System and method for scaling worker node in multi-hybrid clouds

Abstract

The present invention relates to a system and method for scaling a worker node, wherein one or more environments among a public cloud and a private cloud outside a cluster may be selected, a VM may be received from the selected environment through tunneling, and a worker node may be added in the cluster. According to the present invention, it is possible to extend the worker node in the cluster to an external network having a different network environment by utilizing tunneling technology in a multi-hybrid cloud environment.

Description

A system and method for scaling worker nodes in multi- / hybrid clouds.

[0001] This application claims the benefit of Korean Patent Application No. 10-2023-0178833, filed December 11, 2023, with the Korean Intellectual Property Office, the entire contents of which are incorporated herein by reference.

[0002] The present invention relates to a system and method for scaling worker nodes in a multi / hybrid cloud environment, and more particularly, to a system and method for managing the scalability and performance of a cluster by scaling and managing worker nodes by extending them to an external network of an external cloud beyond the same cluster area.

[0003] Multicloud and hybrid cloud both refer to cloud deployments that integrate two or more clouds. However, multicloud and hybrid cloud differ in the types of infrastructure they incorporate. Hybrid cloud infrastructure combines two or more different types of clouds, while multicloud combines different types of clouds of the same type.

[0004] Specifically, multi-cloud refers to the combination or integration of multiple public clouds, while hybrid cloud refers to the combination of public cloud computing and private cloud (also referred to as on-premise in this specification, and is distinguished from legacy on-premise infrastructure that is completely isolated from external networks and is an IT resource built within a company).

[0005] Hybrid cloud deployments, as well as multi-cloud deployments, are becoming increasingly common. Some companies partially migrate IT resources to the public cloud, but maintain some processes, business logic, and data storage on-premises. This is because moving all resources to the public cloud is too costly or resource-intensive.

[0006] Enterprises can adopt a hybrid cloud strategy to leverage the greater resources and lower overhead of public cloud computing while maintaining some processes and data in a more controlled environment (private cloud).

[0007] At this time, when worker node expansion is required, automatic worker node expansion is only possible within the same cluster area in a multi / hybrid cloud environment.

[0008] Therefore, a technology is needed to manage worker nodes beyond the same cluster in a multi / hybrid cloud environment and maintain cluster scalability and performance.

[0009] Therefore, the technical problem that the present invention seeks to solve is to provide a technology for managing the scalability and performance of a cluster by scaling and managing worker nodes by extending them to an external network of an external cloud in a different network environment beyond the same cluster area.

[0010] A worker node scaling method according to one aspect of the present invention for solving the above technical problem may include a step of selecting one or more cloud environments among a public cloud or a private cloud outside a cluster; and a step of adding worker nodes within the cluster by providing a VM (Virtual Machine) through tunneling from the selected environment.

[0011] At this time, the cloud environment outside the cluster may be a different network environment from the cluster.

[0012] At this time, the step of selecting the environment may include a step of evaluating the importance of each factor for each environment-specific evaluation factor, a step of assigning a weight to each environment based on the result of the importance evaluation of each factor, and a step of selecting the environment with the highest weight based on the result of the weight assignment as the most efficient external environment.

[0013] And, the above environment-specific evaluation factors may include one or more of cost, latency, provisioning speed, and stability.

[0014] In addition, the step of selecting the above environment may be a step in which conditions for executing worker node scaling using a cloud environment outside the cluster are pre-entered according to a priority selected by the user, and an external environment to be selected for worker node scaling is pre-set.

[0015] Additionally, the step of selecting the environment may be a step of checking the workload within the cluster and selecting one or more cloud environments outside the cluster when adding worker nodes is required.

[0016] Meanwhile, the step of adding the worker node may include: performing tunneling for the selected external environment; receiving a VM from the selected external environment through the tunneling; and adding a new worker node within the cluster based on the VM.

