Hybrid Virtual Machine Configuration Management via Priority-Based Resource Allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In shared resource pool environments, virtual machines often experience reduced affinity levels to physical resources, leading to decreased performance due to the sharing of physical resources among multiple logical partitions (LPARs), which can result in suboptimal allocation and utilization of resources.
Innovation Solution
Implementing a prioritization scheme where entitled virtual resources are assigned a higher priority than uncapped excess resources, with physical resources having a desired affinity level being mapped and preferentially allocated to high-priority virtual resources, ensuring better affinity and performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If physical resources are shared among multiple LPARs in a shared resource pool environment, then resource utilization efficiency is improved, but affinity levels to physical resources decrease leading to reduced performance
Solution Approach 1:
The patent segments virtual resources into two distinct priority categories: entitled virtual resources (high priority) and uncapped excess virtual resources (low priority). This segmentation allows the system to differentiate between guaranteed resources that require stable affinity and excess resources that can tolerate dynamic allocation, thereby resolving the contradiction between shared resource efficiency and affinity maintenance.
Solution Approach 2:
The patent applies local quality by assigning different affinity requirements to different portions of virtual resources. Entitled virtual resources are mapped to physical resources with high affinity relationships maintained, while uncapped excess resources are allocated from the shared pool without strict affinity constraints. This localized differentiation resolves the contradiction by applying appropriate affinity management only where necessary.
2Reliability
If entitled virtual resources are assigned higher priority than uncapped excess virtual resources, then affinity levels to physical resources are maintained, but resource allocation complexity increases
Solution Approach 1:
The patent implements preliminary action by pre-assigning priority levels to virtual resources based on their entitlement status before allocation occurs. The hypervisor identifies entitled versus uncapped resources and assigns high or low priority accordingly, establishing the allocation hierarchy in advance. This preliminary classification simplifies the allocation process by providing clear priority guidance rather than requiring complex real-time decisions.
Solution Approach 2:
The system employs self-service mechanisms where the hypervisor automatically manages the priority-based allocation process without requiring manual intervention. The hypervisor monitors resource usage, identifies when entitled resources need affinity-preserving allocation, and automatically allocates from the shared pool according to priority rules, reducing operational complexity while maintaining affinity levels.
3Adaptability or versatility
If physical resources are allocated from a shared pool to meet excess resource demands, then resource flexibility is improved, but affinity levels to physical resources are reduced
Solution Approach 1:
The patent applies partial action by allocating shared pool resources only to the extent necessary for uncapped excess virtual resources, while preserving affinity relationships for entitled resources. The hypervisor selectively draws from the shared pool only when allocating low-priority uncapped resources, leaving high-priority entitled resource allocations to maintain stable affinity relationships. This partial application of shared pooling resolves the contradiction by limiting affinity disruption to only where excess capacity is actually needed.
Data Source
AI summary
According to one aspect of the present disclosure, a method and technique for hybrid virtual machine configuration management is disclosed. The method includes: assigning to a first set of virtual resources associated with entitled resources of a virtual machine a first priority; assigning to a second set of virtual resources associated with the virtual machine a second priority lower than the first priority, wherein the first and seconds sets when combined exceed the entitled resources for the virtual machine; mapping the first set of virtual resources to a first physical resource of a pool of shared physical resources allocatable to the first and second sets of virtual resources, wherein the first physical resource comprises a desired affinity level to a second physical resource allocated to the virtual machine; and preferentially allocating the first physical resource to the first set of virtual resources.