Unified Resource Scheduling Coordinator for VM and Container Integration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current cloud computing solutions separate the scheduling of physical resource pools for virtual machines and containers, leading to inefficiencies in network and storage sharing, and complex management, as they operate independently, preventing the reuse of resources and causing bottlenecks.
Innovation Solution
A unified resource scheduling coordinator method that selects a computing node based on resource requirements and utilization mode, allowing virtual machines and containers to share the same physical resources, enabling integrated scheduling and management through a nearest window query method and resource allocation across NUMA nodes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If virtual machines and containers are managed by separate platforms (OpenStack and K8S) with independent physical resource pools, then each platform can be managed independently, but resource reuse is prevented and cross-domain bottlenecks occur in network and storage sharing
Solution Approach 1:
The patent merges the previously separate physical resource pools for virtual machines and containers into a unified resource pool. The scheduling system integrates VM and container resource allocation, allowing both workloads to share the same physical infrastructure (computing, storage, network resources), thereby eliminating resource silos and improving overall resource utilization efficiency
Solution Approach 2:
The patent creates a universal scheduling system that can allocate resources to both virtual machines and containers from a single resource pool. The system provides multi-functional capabilities by supporting heterogeneous workload types (VMs and containers) on the same physical infrastructure, enabling resources to serve multiple purposes and user types
2Device complexity
If separate physical resource pools are used for virtual machines and containers, then management can be simplified for each individual platform, but network and storage sharing becomes complex and creates bottlenecks
Solution Approach 1:
The patent combines separate network and storage resource pools into unified shared resources. The system implements unified network resource management and unified storage resource management, allowing both virtual machines and containers to access the same network and storage infrastructure, thereby eliminating cross-domain bottlenecks and improving sharing efficiency
3Adaptability or versatility
If independent resource pools are maintained for virtual machines and containers, then platform-specific management is easier, but resource fragmentation increases and utilization decreases
Solution Approach 1:
The patent implements a universal resource pool that can dynamically allocate resources to both virtual machines and containers based on demand. The system maintains adaptability by supporting different workload types while eliminating fragmentation through unified resource management, allowing resources to be flexibly assigned to the most appropriate workload at any given time
Data Source
AI summary
A method for creating a virtual machine and/or a container through a unified resource scheduling coordinator, a unified resource scheduling coordinator, a unified resource scheduling system and a computer storage medium. The method includes: receiving an instruction to create the virtual machine and/or the container, wherein the instruction comprises parameters of the virtual machine and/or the container; selecting a first computing node from physical nodes according to the parameters; transmitting the parameters and information of the first computing node to a virtual machine management module and/or a container management module depending on whether the virtual machine or the container is to be created; receiving resource allocation condition from the virtual machine management module and/or the container management module after creation of the virtual machine and/or the container; notifying the resource allocation condition to the container management module and/or the virtual machine management module after recording of resource change condition.