[0017] At this time, the step of adding the worker node may be a step that is automatically executed according to requirements preset by the user when the workload within the cluster is checked and addition of a worker node is required.

[0018] Alternatively, the step of adding the worker node may be a step that checks the workload within the cluster and, if it is necessary to add a worker node, provides a notification display to the user and is executed when the user selects to add a worker node.

[0019] And, the above tunneling may be secure tunneling.

[0020] Meanwhile, a worker node scaling system according to one aspect of the present invention for solving the above technical problem may include: a worker node that creates a container and executes an application; a controller that manages workload and bandwidth within a cluster and communicates about the status of the worker nodes; and a scaler that selects one or more cloud environments from among a public cloud or a private cloud outside the cluster, provides a VM through tunneling from the selected environment, and adds a worker node within the cluster.

[0021] At this time, the cloud environment outside the cluster may be a different network environment from the cluster.

[0022] In addition, the scaler can evaluate the importance of each factor for each evaluation factor of one or more cloud environments outside the cluster, set a priority for each environment based on the result of the importance evaluation for each factor, and select the environment with the highest priority as the most efficient external environment.

[0023] Meanwhile, the above evaluation factors may include one or more of cost, latency, provisioning speed, and stability.

[0024] At this time, the scaler may include a condition setting module that sets a condition for executing worker node scaling; a monitoring module that checks whether additional worker nodes are required through a cloud environment outside the cluster; a tunneling module that executes tunneling for external environment resources to expand worker nodes through the external environment; and a scaling module that executes an analysis for each of one or more cloud environments outside the cluster, selects the most efficient external environment, and creates a new worker node within the cluster based on a VM provided through tunneling from the selected external environment.

[0025] In addition, the scaling module can analyze and evaluate in advance and set the external environment to be selected for scaling the worker node according to the priority conditions for the external environment of the cluster of the input user.

[0026] At this time, the monitoring module checks the workload within the cluster to determine whether additional worker nodes are needed, and the scaling module, when the monitoring module determines that additional worker nodes are needed, performs an analysis on each of one or more cloud environments outside the cluster and selects the most efficient external environment.

[0027] In addition, when the workload within the cluster is checked and it is necessary to add worker nodes, the scaling module can automatically select one or more cloud environments outside the cluster based on requirements preset by the user, and provide VMs from the selected environments through tunneling to add worker nodes within the cluster.

[0028] And when the workload within the cluster is checked and it is necessary to add worker nodes, the monitoring module provides a notification display to the user, and when the user selects to add worker nodes, the scaling module can execute the addition of worker nodes within the cluster from an external environment through tunneling.

[0029] At this time, the tunneling may be secure tunneling.

[0030] As described above, according to the present invention, it is possible to safely access and add new nodes while maintaining security by extending beyond the same cluster area to an external network in a different network environment.

[0031] And according to the present invention, VM provisioning is possible in an external cloud area through secure tunneling, so that new nodes can be added to the system.

[0032] In addition, according to the present invention, weights are distributed among multiple environments of each public cloud and private cloud according to the requirements of each environment, so that the most efficient environment can be selected based on the weight results when scaling worker nodes.

[0033] FIG. 1 is a diagram illustrating the overall configuration of a worker node scaling system in a multi / hybrid cloud environment according to one aspect of the present invention.

[0034] Figure 2 is a configuration diagram of a scaler according to one aspect of the present invention.

[0035] FIG. 3 is an example of a user input screen for auto-scaling setting conditions in a multi / hybrid cloud environment according to one aspect of the present invention.

[0036] FIG. 4 is an example of the state of a currently operating node according to one aspect of the present invention.

[0037] Figure 5 is an example of environmental requirements for each cloud service in a multi-cloud environment according to one aspect of the present invention.

[0038] FIG. 6 is a detailed diagram of a process for adding a node through secure tunneling in a multi / hybrid environment according to one aspect of the present invention.

[0039] Figure 7 is a diagram illustrating a secure tunneling process according to one aspect of the present invention.

