VM Packet Routing Bypassing Virtual Switch Bottleneck
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing communications systems face inefficiencies in packet exchange between virtual machines (VMs) due to latency and throughput limitations caused by the virtual switch, which can lead to degraded performance and increased resource consumption.
Innovation Solution
VMs can cooperate to directly transfer packets between each other, bypassing the virtual switch and virtual network interfaces, using a subset of routing rules to determine the next hop and employing shared memory or a transfer controller for data transfer, thereby reducing latency and resource usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If packets are transferred through the virtual switch between VMs, then routing control and network management are simplified, but latency increases and throughput decreases
Solution Approach 1:
The patent segments packet transfer paths into two types: standard paths through the virtual switch for general traffic, and direct paths between VMs for time-sensitive traffic. This segmentation allows the system to optimize for both management simplicity and low latency by choosing the appropriate path based on traffic requirements.
Solution Approach 2:
The virtual switch acts as an intermediary for standard packet routing, providing centralized control and management. However, for direct transfers, the patent bypasses this intermediary to eliminate the latency it introduces, while still maintaining virtual switch control over routing decisions through software configuration.
2Extent of automation
If packets are transferred through the virtual switch between VMs, then centralized routing control is maintained, but throughput is limited by the virtual switch bottleneck
Solution Approach 1:
The patent divides packet flow into segments that can be handled differently: some segments go through the virtual switch for centralized control, while other segments use direct VM-to-VM paths to bypass the throughput bottleneck. This allows simultaneous achievement of centralized control where needed and high throughput where possible.
Solution Approach 2:
Instead of requiring all packet transfers to go through the virtual switch, the patent applies partial action by allowing only necessary packet segments to use the controlled path, while other segments use direct transfers. This partial bypass maintains essential control while eliminating unnecessary bottlenecks.
3Loss of time
If direct packet transfer between VMs is implemented, then latency is reduced and throughput is increased, but device complexity increases
Solution Approach 1:
The patent introduces a routing decision mechanism that acts as an intermediary to determine whether packets should use direct or virtual switch paths. This centralized decision-making layer manages the complexity of direct transfers without requiring each VM to implement complex transfer logic, thus reducing overall system complexity while maintaining performance benefits.
4Productivity
If direct packet transfer between VMs is implemented, then packet throughput is increased, but resource consumption management becomes more complex
Solution Approach 1:
The patent implements feedback mechanisms where the virtual switch monitors resource usage and packet flow patterns, then dynamically adjusts routing decisions. This feedback loop allows the system to optimize throughput while managing resources effectively, as the virtual switch can identify when direct transfers are beneficial and when virtual switch routing is more appropriate for resource management.
Data Source
Figure 1
Figure 2A
Figure 2B
AI summary
Various embodiments are generally directed to techniques for improving the efficiency of exchanging packets between pairs of VMs within a communications server. An apparatus may include a processor component; a network interface to couple the processor component to a network; a virtual switch to analyze contents of at least one packet of a set of packets to be exchanged between endpoint devices through the network and the communications server, and to route the set of packets through one or more virtual servers of multiple virtual servers based on the contents; and a transfer component of a first virtual server of the multiple virtual servers to determine whether to route the set of packets to the virtual switch or to transfer the set of packets to a second virtual server of the multiple virtual servers in a manner that bypasses the virtual switch based on a routing rule.