CNF Scheduler Selection for Topology-Aware Node Placement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current scheduler systems lack topology awareness and context-awareness when deploying containerized network functions (CNFs), leading to inefficient resource utilization, malfunctions, and imbalances in computing clusters due to improper deployment based on default scheduling schemes.
Innovation Solution
Implementing a custom scheduler that considers the specific requirements and capabilities of computing nodes, using an orchestration system with machine learning models to identify optimal schedulers for deploying CNFs, ensuring they are placed on nodes with appropriate resources to meet performance, latency, and energy efficiency needs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a default scheduling scheme is used to deploy CNFs, then deployment simplicity is improved, but resource utilization efficiency deteriorates
Solution Approach 1:
The patent implements feedback mechanisms where the orchestration system continuously monitors node capabilities, CNF requirements, and deployment outcomes. This feedback loop enables the system to learn from past deployments and optimize future scheduling decisions, resolving the contradiction by maintaining simple deployment operations while improving resource utilization through data-driven insights
Solution Approach 2:
The system dynamically adjusts scheduling parameters based on changing node capabilities and CNF requirements. By modifying deployment parameters such as resource thresholds, affinity rules, and anti-affinity constraints based on real-time conditions, the system achieves both operational simplicity and high resource utilization without requiring complex manual configuration
2Manufacturing precision
If topology awareness is added to the scheduler, then deployment accuracy is improved, but system complexity deteriorates
Solution Approach 1:
The patent introduces an orchestration system as an intermediary layer between the scheduler and the underlying infrastructure. This intermediary handles the complexity of topology awareness and context analysis, while presenting a simplified scheduling interface. The orchestration system translates high-level deployment requirements into detailed scheduling decisions, achieving accurate topology-aware deployment without increasing scheduler complexity
Solution Approach 2:
The system segments the scheduling functionality into distinct modular components: node capability assessment, CNF requirement analysis, matching algorithms, and deployment execution. This segmentation allows each component to handle specific aspects of topology awareness independently, improving deployment accuracy while keeping individual component complexity manageable and enabling parallel processing
3Reliability
If context-awareness is implemented for CNF deployment, then deployment suitability is improved, but computational overhead deteriorates
Solution Approach 1:
The patent implements preliminary action by pre-assessing and caching node capabilities and CNF requirements before actual deployment decisions are made. The system pre-computes compatibility matrices and stores context information for quick retrieval during scheduling. This approach ensures deployment suitability through thorough context analysis while reducing computational overhead during the actual deployment process by avoiding redundant calculations
Solution Approach 2:
The system applies partial context-awareness selectively based on deployment priorities. For critical CNFs requiring high suitability, the system performs comprehensive context analysis. For less critical deployments, it uses simplified matching criteria. This selective approach balances deployment suitability with computational overhead, applying full context-awareness only where necessary rather than uniformly across all deployments
4Productivity
If custom schedulers are used for different CNF types, then deployment optimization is improved, but scheduler management complexity deteriorates
Solution Approach 1:
The patent implements a universal orchestration system that can handle multiple CNF types and deployment scenarios through a single unified interface. The system automatically selects and configures appropriate scheduling strategies based on the specific CNF requirements, eliminating the need for manual scheduler selection and management. This multi-functionality achieves deployment optimization for different CNF types while keeping scheduler management simple and centralized
Data Source
AI summary
In some implementations, an orchestration system may receive information regarding a containerized network function (CNF) to be deployed. The orchestration system may obtain deployment information regarding requirements for deploying the CNF, wherein the requirements include one or more of a computing requirement, a memory requirement, or a network connectivity requirement. The orchestration system may provide a request to obtain one or more recommendations regarding one or more schedulers configured to deploy CNFs associated with the requirements. The orchestration system may obtain the one or more recommendations regarding the one or more schedulers based on providing the request. The orchestration system may provide scheduler information regarding the one or more schedulers to a container management system, wherein the scheduler information is provided to cause the container management system to use the one or more schedulers to deploy the CNF on one or more computing nodes that comply with the requirements.