[0040] FIG. 8 is a detailed diagram of a process in which an AWS cloud environment is selected and added as a new node to an original cluster according to preset user conditions, according to one aspect of the present invention.

[0041] FIG. 9 is an example of the status of all nodes including newly added nodes after worker node scaling according to one aspect of the present invention.

[0042] FIG. 10 is an example of a process for analyzing environmental factors in a multi / hybrid cloud environment and selecting an environment and creating a new node based on the weighted results according to one aspect of the present invention.

[0043] Below, with reference to the attached drawings, embodiments of the present invention are described in detail so that those skilled in the art can easily practice them. However, the present invention may be implemented in various different forms and is not limited to the embodiments described herein. In addition, in the drawings, parts irrelevant to the description are omitted for clarity of description, and similar parts are designated with similar reference numerals throughout the specification.

[0044] Throughout the specification, whenever a part is said to "include" a component, this does not mean that it excludes other components, but rather that it may include other components, unless otherwise stated.

[0045] Additionally, terms such as “unit,” “device,” and “module” described in the specification mean a unit that processes at least one function or operation, which may be implemented by hardware, software, or a combination of hardware and software.

[0046] The devices described in the present invention are comprised of hardware including at least one processor, a memory device, a communication device, and the like, and a program that is executed by being combined with the hardware and stored in a designated location. The hardware has a configuration and performance capable of executing the method of the present invention. The program includes instructions that implement the operating method of the present invention described with reference to the drawings, and executes the present invention by being combined with hardware such as a processor and a memory device.

[0047] In this specification, “transmitting or providing” may include not only direct transmission or providing, but also indirect transmission or providing via another device or by using a bypass route.

[0048] In this specification, expressions described in the singular may be interpreted as singular or plural, unless explicit expressions such as “one” or “single” are used.

[0049] In this specification, the same drawing numbers refer to the same components regardless of the drawings, and “and / or” includes each and every combination of one or more of the mentioned components.

[0050] In this specification, terms including ordinal numbers, such as "first" and "second," may be used to describe various components, but these components are not limited by these terms. These terms are used solely to distinguish one component from another. For example, without departing from the scope of the present disclosure, a first component could be referred to as a "second component," and similarly, a second component could also be referred to as a "first component."

[0051] In the flowcharts described with reference to the drawings in this specification, the order of operations may be changed, several operations may be merged, some operations may be split, and certain operations may not be performed.

[0052] First, the worker node scaling system described in this specification can improve worker node expansion capabilities in a multi / hybrid cloud environment.

[0053] When expanding worker nodes in a conventional hybrid cloud environment, automatic expansion was only possible within the same cluster area.

[0054] However, according to the present invention, in order to improve worker node expansion functionality, it is possible to enable VM (Virtual Machine) provisioning in an external cloud area in a network environment different from that of the cluster by utilizing Secure Tunneling technology.

[0055] Tunneling is a method of making an open network into a secure network by creating a tunnel that prevents external intrusion. Secure tunneling allows you to open a session and establish two-way communication with a remote communication device through a secure connection.

[0056] According to the present invention, VMs can be provisioned while safely accessing and maintaining security by extending beyond the same cluster to an external network.

[0057] At this time, for newly formed nodes via VM provisioning, Kernel Bypass can improve network speed and provide application-optimized routes. When Kernel Bypass is enabled, packets can bypass iptables and LVS distribution functions, enabling fast data transfer for newly formed nodes.

[0058] In the following specification, a node means a worker node unless explicitly stated otherwise.

[0059] Referring to the drawings below, a multi / hybrid cloud environment and worker node scaling system according to one aspect of the present invention are described in detail.

[0060] First, as shown in Fig. 1, a public cloud (10) may exist as a computing environment.

