Dynamic Virtual Resource Request Rate Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In virtualization environments, latency-sensitive workloads often experience increased latency due to being forced to wait behind throughput-sensitive workloads, leading to congestion and inconsistent performance in shared physical computer resources.
Innovation Solution
Implementing dynamic virtual resource request rate controls that introduce random delays between work requests based on the workload of the physical resource queue, ensuring fair sharing and consistent performance by adjusting the number of work requests submitted to the physical resource queue.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple virtual machines are hosted together on a single host to improve resource utilization, then productivity is improved, but latency-sensitive workloads experience increased latency due to contention with throughput-sensitive workloads
Solution Approach 1:
The patent implements dynamic rate control that adjusts the request rate for virtual machines based on real-time queue workload conditions. The rate control parameter is modified dynamically according to the current state of the physical resource queue, allowing the system to adapt between prioritizing throughput and reducing latency depending on workload conditions.
Solution Approach 2:
The system employs feedback mechanisms where the workload condition of the physical resource queue is continuously monitored and used to adjust the request rate for virtual machines. This closed-loop control ensures that latency-sensitive workloads receive appropriate rate control adjustments when queue conditions indicate potential latency issues.
2Device complexity
If request rate control is applied to manage queue workload, then congestion is reduced, but system complexity increases due to dynamic rate adjustment mechanisms
Solution Approach 1:
The patent changes the rate control parameter dynamically based on queue workload conditions. By adjusting this single key parameter according to system state, the patent achieves performance consistency without requiring complex control logic, maintaining relatively simple system architecture while ensuring reliable latency management.
Data Source
AI summary
A virtualization host may implement dynamic virtual resource request rate controls for physical resources. Individual virtual resource request queues may be maintained for different virtual compute instances implemented at a virtualization host for a particular physical computer resource. After placing a work request from one of the individual virtual resource request queues into a physical resource request queue to be performed at the physical computer resource, a delay may be dynamically determined based, at least in part, on the workload of the physical resource request queue. After imposing the delay, a next work request from the individual virtual resource request queue may be placed into the physical resource request queue. In at least some embodiments, the dynamically determined delay may include a randomly added delay.