG. 17 is a flowchart illustrating a multi-cluster provisioning and management method of a cloud platform system according to an embodiment of the present invention.
- FIG. 18 is a diagram illustrating a cluster resource allocation management function of a cloud platform system according to an embodiment of the present invention.
- FIG. 19 is a flowchart illustrating a cluster resource allocation method of a cloud platform system according to an embodiment of the present invention.
- FIG. 20 is a flowchart illustrating a cluster resource usage method of a cloud platform system according to an embodiment of the present invention.
- 21 is a flowchart illustrating a cluster resource return method of a cloud platform system according to an exemplary embodiment of the present invention.
- FIG. 1 is a block diagram of a cloud platform system according to an exemplary embodiment of the present invention
- FIG. 2 schematically illustrates the function of the cloud integrator of FIG. 1
- FIG. 3 briefly illustrates the function of the service manager of FIG. 1.
- 4 schematically illustrates the function of the application orchestration unit of FIG. 1.
- FIG. 5 illustrates a framework of application containerization according to an embodiment of the present invention
- FIGS. 6 to 11 briefly illustrate functions of the development / operation unit of FIG. 1.
- the cloud platform system of FIG. 1 provides a view and a tool for ensuring application availability and scalability and streamlining development and operation based on multi / hybrid cloud integrated management.
- the cloud platform system of the present invention will be referred to as " Cocktail Cloud ".
- the cocktail cloud includes a cloud integration unit (100), a service management unit (Service Management) 110, an application orchestration unit (Orchestration 120), a development / operation unit (DevOps View, 140), and a DB / repository. And 150.
- the cloud integration unit 100 automatically configures the infrastructure of the multi / hybrid cloud to provide the application and synchronize configuration information for management.
- the cloud integrator 100 performs the functions of cloud provisioning and cloud synchronization.
- the cloud provisioning function is a function of configuring and providing a cloud network infrastructure in an application cluster (cocktail cluster), and configuring and providing a cloud computing infrastructure in an application. And for physical infrastructure (bare metal), cluster configuration tool is provided.
- Support Cloud is AWS, Azure, Aliyun, Google Computing Engine for Public, Openstack, VMWear for Private, and On-premise, Datacenter BareMetal Infra.
- the cloud synchronization function is a function of storing and managing cloud infrastructure configuration information in the integrated configuration DB 160 and synchronizing infrastructure change information with the integrated configuration DB 160 during operation.
- the service management unit 110 is a logical group managing an application cluster, and allocates and manages cloud accounts, users, and network resources. In other words, the service manager 110 performs an integrated account management function, a network management function, and a user management function.
- an integrated account management (Cloud Provider) function is a function used to collectively manage multi-cloud accounts and access information, and to configure a network and cloud provisioning.
- Network management is the ability to configure cloud networks and assign them to services.
- it may be a VPC Subnet of AWS.
- One service creates a cluster using a multi-cloud provider's network to configure and operate an application.
- User management function is to manage the team members who manage the service and the authority required for development / operation.
- the authority may include an enterprise service management authority (Admin), an enterprise service inquiry authority (Manager), and a service management authority (DevOps) assigned as a member. Users can participate as members in various services.
- Admin enterprise service management authority
- Manager enterprise service inquiry authority
- DevOps service management authority assigned as a member. Users can participate as members in various services.
- the Application Orchestration Department (120) is responsible for the core functionality of the Cocktail Cluster, with the ability to ensure application deployment, availability, and scalability.
- the application orchestration unit 120 performs an application deployment function, a replication control function, a rolling update function, a scaling function, and a monitoring function.
- the application distribution function is a container image-based distribution that provides ease of requiring no separate setting and configuration, and automatically provisions a cloud infrastructure when the application is distributed.
- an application container (hereinafter referred to as a " container ”) refers to an independent system on an OS virtualized by allocating and isolating host resources to an application process.
- cgroup creates a process group and allocates and manages resources to allocate host resources to processes on the OS.
- a namespace is a technique that isolates a process, network mount, etc. into a specific name space.
- a container is an independent system virtualized on an OS that allocates resources to an application process through cgroups and is isolated with a namespace.
- Container is a lightweight OS virtualization method that does not use a hardware emulator and guest OS. It is a technology suitable for application virtualization because it consumes little host resources and requires little startup time. In addition, virtualization on the OS enables configuration and deployment of infrastructures independent of existing physical servers (bare metals) and virtual servers (virtual machines).
- Converting an existing application to a container requires switching between the configuration of the application and the configuration, rather than the source.
- workload-specific role-specific configurations are common, and multiplexing and scaling through replication Consideration should be given to the design and application of the configuration.
- a cluster-oriented infrastructure for container orchestration must be configured, and computing capacity considering replication and scaling needs to be estimated (minimizing reserve capacity and easily scaleable if necessary), and related to shared storage, security, network, etc. You will need to configure your infrastructure.