[0061] A public cloud (10) is a cloud computing model in which IT infrastructure, such as servers, networking, and storage resources, are provided as virtual resources accessible via the Internet. Public cloud providers maintain a vast network of physical data centers across multiple countries. These data centers house the physical hardware and software tools that support public cloud services, such as servers, storage devices, and network equipment. Representative public cloud providers include AWS (Amazon Web Services), Google Cloud, IBM Cloud, Alibaba Cloud, and Microsoft Azure.

[0062] The public cloud (10) environment may exist alone, such as AWS or Azure, or may exist as a multi-cloud environment that selectively utilizes two or more public cloud providers.

[0063] The worker node scaling system (20) may include a database (201), a master node (203), a worker node (205), a control plane (207), and a scaler (209), and may be divided into an expansion area and a non-expansion area.

[0064] The worker node scaling system (20) runs worker nodes, such as a database (201), in a non-scalable area and may not apply automatic scaling. Therefore, the worker node scaling system (20) can handle stable and predictable worker nodes in a non-scalable area.

[0065] The worker node scaling system (20) can manage cluster expansion and workload in an expansion area, including a master node (203), a worker node (205), a controller (207), and a scaler (209), and can maintain the scalability and performance of the cluster.

[0066] The master node (203) can assign pods to worker nodes (205) and launch containers within the pods. In addition, the master node (203) can manage the status of the cluster and execute scheduling of multiple pods, etc.

[0067] The worker node (205) is a node where the actual application is executed and can create a container using a designated container runtime and image.

[0068] The controller (207) can coordinate communication between each service to maintain communication stability and reliability, apply necessary security policies, and perform performance monitoring and logging.

[0069] The controller (207) can manage the workload and bandwidth within the cluster and communicate with the scaler (209) about the status of the nodes (205).

[0070] The scaler (209) can dynamically expand worker nodes (205) within a cluster by utilizing secure tunneling technology in a multi / hybrid cloud environment.

[0071] When additional nodes are required for workload management, the scaler (209) can access an external public cloud / private cloud (on-premises) area with a different network environment from the cluster through tunneling, thereby allowing external resources to be utilized as additional nodes for the cluster.

[0072] Referring to FIG. 2 below, the scaler (209) is described in detail.

[0073] The scaler (209) may include a condition setting module (2091), a monitoring module (2093), a tunneling module (2095), and a scaling module (2097).

[0074] The condition setting module (2091) can set scaling execution requirements to expand the cluster by adding nodes by accessing external public / private cloud areas with different network environments.

[0075] Additionally, the condition setting module (2091) may assign priorities or weights according to the importance of the requirements.

[0076] For example, the condition setting module (2091) can set whether to automatically execute the scaling function when adding nodes from an external cloud area is required, external clouds that can be targets, and factors to be given priority in selecting targets.

[0077] The monitoring module (2093) can check the workload within the cluster to determine whether additional worker nodes are needed through an external network in a different network environment.

[0078] The monitoring module (2093) can notify the user through the client device (30) or trigger auto scaling when a preset worker node scaling condition via an external network is met.

[0079] The tunneling module (2095) can securely enable access to selected external resources through tunneling technology for worker node expansion through an external network.

[0080] The scaling module (2097) can perform analysis on multiple external environments, including public clouds and on-premises, based on scaling execution requirements and importance for the cloud cluster, and select the most efficient external environment.

[0081] The scaling module (2097) can analyze and evaluate in advance and set the external environment to be selected for worker node scaling according to the priority conditions for the external environment of the user's cluster.

[0082] Alternatively, the scaling module (2097) may run an analysis on each of one or more cloud environments outside the cluster to select the most efficient external environment when the need for adding worker nodes is identified by the monitoring module (2093).

[0083] And the scaling module (2097) can create new nodes and allocate necessary resources according to the selected external environment.

[0084] FIG. 3 is an example of a user screen for setting conditions for executing worker node scaling via an external network according to one aspect of the present invention.

[0085] Specifically, the user can set whether to enable or disable the worker node scaling function. Furthermore, the user can also set whether to automatically execute worker node scaling when the worker node scaling function conditions are met, or whether to manually execute it.

