Zonal Resource Scaling Using Level-Strength Scheduling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing container orchestration systems lack an efficient scheduler that ensures high availability of software applications across multiple zones and databases, necessitating a need for improved scheduling techniques to manage resources effectively.
Innovation Solution
Implementing a level-strength-based scheduling method using a zonal scheduling unit within a container orchestration engine, which assesses the strength of each level in a zonal topology to determine optimal resource allocation and scaling across zones, ensuring high availability and non-deterministic behavior.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional scheduling methods are used in container orchestration systems, then deployment simplicity is maintained, but high availability across multiple zones and databases cannot be ensured
Solution Approach 1:
The patent segments the scheduling problem into hierarchical levels (zone level, database level, container level) and introduces a multi-level scheduling architecture. The system divides the cluster into zones, then databases within zones, and finally schedules containers within databases. This segmentation allows the system to ensure high availability at each level independently, resolving the contradiction between reliability and complexity by making the scheduling process manageable through structured division.
Solution Approach 2:
The patent introduces a new dimensional approach by adding zonal awareness to traditional container scheduling. Instead of flat scheduling, it creates a multi-dimensional scheduling space considering zone location, database identity, and container requirements simultaneously. This dimensional expansion enables the system to achieve high availability across zones while maintaining scheduling control through structured multi-criteria evaluation.
2Reliability
If resources are scaled across multiple zones to ensure high availability, then fault tolerance is improved, but resource management complexity increases
Solution Approach 1:
The patent implements self-service mechanisms where the scheduling system automatically performs zone-aware resource allocation, scaling decisions, and load balancing without manual intervention. The multi-level scheduler autonomously evaluates zone capacity, database requirements, and container priorities to make scheduling decisions. This automation maintains fault tolerance across zones while simplifying resource management operations, as the system self-regulates rather than requiring complex manual coordination.
Solution Approach 2:
The system incorporates feedback mechanisms that continuously monitor zone health, resource utilization, and container performance across the distributed system. This feedback enables dynamic adjustment of scheduling decisions, allowing the system to maintain fault tolerance while adapting to changing conditions automatically. The feedback loop simplifies resource management by providing real-time visibility and automatic response to zone or database issues without requiring complex manual intervention.
3Productivity
If level-strength-based scheduling is implemented to assess zone topology, then optimal resource allocation is achieved, but computing overhead increases
Solution Approach 1:
The patent applies preliminary action by pre-computing and caching zone topology information, database mappings, and strength metrics before scheduling decisions are needed. The system maintains updated zone topologies and pre-evaluates zone strengths based on capacity, load, and health metrics. When scheduling decisions are required, the system leverages these pre-computed data structures rather than calculating everything from scratch, significantly reducing the computing energy required for real-time scheduling while maintaining optimal resource allocation.
Data Source
AI summary
In some embodiments, a method includes receiving zonal topology information related to a zonal topology of a plurality of zones; utilizing the zonal topology information to perform a level strength assessment of each level of a plurality of levels associated with the zonal topology of the plurality of zones; and based on the level strength assessment of each level of the plurality of levels, scaling a target number of resources to at least a first level of the plurality of levels of the zonal topology. In some embodiments of the method, the level strength assessment includes performing a level-by-level breadth analysis of each level of the plurality of levels of the zonal topology.