- containerization is largely divided into analysis and configuration design (S100), container switching (S200), operation transfer (S300).
- the container switching target is selected from existing applications in consideration of the purpose and strategy of container / cloud introduction (S110).
- the target application When the target application is selected, the target application is analyzed (S120). At this time, application status and data survey of application, infrastructure, data, and linkage structure are conducted, and the needs of development, operation, and manager are collected. Then, the direction, issues, and solutions of container composition are drawn.
- the container configuration for each target application is designed in consideration of separation / integration, linkage, availability, scalability, and security (S130). At this point, you can define image build templates such as base images, environment variables, inclusion items, and commands.
- S140 design the infrastructure configuration (S140). Select a transition infrastructure (cloud / bare metal) provider and calculate capacity per application container. It calculates the number of container cluster nodes and infrastructure capacity, and designs storage, network, and security configurations.
- the infrastructure configuration is designed to establish a container switching method (S150).
- S150 container switching method
- the detailed conversion plan for each application is established, the transition task and organization / role are defined, and the transition schedule is established. And reflect reporting and feedback.
- the cocktail cloud platform is installed and configured, and the infrastructure, such as network, shared storage, and security, is configured (provided by cocktail in the case of cloud). Create a cocktail service and cluster and validate the cluster configuration through infrastructure infrastructure assignment and user registration.
- an application container is configured for application switching (S230), and if necessary, the application setting and source are changed. Verifies the function and setting of the switching container, and builds the container deployment image and registers it in the registry. Then create and test the cocktail server.
- Switch the target application container for data conversion (S240), and set the cocktail server through the Persistence volume setting, and extracts data and transmits to the cocktail server.
- this model DB solution data conversion is performed and data consistency is checked.
- data synchronization solutions are applied to minimize downtime.
- the verified container is distributed to the cocktail server, the application function and performance tests are performed, and the test results are reflected in the container and the infrastructure (S250 and S260).
- Operational distribution / opening (S310) is performed for the operation transfer (S300). Specifically, an operation cocktail cluster is created and a cocktail server is generated and linked based on the converted image. Then migrate the operational data and open the application.
- container orchestration The technology for distributing, operating, and managing such application containers is called container orchestration.
- Container orchestration is a technology that deploys, operates, and manages application containers by forming managed clusters in physical and virtual infrastructures, and utilizes the advantages of light, fast mobility and mobility of containers to cloud existing on-premises and data center infrastructure. And application management platforms in private and public clouds.
- the container has the following advantages.
- containers implement lightweight virtualization.
- OS-level virtualization (Non Hypervisor) is possible, fast operation (creation, execution, restart, etc.), and small sized container image is efficient to deploy and update.
- the container is mobile.
- the replication function is faster and more efficient than the OS reboot method by maintaining the initial number of replications (multiplexing) for application stability and availability, and restarting in case of abnormality through the application container health check.
- the cloned application is serviced through load balancing.
- Rolling update function performs update operation such as deployment and infrastructure change without interruption of application service, and configures automation through job management function of DevOps View when there is dependency between several applications.
- Scaling function is to scale up / out of instance through monitoring of application, and to scale up / down of resource capacity in case of application infrastructure.
- the monitoring information then configures the scaling automation.
- the monitoring function monitors an application instance (container + infrastructure) and generates and manages alarms through threshold setting.
- DevOps View includes service status function, cluster map function, monitoring view function, resource management function, metering function, task management function, and enterprise status management / analysis function. Each function will be described with reference to FIGS. 6 to 11 as follows.
- the service status function provides a view (see FIG. 6) that provides a service-oriented view of the status of the entire application cluster of the cocktail cloud. Accordingly, items such as service status, cluster status, and monitoring alarm may be displayed.
- the cluster refers to an application unit and a service refers to a logical group of the cluster.
- the supplier, region, server, cloud component, and monthly usage costs of the cluster can be viewed in the form of a card.
- the usage costs can be excluded.
- the cluster card In the monitoring alarm display function, if an alarm occurs in an application or infrastructure in a cluster, the cluster card can be checked.
- the cluster map function provides a view for visualizing and managing the configuration and state information of the cocktail server (application) in a map form (see FIG. 7).
- the cluster map improves the visibility of configuration information by inquiring and managing the server and cloud component configuration of the cluster in a map form.
- the cluster map may include items such as a cocktail server, a cloud component, and a server group.
- Cocktail Server is a basic unit of application orchestration, consisting of load balancing, application containers, and infrastructure, providing a standardized interface for multi / hybrid cloud management.
- the cocktail server checks application status, replication, resource usage, and manages scaling and rolling updates in the server.
- Cocktail servers are divided into multi- and single-instance types depending on whether they have replication capabilities or not.
- AWS supports multizone options.
