Virtualized Resource Overallocation Prevention
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In virtualized computing environments, existing approaches fail to prevent resource overallocation effectively, as they do not account for powered-off virtual computing instances (VCIs) when allocating resources, leading to situations where not all VCIs can be turned on due to limited resource availability.
Innovation Solution
A system and method that collect data from management platforms to determine the actual available resources by factoring in powered-off VCIs, allowing users to specify preferences for preventing overallocation, enabling all VCIs to be turned on by assigning resources irrespective of their power state, and allowing for global or per-cluster configuration of overallocation preferences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If resources are allocated without considering powered-off VCIs, then resource allocation simplicity is maintained, but resource overallocation occurs leading to inability to turn on all VCIs
Solution Approach 1:
The system performs preliminary identification of powered-off VCIs before resource allocation. By detecting the power state of VCIs in advance and excluding their resources from the available pool, the system prevents overallocation before it occurs, ensuring accurate resource tracking without complex runtime adjustments
Solution Approach 2:
The system implements feedback mechanisms by continuously monitoring VCI power states and dynamically adjusting the available resource pool accordingly. When VCIs are powered off, their resources are automatically added back to the available pool; when powered on, resources are deducted. This closed-loop feedback ensures accurate resource allocation while maintaining manageable complexity through automated updates
2Productivity
If all VCIs are allowed to be turned on simultaneously, then resource utilization is maximized, but resource overallocation occurs
Solution Approach 1:
Before allowing VCI activation, the system preliminarily calculates the total resource requirements of all VCIs and compares it against the actual available resources (excluding powered-off VCIs). This preliminary check prevents overallocation by identifying potential resource shortages before VCIs are turned on, ensuring that resource quantity constraints are respected while maximizing productive utilization
3Measurement precision
If resource allocation does not account for powered-off VCIs, then allocation speed is maintained, but measurement precision of available resources deteriorates
Solution Approach 1:
The system performs preliminary identification and categorization of VCIs by power state before resource allocation. By pre-computing which VCIs are powered off and excluding their resources from the available pool, the system achieves precise measurement of available resources without requiring time-consuming runtime checks during the allocation process
Solution Approach 2:
The system implements self-service by automatically monitoring VCI power states and dynamically updating the available resource pool without manual intervention. The resource management system continuously queries VCI states and adjusts available resources in real-time, ensuring measurement precision is maintained automatically without adding time overhead to the allocation process
Data Source
AI summary
The present disclosure is related to devices, systems, and methods for selectively preventing resource overallocation in a virtualized computing environment. One example includes instructions to receive a request to prevent overallocation of a resource in a software-defined datacenter associated with a customer, determine an amount of the resource available to the customer, and assign a respective portion of the amount of the resource available to the customer to each of a plurality of virtual computing instances (VCIs) irrespective of a power state of each of the plurality of VCIs.