Virtual Ethernet Bridge for Low-Latency VM Traffic Routing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional virtual machine networking solutions face challenges in balancing latency, throughput, and resource utilization, as they often require external network switches that increase latency and consume more network bandwidth, while also being costly and complex to implement directly in network adapters.
Innovation Solution
A system and method for virtual machine networking that employs an Entity for Flexible Virtual Machine Communications (EFVC) within a network adapter to determine the optimal path for traffic handling based on traffic characteristics, available resources, and management information, allowing for internal or external processing of traffic, and utilizing an external network device to support additional features and management functions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If virtual machine traffic is handled via external network switches, then network management and control functions are improved, but latency and network bandwidth consumption increase
Solution Approach 1:
The patent segments traffic handling into two paths: simple inter-VM traffic is handled internally by the VEB within the network adapter (bypassing external switches), while complex traffic requiring management functions is forwarded externally. This segmentation resolves the contradiction by allowing latency-sensitive traffic to be processed locally while still enabling external management when needed.
Solution Approach 2:
The VEB acts as an intermediary between VMs and external network switches, providing intelligent routing decisions. It mediates by determining which traffic should be handled internally versus externally, thus reducing unnecessary external switch involvement and latency while preserving management capabilities for appropriate traffic types.
2Speed
If virtual machine networking functionality is implemented directly in network adapters, then latency is reduced and throughput is improved, but device complexity and cost increase
Solution Approach 1:
The VEB provides multi-functionality by combining virtual Ethernet bridging, MAC address learning, and intelligent traffic routing within the network adapter. This allows a single device to handle both simple switching and complex routing decisions, improving throughput without proportionally increasing complexity compared to implementing all functions externally.
Solution Approach 2:
The VEB dynamically adapts its behavior based on traffic characteristics and configuration, switching between internal handling and external forwarding modes. This dynamic operation allows the system to optimize performance for different traffic types while maintaining manageable complexity through software-configurable behavior.
3Loss of time
If all virtual machine traffic is processed internally within the network adapter, then latency is reduced, but network management capabilities and external resource utilization are limited
Solution Approach 1:
The VEB dynamically determines the optimal processing path for each traffic flow based on characteristics such as source/destination MAC addresses, VLAN tags, and configured policies. This dynamic decision-making enables low-latency internal processing for appropriate traffic while maintaining adaptability to leverage external management resources when needed.
Solution Approach 2:
The system changes operational parameters (internal vs. external routing) based on traffic conditions and configuration. By adjusting these parameters dynamically, the VEB achieves low latency for time-sensitive traffic while preserving network management capabilities for traffic requiring external processing.
Data Source
AI summary
Aspects of a method and system for networking are provided. In this regard, one or more circuits and/or processors in a network adapter of a first network device may determine whether to communicate traffic between virtual machines running on the first network device via a path that resides solely within the first network device, or via a path that comprises a second network device that is external to the first network device. The determination may be based, at least in part, on characteristics of the traffic. The determination may be based, at least in part, on capabilities and/or available resources of the network adapter. The determination may be based, at least in part, on management information exchanged between the one or more circuits and/or processors and one or more of: software running on the first network device, the second network device, and a third network device.