- Cloud components manage PaaS services provided by providers.
- it can be RDS, a DB service of AWS.
- Server groups provide administrative convenience for logical groups of server configurations.
- the monitoring view function checks the resource capacity and status of applications and infrastructure in the cluster and provides information for checking the status of cloud resources (see FIG. 8).
- the monitoring view visualizes and provides monitoring information about the applications and infrastructure in the cluster, and provides the average and TOP information of CPU, memory, and disk so that resource usage can be checked and responded to in operation.
- the monitoring view may include a view switch (trend / data) item, a target switch (server / resource) item, and the like.
- Trend View provides hourly monitoring information about servers and replicated instance and application containers
- Data View provides average and TOP monitoring values of the current time.
- the monitored targets are divided into servers in the cluster and resources in the cloud infrastructure.
- Cloud resources use information provided by providers.
- the resource management function provides a view (hereinafter referred to as a " resource management view") for identifying resources of the cloud infrastructure constituting the application and adjusting detailed settings as necessary (see FIG. 9).
- the resource management view allows you to view the cloud infrastructure resources that make up the cocktail server and to change settings in detail.
- the Cocktail Server automatically performs the basic configuration for application orchestration, but is used when you need to adjust cloud resources yourself if necessary.
- the resource management view includes resource information / action items.
- the application manages container setting and distribution information.
- Cloud resource information is composed of load balancer, instance (VM) and security, and instance manages capacity and volume. Resource information that needs to be adjusted is performed through actions.
- the metering function provides a view (hereinafter referred to as a "metering view ”) that allows you to check the cost information of the cloud infrastructure resources used by the application (see FIG. 10).
- the metering view may include cluster infrastructure usage cost items, server and resource cost items, and the like.
- the cost category by server and resource provides the cost of cloud resources used by each cocktail server based on the TOP, and the cost of using cloud resources by type based on the TOP.
- the job management function provides an administrative view (hereinafter referred to as " job management view") for scheduling / automating operational tasks such as distribution, remote command, resource management, and the like (see FIG. 11).
- job management view an administrative view for scheduling / automating operational tasks such as distribution, remote command, resource management, and the like (see FIG. 11).
- Job management views provide scheduling and batch processing for the operation of applications and infrastructure.
- the work management view may include a work status item, a work management item, and the like.
- the task status items in the task management view are divided into distribution, remote command, and resource management tasks, and are composed by combining each task.
- Deployment refers to application deployment, remote commands to perform OS commands remotely, and resource management to scale and state / configuration changes.
- work management items can be set up according to immediate execution, scheduling, and alarm occurrence.
- Execution according to the alarm occurrence is used, for example, automatic scaling according to the capacity monitoring standard.
- the task management section provides a check of the execution status and logs of the task.
- Enterprise Status Management / Analysis provides a Cocktail Dashboard for identifying and analyzing enterprise application, cloud and cost status.
- the Cocktail Dashboard is a view of the application and cloud infrastructure at the enterprise level, providing cost / budget management, cost optimization analysis, and statistical reports.
- the cocktail dashboard may include application status items, cloud status items, cost / budget management, cost optimization analysis items, and statistics / report items.
- the application and infrastructure status can be identified and viewed company-wide based on the standardized elements of the cocktail server, cluster, and cloud component, and the service-oriented status view is provided.
- the cloud used by the company can be identified by provider, region, and resource, and the infrastructure-oriented status view is provided.
- the company can identify the enterprise cloud cost status and provide the information to make cloud resource cost effective through budget allocation / control and optimization analysis by service.
- Statistics / Report item provides statistical information and report view for analysis and reporting.
- Image storage (registry) 180 in the DB / repository 150 manages the registration, sharing, download, search, version of the application container
- monitoring DB 170 manages the monitoring information of the application and infrastructure
- the DB Configuration Management DB, CMDB, 160
- FIG. 12 illustrates an architecture of a cloud platform according to an exemplary embodiment of the present invention
- FIG. 13 illustrates a configuration of a cocktail server and its surrounding architecture.
- the cocktail cloud includes a cocktail cluster 200, a provider plug-in 210, a server manager 220, a DevOps manager, a CMDB 160, a monitoring DB 170, an image registry 180, an API server ( 290, a user console 300.
- the cocktail cluster 200 provides an orchestration-based architecture and the provider plug-in 210 is used as a basic module for integrated management through the cloud provider API 280.
- the cluster 200 is composed of a node and a master.
- the cluster 200 processes a command of a master through a worker 310.
- the worker 310 is in charge of communication with the master and is supported by the executor according to the execution instruction.
- the monitoring executor 320 collects node and container monitoring information, and the command executor 330 executes an OS and a container command.
- Container Engine Docker, 340.
- the provider plug-in 210 is an API Rapper for Kubernetes API support for multi-cloud and bare metal, and is composed of a plug-in module for provider extension.
