VCI Placement Using Physical NIC Queue Availability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing VCI placement techniques in SDN do not effectively consider physical NIC queue availability, leading to inefficiencies and errors, particularly for high-performance workloads that require dedicated PNIC queues and receive side scaling.
Innovation Solution
A resource optimization system that collects and processes PNIC queue availability information to determine optimal host placement for VCIs, taking into account dedicated queues, RSS engines, and queue sizes, ensuring high-performance network function virtualization capabilities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If VCI placement is based on processor and memory utilization only, then general load balancing is achieved, but network performance requirements for high-performance workloads are not met
Solution Approach 1:
The patent extends the VCI placement decision parameters from traditional CPU and memory utilization to include physical NIC queue availability, RSS engine availability, and queue sizes. This allows the placement system to adapt to specific network performance requirements of different workload types while maintaining compatibility with existing load balancing mechanisms.
2Productivity
If VCIs are placed without considering PNIC queue availability, then placement simplicity is maintained, but high-performance workloads experience network bottlenecks
Solution Approach 1:
The patent collects and stores PNIC queue availability information, RSS engine availability, and queue sizes in advance before VCI placement decisions are made. This preliminary gathering of network resource information allows the placement system to make informed decisions without adding complexity to the actual placement process, ensuring high-performance workloads are assigned to hosts with adequate network resources.
3Reliability
If hosts without adequate PNIC queue resources are selected for VCI placement, then resource utilization is maximized, but network function virtualization performance deteriorates
Solution Approach 1:
The patent applies different placement criteria based on the specific network performance requirements of individual VCIs. High-performance workloads requiring dedicated PNIC queues and RSS engines are placed on hosts with adequate network resources, while less demanding workloads can be placed on hosts with lower network resource availability. This localized quality approach ensures each VCI receives the network resources it specifically needs.
Data Source
AI summary
The disclosure provides an approach for virtual computing instance (VCI) placement. Embodiments include receiving, by a resource optimization system, physical network interface (NIC) queue availability information relating to a plurality of host computers. Embodiments include determining, by the resource optimization system, physical NIC queue requirements of a VCI. Embodiments include selecting, by the resource optimization system, a target host computer for the VCI from the plurality of host computers based on the physical NIC queue availability information and the physical NIC queue requirements of the VCI. Embodiments include loading, by the resource optimization system, the VCI on the target host computer.