Virtual Machine Host-Group Mapping via Distributed Resource Scheduler
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manual mapping of host computers into host-groups in virtual machine clusters is inefficient and slow, relying heavily on administrative expertise, making it difficult to adapt to fluctuating resource demands.
Innovation Solution
A system that automatically determines if a virtual machine entity needs additional resources and maps an available host computer to the associated host-group, and reverses this mapping when resources are no longer needed, using a distributed resource scheduler to manage host computers and maintain desired configurations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If manual mapping of host computers to host-groups is used, then administrative control and precision are improved, but efficiency and responsiveness to demand fluctuations deteriorate
Solution Approach 1:
The system enables self-service by allowing the distributed resource scheduler to automatically perform host computer mapping to host-groups based on resource demand, eliminating the need for manual administrative intervention. The scheduler monitors resource requirements and autonomously maps host computers to appropriate host-groups, achieving both precision through automated decision logic and efficiency through elimination of manual processes.
Solution Approach 2:
The patent replaces the mechanical manual mapping process with an automated software-based system. The distributed resource scheduler uses algorithmic logic to determine optimal host computer mappings, substituting human administrative actions with automated computational processes that can rapidly respond to changing resource demands while maintaining precise control over host-group assignments.
2Ease of operation
If manual mapping by administrators is used, then expertise-based control is improved, but speed and adaptability to demand fluctuations deteriorate
Solution Approach 1:
The system implements feedback mechanisms where the distributed resource scheduler continuously monitors resource demand and host computer performance metrics. Based on this feedback, the scheduler dynamically adjusts host computer mappings to host-groups, ensuring high-quality control through data-driven decisions while achieving rapid adaptation to demand fluctuations. The feedback loop enables the system to respond automatically to changing conditions without manual intervention.
Solution Approach 2:
The patent introduces dynamics by enabling the host computer mapping system to adapt in real-time to changing resource demands. Instead of static manual mappings, the distributed resource scheduler dynamically reconfigures host-group assignments based on current system conditions, allowing the system to flexibly respond to demand fluctuations while maintaining operational control through automated policy enforcement.
3Stability of the object's composition
If static host-group creation is used, then system stability is improved, but adaptability to resource demand changes deteriorates
Solution Approach 1:
The system resolves the contradiction between stability and adaptability by making host-group mappings dynamic rather than static. The distributed resource scheduler maintains stable host-group definitions but dynamically assigns host computers to these groups based on real-time resource demand. This approach preserves the structural stability of host-groups while enabling flexible adaptation to changing conditions through automated reassignment of host computers.
Solution Approach 2:
The patent applies preliminary action by pre-defining host-groups with their target resource characteristics and capacity thresholds. These pre-configured host-groups serve as stable templates that guide dynamic mapping decisions. When resource demand changes, the scheduler refers to these pre-established group definitions to determine appropriate mappings, combining the stability of predefined structures with the adaptability of dynamic assignment based on actual resource needs.
Data Source
AI summary
Embodiments of a non-transitory computer-readable storage medium and a computer system are disclosed. In an embodiment, a non-transitory computer-readable storage medium containing program instructions for managing host computers that run virtual machines into host-groups within a cluster is disclosed. When executed, the instructions cause one or more processors to perform steps including determining if a virtual machine entity needs additional resources and, if the virtual machine entity needs additional resources, mapping a host computer to a host-group with which the virtual machine entity is associated.