[0086] The user screen displays external computing environments that can be targeted when executing worker node scaling, and the user can preset the factors that are given the highest priority when selecting a target.

[0087] For example, users can pre-select one of cost, latency, provisioning speed, and reliability as priority factors in selecting the most efficient external environment.

[0088] According to Figure 3, the user enabled the worker node scaling feature and set it to auto-scaling mode. The external clouds targeted for auto-scaling include on-premises, Azure, Google, and AWS, and low cost was set as the most important factor in selecting the target.

[0089] Accordingly, according to FIG. 3, when the system (20) determines that it is necessary to add a worker node from an external cluster area, a worker node may be added from the external cloud with the lowest cost among the external cloud environments that the user has.

[0090] Alternatively, if the user chooses to manually scale worker nodes, worker nodes may be added from one or more external cloud environments of the user's choice.

[0091] Alternatively, if the user selects to manually execute worker node scaling, the system (20) provides the user with a notification regarding the need for worker node scaling, and if the user selects to execute the worker node scaling function, the system can add worker nodes from an external cloud according to the priority set by the user among the external cloud environments.

[0092] Meanwhile, as illustrated in Figure 4, 14 nodes in the current cluster are in a Ready state and pods can be assigned to them. However, as illustrated, two of the 14 nodes are already experiencing significant resource usage compared to the remaining 12 nodes.

[0093] At this time, according to the present invention, the system (20) can access an external public / private cloud area and execute worker node scaling.

[0094] As a specific example, when the indicators for cost, latency, provisioning, and stability for each cloud environment are as shown in Figure 5, AWS is confirmed to have the lowest cost.

[0095] At this time, as illustrated in Figure 3, low cost is set as the most important factor in selecting a target external cloud, so AWS can be selected as the target external cloud when scaling worker nodes.

[0096] Therefore, as illustrated in FIG. 6, the scaler (209) can dynamically expand worker nodes (205) within the cluster by accessing AWS as an external public cloud using tunneling technology.

[0097] To enhance worker node expansion capabilities, the scaler (209) utilizes secure tunneling technology to execute VM provisioning in an external cloud region, so that worker nodes (205) can be added within the cluster.

[0098] The added worker node (205) can provide network speed improvements and application-optimized paths through kernel bypass, and can also support fast data transmission. In this case, network processing is performed directly by the application, enabling fast network processing and resulting in improved performance.

[0099] Figure 7 describes in more detail, for example, the process of tunneling construction executed by the tunneling module (2095) to securely access external resources.

[0100] First, an initial request to establish a tunnel can be made from the user's domain (S100) when creating a tunnel. For example, a VPN or SSH tunneling request can be made.

[0101] The following initial request for tunnel setup is transmitted to the kernel, and an instruction to prepare resources required for tunnel setup is executed (S200), and the location of resources required for tunneling can be identified (S201).

[0102] In addition, a check for incorrect requests or errors is performed (S203), and a check for whether access to the resource is authorized can be performed through discretionary access control (DAC) (S205).

[0103] At this time, whether the tunneling request is safe can be verified through the security enhancement policy and whether to allow the setting can be determined, and the security enhancement module can allow or deny the tunneling request (S207, S209).

[0104] Additionally, additional logging and security checks may be performed to determine whether tunneling is permitted (S211). At this time, relevant log information may be stored, for example, in 'sys / fs / croup', or as subdirectories in 'sys / fs / croup / child1' and 'sys / fs / croup / child2', for use in debugging or monitoring (S213).

[0105] Next, metadata for tunneling can be secured and setup completed through inode access (S215).

[0106] Accordingly, as illustrated in FIG. 8, the system (20) can utilize the tunneling technology to execute VM provisioning from an external cloud and add a worker node (205) to the corresponding cluster area.