- the cocktail server is a basic unit of application orchestration, and performs replication, scaling, and rolling updates of containers and cloud infrastructure through the cluster master 200 and the provider plug-in 210.
- the cocktail server is composed of a container and a cloud infrastructure as shown in FIG. 13, and is composed of a load balancer, an instance (node), a container, a volume, security, and the like, and examples of AWS include ELB, EC2 Instance, and Security Group. It may be an ESB.
- Cocktail Server provides cloud components for PaaS of cloud providers. For example, it can be RDS from AWS.
- the server manager 220 is a control module that orchestrates application containers and infrastructure in a server, and performs replication control for restarting / recovering abnormally terminated containers, scaling in / out, and scaling up and down through instance types and volume expansion. It provides a rolling update function that continuously and nondisruptively distributes application containers.
- DevOps Manager includes configuration management for provisioning multi-cloud infrastructure (Configuration Manager, 230), metering management (Metering Manager, 240) for multi-cloud resource usage and cost management, and resource management for multi-cloud resource status and configuration management. Manager, 250), monitoring and management for collecting and managing container / infrastructure monitoring information (Monitoring Manager, 260), and a combination of several task tasks for immediate execution, execution time, and event occurrence. It is a manager module for DevOps that provides job management (Job Manager, 270) for remote command tasks.
- Cocktail Cloud provides a DB for managing configuration information, monitoring information management, and application container image management of applications and infrastructure, and provides an interface for users and programming.
- the CMDB 160 manages configuration information of provider networks, services, clusters, servers, components, and cloud resources.
- the monitoring DB 170 manages monitoring information of applications and infrastructure.
- the image registry 180 manages registration, sharing, download, search, and version of application containers.
- the API server 290 provides all the functions of the cocktail cloud to the API 280, and supports customization according to the corporate strategy and linkage with other solutions.
- the user console 300 is provided in the form of a Web GUI.
- This cocktail cloud can be utilized as follows.
- Cocktail Cloud is a platform for integrated management of heterogeneous and complex multicloud environments through standardized components. It also implements the entire application-oriented enterprise cloud. Specifically, Cocktail Cloud is a standardized management component that standardizes managed objects through providers, networks, services, clusters, servers, and cloud components, and integrates and manages heterogeneous and complex multi-cloud resources (integrated accounts, resources, and costs). In addition, applications are a key resource of the business. Cocktail clusters can be used to increase application availability and scalability, and cocktail-driven devOps View can streamline development / operational tasks to enable an application-centric enterprise cloud.
- Cocktail Cloud provides the foundation for building / operating hybrid cloud through cloudization of in-house and data center bare metal infrastructure. It also provides integrated management and efficient development / operation of complex hybrid infrastructures.
- application clusters are built in-house and in the data center bare metal infrastructure to create a container-based cloud environment, eliminating the need for a platform for virtualization, providing scalability such as availability and scaling, and integrating existing private and public clouds. You can implement cloudization of a manageable physical infrastructure.
- Cocktail Cloud DevOps View It is also managed through standard components of the Cocktail Cloud and provides streamlined development and operation tasks through the Cocktail Cloud DevOps View.
- Cocktail Cloud provides an efficient management of applications on the cloud and a microservice construction and operation platform through automation for containers and CI / CD.
- Cocktail Cluster provides container-based application deployment and management environment (cloud-native applications) in the cloud infrastructure. Cocktail clusters are the basic units for building and managing microservices.
- Task management in the Cocktail DevOps view provides an automation foundation for building and deploying applications, and containers are a lighter and easier way to perform CI / CD.
- Cocktail Cloud provides a platform for deploying / operating applications on multi / hybrid clouds.
- Cocktail Cloud can also be used as an infrastructure resale and service delivery platform for cloud service brokers.
- CSB Build and operate a platform for CSB as a cocktail cloud that integrates and manages public cloud data center infrastructure and provides users with resale and cloud management platforms as a service, and provides multi-tenancy and billing systems for SaaS. Can be used as an affiliate cloud delivery and management platform.
- PaaS fire cloud components
- FIGS. 14 to 16 are diagrams for explaining a multi-cluster provisioning and management function of a cloud platform system according to an embodiment of the present invention.
- Cocktail Cloud a cloud platform system according to the present invention, is provided with a multi-cluster provisioning and management function that automatically creates a cluster environment in which container-based applications can operate in various infrastructures such as bare metal, cloud platform, and public cloud.
- This feature creates a container application operating environment by remotely provisioning multiple clusters centrally (cocktail clouds) in a multi-cloud environment, and improves operational efficiency by managing version upgrades of clusters remotely (see FIG. 14).
- FIG. 16 illustrates an example screen for registering an account information of a public cloud in which a cluster is configured to remotely control a cluster to perform application distribution, operation management, and cluster monitoring.
