Search

JP-2026075270-A - Information processing system, control device, control method, and program

JP2026075270AJP 2026075270 AJP2026075270 AJP 2026075270AJP-2026075270-A

Abstract

[Challenge] To enable dynamic management of multiple clusters. [Solution] An information processing system that functions as a cluster, comprising: a data search controller that receives a process request from a user, searches within the information processing system to determine whether or not the process exists within the information processing system, and executes processing according to the determination result; and a gateway device for establishing a tunnel between another cluster and the cluster. [Selection Diagram] Figure 1

Inventors

  • 新谷 翔
  • 沈 文裕
  • 和田 雄一郎
  • 上野山 実
  • 米倉 倫之介

Assignees

  • ＮＴＴドコモビジネス株式会社

Dates

Publication Date
20260508
Application Date
20241022

Claims (9)

  1. An information processing system that functions as a cluster, A data search controller receives a process request from a user, searches within the information processing system to determine whether the process exists within the information processing system, and executes processing according to the determination result. An information processing system comprising a gateway device for establishing a tunnel between another cluster and the aforementioned cluster.
  2. If the process does not exist within the information processing system, the data search controller creates the process within the information processing system, as described in claim 1.
  3. The information processing system according to claim 1, wherein if the process does not exist within the information processing system and it is not possible to create the process within the information processing system, the system requests the process from the inter-cluster distributed controller, and the inter-cluster distributed controller creates a virtual process within the information processing system that utilizes the process in another cluster.
  4. A control device that controls multiple clusters having a first cluster and a second cluster, A receiving unit that receives process requests from the first cluster, A control device comprising a control unit that creates a tunnel between the second cluster having the process and the first cluster, and creates a virtual process in the first cluster that utilizes the process.
  5. The control device according to claim 4, wherein the control unit deletes the virtual process and the process corresponding to the virtual process based on the number of times the virtual process has been referenced.
  6. The control device according to claim 4, wherein the control unit monitors each cluster and, when it detects an abnormality in a cluster, migrates the process of that cluster to another cluster.
  7. The control device according to claim 4, wherein the control unit monitors the resource utilization rate of each cluster, and when it determines that load balancing is necessary in a certain cluster, it creates a virtual process in that cluster corresponding to a specific process in that cluster, creates the specific process in another cluster, and deletes the specific process in that cluster.
  8. A control method executed by a control device that controls multiple clusters having a first cluster and a second cluster, The steps include receiving a process request from the first cluster, A control method comprising the steps of creating a tunnel between the second cluster having the process and the first cluster, and creating a virtual process in the first cluster that utilizes the process.
  9. A program for causing a computer to function as a component of the control device described in any one of claims 4 to 7.

Description

This invention relates to cloud computing technology. In recent years, cloud computing technology has become widespread (for example, Non-Patent Document 1). In cloud computing, services are provided through clusters built using orchestration tools such as Kubernetes (registered trademark). Cluster configurations include single-cluster configurations using one cluster and multi-cluster configurations using multiple clusters. Single-cluster configurations are currently the mainstream. When using a single-cluster configuration, there are concerns that insufficient processing power for workloads in a single cluster, or a failure in a single cluster, could impact the entire system. Therefore, there is a growing need for technologies that utilize a multi-cluster configuration rather than a single-cluster configuration for distributed management. While technologies exist for creating multi-clusters, these technologies are static, requiring administrators to manually manage the creation and operation of multiple clusters and processes at all times. NTT Communications Smart Data Platform Edge service, https://www.ntt.com/business/services/data-utilization/dxplatform/sdpf/edge.html, accessed October 17, 2024. This figure shows an example of the configuration of a communication system in the first embodiment.This is a flowchart of the first embodiment.This is a diagram to explain the details of S1.This is a diagram illustrating the details of S2-a.This is a diagram to explain the details of S2-b.This is a diagram to explain the details of S2-c.This is a diagram to explain the details of S3.This is a diagram to explain the details of S4.This figure shows an example of the configuration of the communication system in the second embodiment.This is a flowchart of the second embodiment.This is a diagram to explain the details of S11 and S12.This is a diagram to explain the details of S11 and S12.This is a diagram showing an example of the configuration of a communication system in the third embodiment.This is a flowchart of the third embodiment.This is a diagram to explain the details of S21.This is a diagram to explain the details of S22.This figure shows an example of the configuration of the communication system in the fourth embodiment.This is a flowchart of the fourth embodiment.This is a diagram to explain the details of S31.This is a diagram to explain the details of S32.This is a diagram to explain the details of S33.This is a diagram showing an example configuration of the control device 200.This figure shows an example of the device's hardware configuration. The embodiments of the present invention (this embodiment) will be described below with reference to the drawings. The embodiments described below are merely examples, and the embodiments to which the present invention applies are not limited to these embodiments. In this embodiment, a "cluster" is a collection of computing resources. These computing resources may be physical computers, virtual machines, containers, CPUs, GPUs, or other resources. Furthermore, the cluster in this embodiment may be a cluster within Kubernetes®. Also, a "cluster" may be a single computer (server, etc.). The term "cluster" may also be referred to as an information processing system. Furthermore, a "process" created and executed within a cluster is, for example, an application that provides services to users. A "process" may also be a container. Creating a "process" may mean starting an application or a container. (Summary of the embodiment) This embodiment achieves dynamic multi-cluster management, which was not possible with conventional technology. Specifically, it achieves dynamic multi-cluster management by introducing a data discovery controller that records user requests and creates processes that match user requests, and an inter-cluster distribution controller that connects clusters. The configuration and operation of the communication system in this embodiment will be described in detail below using the first to fourth embodiments. The first to fourth embodiments can be implemented in any combination. (First Embodiment) First, the first embodiment will be described. Figure 1 shows an example of the configuration of the communication system in the first embodiment. As shown in Figure 1, this communication system has cluster 100A, cluster 100B, and an inter-cluster distributed controller 200. The clusters are connected by VPN 400. Figure 1 also shows an end user 300. Specifically, the end user 300 is a user's terminal, etc. Also, for convenience of description, the end user 300 may be referred to as "user". The inter-cluster distributed controller may be referred to as a "control device". Cluster 100A has a gateway device 110A for communication between clusters. Furthermore, in the state shown in Figure 1, cluster 100A has actual processes. Cluster 100B has a gateway device 110B and a data discovery controller 120B. Furthermore, in the state shown in Figure 1, cluster 10