[0107] More specifically, the system (20) can utilize the tunneling technology to add worker nodes (205) from AWS selected according to priority settings for low cost, thereby maintaining the scalability and performance of the cluster.

[0108] That is, according to the present invention, in order to intelligently optimize resources and manage traffic in a multi / hybrid environment, requirements for scaling execution can be set for a cloud cluster and weights can be assigned according to the importance of the requirements.

[0109] And according to the present invention, when worker node scaling is performed according to set requirements, analysis can be performed on multiple environments including each public cloud and private cloud.

[0110] The system (20) can select the most efficient environment for the user based on the weighted results, create a new node (205) according to the selected environment, and allocate necessary resources to normalize traffic.

[0111] Accordingly, as illustrated in FIG. 9, the system (20) can expand into a cluster including a total of 16 nodes, including 14 nodes currently in operation in the existing on-premises cluster, by additionally creating two nodes from AWS, which is a selected external environment.

[0112] Referring to FIG. 10 below, a process of selecting the most efficient environment for scaling execution and creating a new node by evaluating environmental factors according to set importance in a multi / hybrid cloud environment according to one aspect of the present invention is summarized and explained.

[0113] First, the system (20) can select a hybrid environment including multi-cloud or on-premise as an external environment to be subject to worker node scaling, and check evaluation factors for each environment (S300, S301).

[0114] For example, the system (20) can check cost, latency, provisioning speed, and stability for each environment.

[0115] The system (20) can evaluate the importance of each element based on preset requirements. For example, if cost is important, the highest weight can be given to the environment with the lowest cost (S303-1, S303-5, S303-6, S303-7, S307).

[0116] Also, for example, the system (20) may assign the highest weight to the environment with the lowest latency or the fastest provisioning speed (S307) when latency is important (S303-2, S303-6, S303-7) or provisioning speed is important (S303-3, S303-7). In addition, the system (20) may assign the highest weight to the environment with the best stability when stability is important (S303-4, S307).

[0117] In addition, the system (20) can assign a high weight to an environment with low cost and short latency (S307) when cost and latency are important, so that the element-by-element importance evaluation can be executed in various embodiments as illustrated.

[0118] Meanwhile, the system (20) may also execute weight redistribution if necessary based on the results of the importance evaluation for each evaluation factor for each environment (S305).

[0119] The system (20) can select the most efficient external environment based on the weighted results (S309). Furthermore, the system (20) can utilize tunneling technology based on the selected environment to create a new node in the cluster and allocate necessary resources (S311).

[0120] Therefore, according to the present invention, the system (20) can access an external environment to expand a cluster, and normalize traffic by applying a traffic management strategy such as load balancing or auto-scaling in the expanded cluster (S313).

[0121] The embodiments of the present disclosure described above are not implemented only through devices and methods, but may also be implemented through a program that realizes a function corresponding to the configuration of the embodiments of the present disclosure or a recording medium on which the program is recorded.

[0122] Although the embodiments of the present disclosure have been described in detail above, the scope of the present disclosure is not limited thereto, and various modifications and improvements made by those skilled in the art using the basic concepts of the present disclosure defined in the following claims also fall within the scope of the present disclosure.

[0123] [Explanation of symbols]

[0124] 10: Public Cloud

[0125] 20: Worker Node Scaling System

[0126] 30: Client

[0127] 201: Database

[0128] 203: Masternode

[0129] 205: Worker Node

[0130] 207: Control Plane

[0131] 209: Scaler

[0132] 2091: Condition Setting Module

[0133] 2093: Monitoring Module

[0134] 2095: Tunneling Module

[0135] 2097: Scaling Module

Claims

1. As a method for scaling worker nodes of the system, Step of selecting one or more cloud environments, either public cloud or private cloud, outside the cluster; and A method comprising: a step of providing a VM (Virtual Machine) through tunneling from the selected environment and adding a worker node within the cluster; 2. In paragraph 1, A method wherein a cloud environment outside the above cluster is a different network environment from the above cluster.