- FIG. 17 is a flowchart illustrating a multi-cluster provisioning and management method of a cloud platform system according to an embodiment of the present invention.
- the cloud platform system When a user's cluster provisioning is requested by the provisioning and management tool (S400), the cloud platform system enables input of cluster type information (eg, bare metal, public cloud, cloud platform, etc.) (S410).
- cluster type information eg, bare metal, public cloud, cloud platform, etc.
- cluster configuration information is generated (S420).
- the cluster configuration information includes at least one of an instance number, an instance specification (GPU, a memory type), network configuration information, and storage configuration information.
- the cloud platform system checks whether to create or change a cluster (S440, S490).
- the cloud platform system requests and configures the creation of instances, networks, and storage (S450). Thereafter, the container runtime software is installed (S460). Cluster configuration information is set (S470). Then cluster provisioning is completed (S480).
- the cloud platform system checks the cluster configuration history information and updates the cluster configuration (S500, S510). Then, the cluster change is completed (S520).
- the cloud platform system according to an exemplary embodiment of the present invention is developed, tested, and operated by a service or an organization by an administrator, and provides a cluster resource allocation management function for allocating a cluster infrastructure according to resource capacity of each cluster.
- FIG. 18 is a diagram illustrating a cluster resource allocation management function of a cloud platform system according to an embodiment of the present invention.
- the cloud platform system allocates a cluster to an assigned user when a service / organization use group is registered and a user belonging to a service / organization is assigned. At this time, the cloud platform system queries and allocates available cluster resources, and the amount of cluster resources for each service / organization is limited and managed.
- FIG. 19 is a flowchart illustrating a cluster resource allocation method of a cloud platform system according to an embodiment of the present invention.
- the cluster to be allocated to the next service group is selected and allocation information (for example, CPU and memory resource allocation) is input (S430).
- allocation information for example, CPU and memory resource allocation
- the cloud platform system inquires the cluster resource capacity (S440).
- the cloud platform system registers the resource allocation information in the service group and allocates the resources (S450, S460).
- the cluster resource allocation is completed (S470).
- the cloud platform system checks whether cluster resources are added (S480). If the cluster resource can be added, the cloud platform system additionally registers the cluster resource (S490). Thereafter, the cloud platform system registers resource allocation information in the service group and allocates resources to complete cluster resource allocation (S450, S460, and S470).
- step S430 If the cluster resource is not addable, the process returns to step S430 to reselect an available cluster.
- FIG. 20 is a flowchart illustrating a cluster resource usage method of a cloud platform system according to an embodiment of the present invention.
- the cloud platform system inquires the cluster resource capacity (S520).
- the application resource is allocated and activated (S530). At this time, application resource usage status monitoring and control is performed.
- the remaining resource allocation information is updated (S540). Then, the use of the cluster resource is completed (S550).
- 21 is a flowchart illustrating a cluster resource return method of a cloud platform system according to an exemplary embodiment of the present invention.
- the application is terminated to return the cluster resources (S560 and S570).
- the cloud platform system reclaims allocated resources (S580) and updates cluster available resource information (S590). Then, the cluster resource return is completed (S600).
- the above-described embodiments of the present invention can be written as a program that can be executed in a computer, and can be implemented in a general-purpose digital computer that operates the program using a computer-readable recording medium.
- the computer-readable recording medium may be a magnetic storage medium (for example, a ROM, a floppy disk, a hard disk, etc.), an optical reading medium (for example, a CD-ROM, a DVD, etc.) and a carrier wave (for example, over the Internet Storage media).
- the method of containerizing an application in the cloud platform provides an isolated application execution environment, enables independent resource allocation, enables multiple applications to operate on the same host, and enables fast operation with OS-level virtualization.
- This small, container-size image is efficient to deploy and update, and can be moved anywhere.
- the multi-cluster provisioning and management method in the cloud platform according to the present invention it is possible to automatically create a multi-cluster environment in which container-based applications can operate in various infrastructures, and to improve operational efficiency by managing version upgrades of clusters remotely. You can.
