Cloud Job Scheduling With Mixed Deployment for Service Availability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In current cloud computing platforms, the centralized deployment of service copies on the same resource leads to poor service availability due to aggravation of implementation units.
Innovation Solution
Implement mixed-deployment of implementation units of multiple to-be-scheduled jobs in one implementation combination, reducing the aggravation degree and improving service dispersion, ensuring high availability through a job scheduling method that includes obtaining and scheduling implementation units based on resource demands and idle resource specifications.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If centralized deployment of service copies is used on the same resource, then resource utilization is simplified, but service availability deteriorates due to aggravation of implementation units
Solution Approach 1:
The patent segments implementation units of the same job into different deployment groups. Each group is deployed on different resource nodes, preventing all copies from being concentrated on a single resource. This segmentation reduces the aggravation effect and improves service availability while maintaining manageable deployment complexity through automated group management.
Solution Approach 2:
The patent applies local quality by differentiating deployment strategies for different groups of implementation units. Each deployment group is assigned to specific resource nodes based on local resource characteristics and job requirements. This allows optimized placement that balances resource utilization with availability requirements, addressing the contradiction between simplified deployment and service reliability.
2Reliability
If implementation units of multiple jobs are deployed separately, then service isolation is improved, but resource utilization efficiency deteriorates
Solution Approach 1:
The patent merges implementation units from different jobs into shared resource pools while maintaining logical separation through deployment groups. Multiple jobs can share the same underlying resource infrastructure, improving resource utilization efficiency. Service isolation is preserved through group-level management and controlled access, resolving the contradiction between isolation and efficiency.
Solution Approach 2:
The patent creates universal resource pools that can serve multiple jobs simultaneously. The same resource nodes can host implementation units from different jobs through the group deployment mechanism, enabling multi-functionality. This improves resource utilization while maintaining service isolation through proper group management and resource allocation policies.
3Reliability
If mixed-deployment of implementation units is implemented, then service dispersion is improved, but deployment complexity increases
Solution Approach 1:
The patent introduces dynamic deployment group management that automatically adjusts the distribution of implementation units across resource nodes. The system dynamically balances service dispersion requirements with deployment complexity by using automated algorithms to manage group assignments. This allows improved service dispersion without proportionally increasing manual deployment complexity, as the system self-manages the complexity through automation.
Data Source
AI summary
A job scheduling method including: obtaining a plurality of implementation combinations based on resource demands of implementation units that need to be deployed for a plurality of to-be-scheduled jobs, where at least a part of the plurality of implementation combinations include implementation units of at least two to-be-scheduled jobs, and a first proportion of a resource demand of the implementation combination to an idle resource specification of a resource node in a resource cluster is greater than a preset value; obtaining a target implementation set based on the plurality of implementation combinations, where the target implementation set includes at least one implementation combination, and includes all implementation units of the plurality of to-be-scheduled jobs; and scheduling, to the resource node based on the target implementation set, the implementation units that need to be deployed for the plurality of to-be-scheduled jobs.