3. In paragraph 2, The steps for selecting the above environment are: For each environmental evaluation factor, a step is taken to evaluate the importance of each factor. A step of assigning weights to each environment based on the importance evaluation results for each of the above elements; and A method comprising: selecting an environment with the highest weight as the most efficient external environment based on the above weighting results.

4. In paragraph 3, A method wherein the above environmental evaluation factors include one or more of cost, latency, provisioning speed, and stability.

5. In paragraph 4, The steps for selecting the above environment are: A method in which conditions for executing worker node scaling using a cloud environment outside the cluster are pre-entered according to a priority selected by a user, and an external environment to be selected for worker node scaling is pre-set.

6. In paragraph 4, The steps for selecting the above environment are: A method for selecting one or more cloud environments outside the cluster by checking the workload within the cluster and when additional worker nodes are required.

7. In paragraph 5 or 6, The steps to add the above worker node are: A step of executing tunneling for the above selected external environment; A step of providing a VM from the selected external environment through the tunneling; and A method comprising: adding a new worker node within a cluster based on the VM; 8. In paragraph 7, The steps to add the above worker node are: A method that automatically executes steps based on requirements preset by the user when checking the workload within the cluster and requiring the addition of worker nodes.

9. In paragraph 7, The steps to add the above worker node are: A method for checking the workload within a cluster to determine if adding worker nodes is necessary, and providing a notification display to the user, which is a step to be executed when the user selects to add worker nodes.

10. In paragraph 7, The above tunneling is a secure tunneling method.

11. As a worker node scaling system, Worker nodes that create containers and run applications; A controller that manages workload and bandwidth within the cluster and communicates about the status of the worker nodes; and A system comprising: a scaler for selecting one or more cloud environments, either public or private, outside the cluster, and adding worker nodes within the cluster by provisioning VMs via tunneling from the selected environments.

12. In paragraph 11, A cloud environment outside the above cluster is a system with a different network environment from the above cluster.

13. In paragraph 12, The above scaler, A system that evaluates the importance of each evaluation factor for one or more cloud environments outside a cluster, sets priorities for each environment based on the results of the importance evaluation for each factor, and selects the environment with the highest priority as the most efficient external environment.

14. In paragraph 13, A system wherein the above evaluation factors include one or more of cost, latency, provisioning speed, and stability.

15. In paragraph 14, The above scaler, Condition setting module that sets the conditions for executing worker node scaling; A monitoring module that checks whether additional worker nodes are needed through a cloud environment outside the cluster; A tunneling module that performs tunneling to external environment resources for worker node expansion through the external environment; and A system comprising a scaling module that runs an analysis on each of one or more cloud environments outside the cluster, selects the most efficient external environment, and creates new worker nodes within the cluster based on VMs tunneled from the selected external environment.

16. In paragraph 15, The above scaling module, A system that analyzes and evaluates in advance the external environment to be selected for scaling the worker node, based on the priority conditions for the external environment of the cluster of the input user.

17. In paragraph 15, The above monitoring module checks the workload within the cluster to determine whether additional worker nodes are needed. The scaling module is a system that runs an analysis on each of one or more cloud environments outside the cluster when the need for adding worker nodes is identified by the monitoring module, and selects the most efficient external environment.

18. In any one of paragraphs 15 to 17, Check the workload within the cluster to see if adding worker nodes is necessary. A system in which the scaling module automatically selects one or more cloud environments outside the cluster according to requirements preset by the user, provides VMs from the selected environments through tunneling, and adds worker nodes within the cluster.

19. In any one of paragraphs 15 to 17, Check the workload within the cluster to see if adding worker nodes is necessary. The above monitoring module provides a notification display to the user, A system in which the scaling module executes the addition of worker nodes within the cluster from an external environment via tunneling when the user selects to add worker nodes.

20. In any one of paragraphs 15 to 17, The above tunneling is a secure tunneling system.

Images