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Stored Programmes (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
Description
Claims (3)
- 클러스터 리소스 할당 요청이 입력되면, 클라우드 플랫폼 시스템이 서비스 그룹 정보를 생성하는 단계;상기 클라우드 플랫폼 시스템이 상기 서비스 그룹의 사용자를 추가/삭제하는 단계;상기 서비스 그룹의 사용자에게 할당할 클러스터가 선택되고 할당 정보가 입력되는 단계;상기 클라우드 플랫폼 시스템이 클러스터 자원 가용량을 조회하는 단계;상기 자원 가용량 조회 결과 자원이 사용 가능하면, 상기 클라우드 플랫폼 시스템이 상기 서비스 그룹에 자원 할당 정보를 등록하고 자원을 할당하여 클러스터 리소스 할당을 완료하는 단계;상기 자원 가용량 조회 결과 자원이 부족하면, 상기 클라우드 플랫폼 시스템이 클러스터 자원이 추가되는지 확인하는 단계;상기 클러스터 자원이 추가 가능하면, 상기 클라우드 플랫폼 시스템이 상기 클러스터 자원을 추가 등록하는 단계;상기 클라우드 플랫폼 시스템이 상기 서비스 그룹에 자원 할당 정보를 등록하고 자원을 할당하여 클러스터 리소스 할당을 완료하는 단계; 및상기 클러스터 자원이 추가 가능하지 않으면, 가용 클러스터를 재선택하는 단계를 포함하는 클라우드 플랫폼에서의 클러스터 리소스 할당 및 관리 방법.
- 제1항에 있어서,상기 클러스터 리소스 사용을 위해 어플리케이션 자원 사용량이 등록되는 단계;상기 클라우드 플랫폼 시스템이 클러스터 자원 가용량을 조회하는 단계;자원 사용이 가능하면 상기 클라우드 플랫폼 시스템이 어플리케이션 자원을 할당하고 기동하는 단계;상기 클라우드 플랫폼 시스템이 잔여 리소스 할당 정보를 업데이트하는 단계; 및상기 클러스터 리소스 사용이 완료되는 단계를 더 포함하는 클라우드 플랫폼에서의 클러스터 리소스 할당 및 관리 방법.
- 제1항에 있어서,상기 클러스터 리소스 반납을 위해 어플리케이션이 종료되는 단계;상기 클라우드 플랫폼 시스템이 할당 리소스를 회수하는 단계;상기 클라우드 플랫폼 시스템이 클러스터 가용 리소스 정보를 업데이트하는 단계; 및상기 클러스터 리소스 반납이 완료되는 단계를 더 포함하는 클라우드 플랫폼에서의 클러스터 리소스 할당 및 관리 방법.
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US17/257,484 US11520639B2 (en) | 2018-07-19 | 2019-07-15 | Method for allocating and managing cluster resource on cloud platform |
CN201980047562.2A CN112424751A (zh) | 2018-07-19 | 2019-07-15 | 云平台上的集群资源分配与管理方法 |
SG11202100287SA SG11202100287SA (en) | 2018-07-19 | 2019-07-15 | Method for allocating and managing cluster resource on cloud platform |
JP2021502745A JP2021530801A (ja) | 2018-07-19 | 2019-07-15 | クラウドプラットフォームでのクラスターリソース割り当て及び管理方法 |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR10-2018-0084013 | 2018-07-19 | ||
KR1020180084013A KR101987661B1 (ko) | 2018-07-19 | 2018-07-19 | 클라우드 플랫폼에서의 클러스터 리소스 할당 및 관리 방법 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2020017843A1 true WO2020017843A1 (ko) | 2020-01-23 |
Family
ID=66847128
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/KR2019/008697 WO2020017843A1 (ko) | 2018-07-19 | 2019-07-15 | 클라우드 플랫폼에서의 클러스터 리소스 할당 및 관리 방법 |
Country Status (6)
Country | Link |
---|---|
US (1) | US11520639B2 (ko) |
JP (1) | JP2021530801A (ko) |
KR (1) | KR101987661B1 (ko) |
CN (1) | CN112424751A (ko) |
SG (1) | SG11202100287SA (ko) |
WO (1) | WO2020017843A1 (ko) |
Families Citing this family (22)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR101987661B1 (ko) * | 2018-07-19 | 2019-06-11 | 나무기술 주식회사 | 클라우드 플랫폼에서의 클러스터 리소스 할당 및 관리 방법 |
US10893097B1 (en) * | 2019-07-31 | 2021-01-12 | Verizon Patent And Licensing Inc. | Methods and devices for discovering and employing distributed computing resources to balance performance priorities |
KR102284264B1 (ko) * | 2019-11-14 | 2021-08-02 | 에스피테크놀러지 주식회사 | 클라우드 시스템 및 그 제어방법 |
KR102154446B1 (ko) | 2019-11-14 | 2020-09-09 | 한국전자기술연구원 | 분산·협업형 컨테이너 플랫폼 환경에서의 자원 균등 배분을 위한 고속 스케줄링 방법 |
KR102172607B1 (ko) * | 2019-11-15 | 2020-11-02 | 한국전자기술연구원 | 분산·협업형 컨테이너 플랫폼 환경에서의 자원간 균형적 스케일아웃 방법 |
