Task Orchestration via Auxiliary Container File System Sharing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current cloud native container orchestration engines, such as Kubernetes, fail to decouple task orchestration from resource orchestration, do not support container program hot updates, and lack support for data delivery tasks, limiting their ability to manage complex orchestration and scheduling needs, especially for large-scale applications.
Innovation Solution
A method and system for task orchestration that involves a task master control unit providing execution instructions to a node agent service unit in an auxiliary application container, which shares a file system with the main application container, allowing for decoupled resource management, hot updates, and unified scheduling of various task types, including deployment, timed, and data delivery tasks, while ensuring fault-tolerance and high concurrent performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If cloud native container orchestration engine (e.g., Kubernetes) is used for task orchestration, then resource orchestration and task orchestration are coupled together, but this fails to decouple resource management and task orchestration management
Solution Approach 1:
The system divides the orchestration functionality into two independent components: a container orchestration engine for resource management and a task orchestration system for task management. The task orchestration system includes a task master control unit, auxiliary application containers with node agent service units, which operate independently from the container orchestration engine, achieving decoupling of resource and task orchestration.
2Productivity
If traditional container orchestration engine is used, then basic container deployment is supported, but container program hot update is not supported
Solution Approach 1:
The system prepares update packages in advance and stores them in the auxiliary application container's file system. When a hot update is needed, the node agent service unit directly applies these pre-prepared updates without requiring container recreation, enabling seamless hot updates while maintaining deployment efficiency.
3Ease of operation
If cloud native container orchestration engine is used, then container management is provided, but data delivery tasks are not supported
Solution Approach 1:
The task orchestration system implements a universal task execution framework that can handle multiple task types including deployment tasks, timed tasks, data delivery tasks, and one-off command tasks. The node agent service unit executes different types of tasks through a unified interface, enabling the system to perform both container management and data delivery operations.
4Device complexity
If container granularity update is implemented, then basic update functionality is achieved, but module granularity hot change is not supported
Solution Approach 1:
The system implements module-granularity hot change by allowing different modules within an application to be updated independently. The auxiliary application container maintains a file system that supports individual module replacement, and the node agent service unit can apply updates to specific modules without affecting other parts of the application, enabling fine-grained hot changes.
Data Source
AI summary
Embodiments of the disclosure provide a method and system for task orchestration. A method may include: providing, by a task master control unit, an execution instruction of a task related to a module in an application container to a node agent service unit in an auxiliary application container bound to the application container, the auxiliary application container sharing a file system with the application container; and executing, by the node agent service unit, a command for completing the task, in response to acquiring the execution instruction of the task.