Auto-scaling Virtual Desktop Capacity via Latency Monitoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Public cloud environments face challenges in dynamically scaling virtual desktop capacity to match fluctuating demand, leading to poor user experience due to undersized resources and high costs from excess capacity, with existing tools lacking effective dynamic scaling and cost optimization capabilities.
Innovation Solution
The system implements an auto-scaling mechanism that dynamically adjusts virtual desktop capacity by powering on or off VMs and creating burst capacity on demand, using load balancing and user-defined triggers based on CPU usage, active sessions, available sessions, and user experience to optimize resource allocation and cost efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If allocated capacity is increased to meet peak demand, then user experience is improved, but operating costs increase due to paying for excess capacity during non-peak times
Solution Approach 1:
The patent implements dynamic capacity allocation that automatically adjusts virtual desktop resources based on real-time demand conditions. The system transitions from static capacity allocation to dynamic scaling, where capacity is increased during peak demand periods and decreased during non-peak periods, resolving the contradiction between maintaining reliable user experience and reducing operating costs.
Solution Approach 2:
The system changes the parameter of capacity allocation from fixed to variable based on demand conditions. By monitoring usage patterns and automatically adjusting the number of active virtual desktop instances, the system adapts capacity parameters to match actual needs, ensuring good user experience during peak times while minimizing costs during low-utilization periods.
2Loss of energy
If allocated capacity is decreased to reduce costs, then operating costs are reduced, but user experience deteriorates due to insufficient resources during peak demand
Solution Approach 1:
The system implements dynamic capacity adjustment that responds to real-time demand signals. Rather than maintaining fixed reduced capacity, the system automatically scales up resources when demand increases, ensuring user experience is maintained during peak periods while achieving cost savings during non-peak periods through reduced capacity allocation.
Solution Approach 2:
The patent incorporates feedback mechanisms that monitor user demand patterns, system performance metrics, and usage statistics. This feedback loop enables the system to detect when capacity reduction begins to impact user experience and automatically adjusts allocation accordingly, resolving the contradiction between cost reduction and service quality.
3Device complexity
If static capacity allocation is used to simplify management, then system complexity is reduced, but adaptability to fluctuating demand decreases
Solution Approach 1:
The patent implements self-service capacity management through automated scaling algorithms that monitor demand and adjust resources without manual intervention. The system autonomously performs capacity planning, provisioning, and optimization tasks, reducing the complexity burden on administrators while simultaneously improving adaptability to fluctuating demand through continuous automated adjustment.
Solution Approach 2:
The system transforms static capacity management into a dynamic automated process. By implementing real-time monitoring and automatic scaling capabilities, the system maintains simplicity of operation for users while achieving high adaptability to demand changes through algorithm-driven resource adjustment.
Data Source
AI summary
A system including: VMs selectively powered on and off, wherein the VMs store files on a storage volume whose size can be increased and decreased; and a controller that, in response to storage volume usage, adjusts the storage volume to ensure that it is no less than the current storage volume usage plus a minimum free space and no higher than a maximum storage volume size; monitors a latency; when the latency exceeds a high threshold for a given duration, increases the storage volume size until either the latency is below a high threshold or the storage volume size reaches a maximum storage volume size; and when the latency is below a low threshold for a given duration, reduces the storage volume size until either the latency is above the low threshold or the storage volume size is the current storage volume usage plus the minimum free space.