US11991041B2 (en) * | 2019-12-30 | 2024-05-21 | Oracle International Corporation | Autonomous terraforming on cloud infrastructures |
CN111209118A (zh) * | 2020-01-14 | 2020-05-29 | 北京三快在线科技有限公司 | 确定资源余量的方法、装置、存储介质和电子设备 |
KR102289100B1 (ko) * | 2020-05-07 | 2021-08-11 | 한전케이디엔주식회사 | 빅데이터 분석을 위한 컨테이너 기반의 클러스터 구축 방법 및 클러스터 장치 |
US11550633B2 (en) | 2020-10-31 | 2023-01-10 | Nutanix, Inc. | Intra-footprint computing cluster bring-up |
CN113485788B (zh) * | 2021-06-30 | 2023-08-29 | 中国民航信息网络股份有限公司 | 容器资源的分配方法、装置、服务器及计算机存储介质 |
KR102338422B1 (ko) * | 2021-07-30 | 2021-12-10 | 클로우플레이크(주) | Did 기반 클라우드 환경의 협업툴 서비스 제공 방법, 장치 및 시스템 |
US11900172B2 (en) | 2021-07-30 | 2024-02-13 | Nutanix, Inc. | Computing cluster bring-up on public cloud infrastructure using expressed intents |
CN113641503B (zh) * | 2021-09-01 | 2024-05-14 | 上海联蔚盘云科技有限公司 | 多云多集群的Kubernetes管理系统及方法与设备 |
US11829780B2 (en) * | 2021-09-22 | 2023-11-28 | International Business Machines Corporation | Open-source container cluster hot plug adapter |
CN113824790B (zh) * | 2021-09-23 | 2024-04-26 | 信华信技术股份有限公司 | 一种支持企业级应用的云原生PaaS管理平台 |
CN114285850A (zh) * | 2021-12-27 | 2022-04-05 | 北银金融科技有限责任公司 | 一种基于容器平台的跨集群多租户资源管理系统 |
CN114090268B (zh) * | 2022-01-11 | 2022-04-15 | 北京九章云极科技有限公司 | 容器管理方法及容器管理系统 |
CN114915460B (zh) * | 2022-04-28 | 2023-05-05 | 中国人民解放军战略支援部队信息工程大学 | 一种面向容器云的异构动态扩缩容装置及方法 |
KR20240003537A (ko) * | 2022-07-01 | 2024-01-09 | 몬드리안에이아이 주식회사 | 공유자원 기반 원격 분산 학습 시스템 |
CN115086335A (zh) * | 2022-07-27 | 2022-09-20 | 北京思和科创软件有限公司 | 容器云的节点动态添加方法、装置、电子设备及存储介质 |
CN115834600B (zh) * | 2022-11-03 | 2023-07-07 | 中国烟草总公司贵州省公司 | 一种多云纳管数据同步方法、装置、电子设备和存储介质 |
CN117278635B (zh) * | 2023-11-03 | 2024-03-19 | 烽台科技(北京)有限公司 | 工业靶场的调度方法和工业靶场系统 |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20140086436A (ko) * | 2012-12-28 | 2014-07-08 | 삼성에스디에스 주식회사 | 가상 클러스터의 동적 확장 시스템, 방법 및 그 프로그램이 저장된 기록매체 |
KR20140117965A (ko) * | 2013-03-27 | 2014-10-08 | 주식회사 케이티 | 클라우드 api 키를 이용한 자원 할당 방법 및 이를 위한 장치 |
KR101582751B1 (ko) * | 2014-07-01 | 2016-01-06 | 울산대학교 산학협력단 | 협업 작업공간 제공 방법 및 시스템 |
KR101807806B1 (ko) * | 2017-05-02 | 2017-12-11 | 나무기술 주식회사 | 클라우드 플랫폼에서 어플리케이션을 컨테이너화하는 방법 |
KR101826498B1 (ko) * | 2017-05-02 | 2018-02-07 | 나무기술 주식회사 | 클라우드 플랫폼 시스템 |
KR101858715B1 (ko) * | 2017-11-10 | 2018-05-16 | 베스핀글로벌 주식회사 | 서비스자원 관리시스템 및 그 방법 |
KR101987661B1 (ko) * | 2018-07-19 | 2019-06-11 | 나무기술 주식회사 | 클라우드 플랫폼에서의 클러스터 리소스 할당 및 관리 방법 |
Family Cites Families (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7900206B1 (en) * | 2004-03-31 | 2011-03-01 | Symantec Operating Corporation | Information technology process workflow for data centers |
KR20150142871A (ko) | 2014-06-12 | 2015-12-23 | 주홍찬 | 인터렉티브 푸쉬 클라우드 시스템 장치 및 방법 |
JP6320317B2 (ja) * | 2015-02-09 | 2018-05-09 | 日本電信電話株式会社 | リソース融通システム及び方法 |
JP2016189145A (ja) * | 2015-03-30 | 2016-11-04 | 株式会社ミロク情報サービス | 管理装置及び管理方法 |
US10341128B2 (en) * | 2016-03-12 | 2019-07-02 | Wipro Limited | Method and system for optimizing usage of network resources in a communication network |
