Infrastructure Driven Workload Scaling via Dynamic Resource Utilization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Data centers often experience unused capacity due to peak utilization challenges, leading to inefficiencies in resource allocation and compute density.
Innovation Solution
A method for scaling workloads by determining the quantity of virtual machines based on resource utilization, using a scaling agent to adjust the number of VMs and containerized workloads in response to changing resource demands, thereby optimizing resource usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If capacity is provisioned based on anticipated peak utilization, then reliability is improved, but productivity deteriorates due to unused capacity
Solution Approach 1:
The patent implements dynamic capacity provisioning that automatically adjusts resource allocation based on real-time workload demands. The system monitors utilization metrics and scales capacity up or down dynamically, transforming the static peak-based provisioning model into a dynamic adaptive model that matches actual usage patterns.
Solution Approach 2:
The system employs self-service automation where the infrastructure automatically detects utilization patterns and provisions or de-provisions capacity without human intervention. The intelligent system monitors its own performance metrics and autonomously adjusts resource allocation to maintain optimal utilization while ensuring reliability.
2Reliability
If virtual machines are over-provisioned to handle peak loads, then reliability is improved, but device complexity increases
Solution Approach 1:
The patent implements feedback mechanisms that continuously monitor infrastructure utilization and workload demands. The system uses this feedback to automatically adjust VM provisioning, replacing complex manual management with automated closed-loop control that simplifies infrastructure operations while maintaining peak load reliability.
Solution Approach 2:
The system introduces an intelligent intermediary layer that sits between the infrastructure and workloads, automatically managing VM provisioning and allocation. This intermediary handles the complexity of peak load management, shielding operators from intricate infrastructure decisions while ensuring reliable peak performance.
3Device complexity
If static capacity provisioning is used, then device complexity is reduced, but adaptability deteriorates
Solution Approach 1:
The patent transforms static capacity provisioning into a dynamic system that automatically adapts to changing workload patterns. The infrastructure continuously monitors utilization and adjusts resource allocation in real-time, providing flexibility and adaptability while maintaining manageable complexity through automation.
Data Source
AI summary
The disclosure provides a method for scaling workloads. The method includes receiving information regarding resources of one or more host machines running one or more virtual machines. The method further includes determining, based on the information, to change a quantity of the one or more virtual machines running on the one or more host machines. The method further includes determining an amount to change the quantity of the one or more virtual machines running on the one or more host machines based on utilization of one or more resource types of the one or more host machines, the utilization indicated by the information. The method further includes causing a change in the quantity of the one or more virtual machines running on the one or more host machines by the determined amount.