Multi-cluster container orchestration with predictive node availability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cloud computing environments face challenges in providing consistent service levels and security while maintaining cost-effectiveness, as they lack direct control and are prone to changes in service offerings and pricing, necessitating efficient management of container-based virtual machines to ensure high performance and resource utilization.
Innovation Solution
A method involving a manager node that receives metrics data from compute nodes, uses orchestrator availability data to select suitable nodes for hosting container-based applications, and sends command data to respawn applications on available nodes across different computing environments, leveraging machine learning for predictive analytics to optimize resource allocation and availability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If cloud service providers are used to provide computing services, then service speed and cost-effectiveness are improved, but control over service levels and security is reduced
Solution Approach 1:
The system segments the computing environment into multiple isolated clusters (first computing environment, second computing environment) that can be independently managed. Each cluster maintains its own service levels and security policies, allowing the organization to leverage cloud providers for speed while preserving control over critical service parameters through localized management policies.
Solution Approach 2:
The system applies different management policies and control mechanisms to different clusters based on their specific requirements. The first cluster may use cloud providers for rapid provisioning, while the second cluster can enforce stricter security and service level controls, allowing each environment to have optimized local characteristics rather than a one-size-fits-all approach.
2Productivity
If container-based virtualization is used to increase resource density, then performance and resource utilization are improved, but management complexity across multiple clusters increases
Solution Approach 1:
The system implements a universal container management approach that works across multiple different computing environments and cluster types. The same containerization technology and management principles are applied consistently across the first and second clusters, enabling portable container images and unified orchestration despite the underlying hardware and cloud provider differences.
Solution Approach 2:
The system introduces an intermediary layer (the patent system itself) that sits between the diverse cloud providers and the container workloads. This intermediary standardizes the interface for deploying and managing containers across different clusters, abstracting away the underlying complexity of each cloud provider while maintaining efficient resource utilization through containerization.
3Reliability
If applications are respawned on remote nodes in different computing environments, then availability is improved, but network latency and communication overhead increase
Solution Approach 1:
The system performs preliminary actions by pre-positioning container images and configuration data in both the first and second computing environments before failures occur. When a failure is detected, the system can immediately respawn applications on remote nodes without waiting for image retrieval, significantly reducing respawn time while maintaining high availability through pre-prepared resources.
Solution Approach 2:
The system implements a nested architecture where container images are embedded within a standardized format that can be efficiently transferred and executed across different computing environments. This nested structure allows the same container package to be rapidly deployed to remote nodes in either cluster, minimizing transfer overhead and enabling fast failover while maintaining application availability.
Data Source
AI summary
Methods, computer program products, and systems are presented. The method computer program products, and systems can include, for instance: receiving, by a manager node, from a plurality of compute nodes metrics data, the manager node and the plurality of compute nodes defining a first local cluster of a first computing environment, wherein nodes of the compute nodes defining the first local cluster have running thereon container based applications, wherein a first container based application runs on a first compute node of the plurality of compute nodes defining the first local cluster, and wherein a second compute node of the plurality of compute nodes defining the first local cluster runs a second container based application; wherein the manager node has received from an orchestrator availability data specifying a set of compute nodes available for hosting the first application.