US10805414B2 (en) * | 2016-07-15 | 2020-10-13 | International Business Machines Corporation | Dynamic resource broker services |
-
2018
- 2018-07-19 KR KR1020180084013A patent/KR101987661B1/ko active IP Right Grant
-
2019
- 2019-07-15 CN CN201980047562.2A patent/CN112424751A/zh active Pending
- 2019-07-15 SG SG11202100287SA patent/SG11202100287SA/en unknown
- 2019-07-15 WO PCT/KR2019/008697 patent/WO2020017843A1/ko active Application Filing
- 2019-07-15 JP JP2021502745A patent/JP2021530801A/ja active Pending
- 2019-07-15 US US17/257,484 patent/US11520639B2/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20140086436A (ko) * | 2012-12-28 | 2014-07-08 | 삼성에스디에스 주식회사 | 가상 클러스터의 동적 확장 시스템, 방법 및 그 프로그램이 저장된 기록매체 |
KR20140117965A (ko) * | 2013-03-27 | 2014-10-08 | 주식회사 케이티 | 클라우드 api 키를 이용한 자원 할당 방법 및 이를 위한 장치 |
KR101582751B1 (ko) * | 2014-07-01 | 2016-01-06 | 울산대학교 산학협력단 | 협업 작업공간 제공 방법 및 시스템 |
KR101807806B1 (ko) * | 2017-05-02 | 2017-12-11 | 나무기술 주식회사 | 클라우드 플랫폼에서 어플리케이션을 컨테이너화하는 방법 |
KR101826498B1 (ko) * | 2017-05-02 | 2018-02-07 | 나무기술 주식회사 | 클라우드 플랫폼 시스템 |
KR101858715B1 (ko) * | 2017-11-10 | 2018-05-16 | 베스핀글로벌 주식회사 | 서비스자원 관리시스템 및 그 방법 |
KR101987661B1 (ko) * | 2018-07-19 | 2019-06-11 | 나무기술 주식회사 | 클라우드 플랫폼에서의 클러스터 리소스 할당 및 관리 방법 |
Also Published As
Publication number | Publication date |
---|---|
CN112424751A (zh) | 2021-02-26 |
SG11202100287SA (en) | 2021-02-25 |
JP2021530801A (ja) | 2021-11-11 |
US20210124624A1 (en) | 2021-04-29 |
US11520639B2 (en) | 2022-12-06 |
KR101987661B1 (ko) | 2019-06-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2020017843A1 (ko) | 클라우드 플랫폼에서의 클러스터 리소스 할당 및 관리 방법 | |
WO2020017847A1 (ko) | 클라우드 플랫폼에서의 멀티 클러스터 프로비저닝 및 관리 방법 | |
WO2020017844A1 (ko) | 클라우드 플랫폼에서 복수의 클러스터 및 어플리케이션을 모니터링하는 방법 | |
WO2018203635A1 (ko) | 클라우드 플랫폼에서 어플리케이션을 컨테이너화하는 방법 | |
WO2020017846A1 (ko) | 클라우드 플랫폼에서 어플리케이션 컨테이너의 볼륨(스토리지) 프로비저닝 방법 | |
WO2018203634A1 (ko) | 클라우드 플랫폼 시스템 | |
US9338067B2 (en) | Network resource deployment for cloud-based services | |
WO2009098909A1 (ja) | 仮想アプライアンス配備システム | |
US20090204961A1 (en) | Systems and methods for distributing and managing virtual machines | |
CN105426208A (zh) | 一种云计算系统OpenStack离线自动安装方法 | |
WO2016153288A1 (ko) | 가상 클러스터 관리 시스템 및 이를 제어하기 위한 방법 | |
WO2023128699A1 (ko) | 서비스 레벨 협약 기반의 클라우드 네이티브 네트워크 기능의 자원 할당 장치 및 방법 | |
WO2024063585A1 (ko) | 거대규모 워크플로우를 위한 분산 및 병렬처리가 가능한 클라우드 플랫폼 시스템 및 서비스 방법 | |
WO2023085738A1 (ko) | Ai 모델 서빙 시스템 및 방법 | |
WO2018088629A1 (ko) | 읽기 및 쓰기가 가능한 가상 디스크의 병합 마운팅을 통한 가상 클러스터 구축 방법 및 장치 | |
Ward et al. | Deploying Azure SQL | |
Syrewicze et al. | Migrating Virtual Machines Between Hosts | |
CN115599500A (zh) | 基于sdn网络同步的安全容器热迁移方法和系统 | |
Zhao et al. | A Remote and Unified Software Automated Deployment Platform |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 19838733 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2021502745 Country of ref document: JP Kind code of ref document: A |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 210521) |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 19838733 Country of ref document: EP Kind code of ref document: A1 |