Supervisor Cluster Control Plane for Containerized Scaling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing containerized application management systems are complex and resource-intensive, requiring administrator credentials and lacking in scalability and ease of deployment, especially in environments with high churn and utilization.
Innovation Solution
A low-footprint control plane system utilizing worker pods with fast on-demand deployment, customizable and deployable in a self-serve fashion, leveraging a Supervisor Cluster for Kubernetes management, with optimized resource allocation and minimalistic control plane components.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If traditional containerized application management systems are used, then applications can be deployed and managed, but the systems become complex and resource-intensive
Solution Approach 1:
The system segments the control plane into minimalistic components that run inside worker pods rather than as a separate complex infrastructure. Each worker pod contains only the essential control plane components needed for its function, reducing overall system complexity while maintaining deployment capabilities.
Solution Approach 2:
The control plane components are extracted from traditional separate control plane infrastructure and integrated inside worker pods. This extraction eliminates the need for complex dedicated control plane machinery and reduces resource consumption while maintaining application management functionality.
2Ease of operation
If traditional containerized application management systems are used, then applications can be deployed and managed, but administrator credentials and manual intervention are required
Solution Approach 1:
The system enables self-serve deployment where worker pods automatically discover and register themselves with the control plane without requiring administrator credentials or manual intervention. Pods can be deployed and managed autonomously through standardized APIs and service discovery mechanisms.
Solution Approach 2:
The control plane pre-configures worker pods with necessary credentials, certificates, and connection information before deployment. This preliminary setup allows pods to automatically authenticate and register with the control plane without requiring manual administrator intervention during deployment.
3Productivity
If traditional containerized application management systems are used, then applications can be deployed, but scalability and resource density are limited
Solution Approach 1:
The system merges control plane functionality with worker pods, eliminating the need for separate control plane infrastructure. This integration allows for more efficient resource utilization and faster scaling since control plane components are co-located with their workloads, reducing the overhead of resource allocation and communication.
Solution Approach 2:
The system changes the resource allocation parameters by using dynamic resource requests and limits in worker pods rather than static resource allocations. This allows the system to optimize resource density by allocating resources based on actual workload needs and enables faster scaling through dynamic pod creation and termination.
Data Source
AI summary
A computer-implemented method, computer-readable medium and computer system to execute containerized applications includes initiating a Supervisor Cluster on top of a SDDC to support execution of containerized applications. A supervisor cluster namespace is created on the Supervisor Cluster. A storage policy is attached to the supervisor cluster namespace. Then, a control plane is bootstrapped, and containerized applications are executed in a virtual machine cluster using vSphere pods as the worker nodes in the virtual machine cluster.