Hypervisor Multicast Proxy for Logical Network Traffic

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In virtualized computing environments, multicast packets are inefficiently handled as unknown unicast or broadcast packets, leading to resource wastage and network performance issues due to unnecessary traffic flooding, especially in applications like IP television and large file distribution.

Innovation Solution

Implementing a method where hypervisors in a logical network act as multicast proxies, leveraging multicast-enabled network devices to forward encapsulated multicast packets only to destinations that have joined the multicast group address, reducing unnecessary traffic by using multicast snooping and IGMP protocols to manage membership and forwarding.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If multicast packets are treated as unknown unicast or broadcast packets in logical networks, then compatibility with existing network infrastructure is maintained, but network performance deteriorates due to unnecessary traffic flooding and resource wastage

Engineering Contradiction:
ImprovecompatibilityVSAvoidnetwork performance
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent introduces a multicast proxy (hypervisor) as an intermediary between virtual machines and the physical network. This proxy translates virtual multicast traffic into efficient physical multicast packets, mediating between the logical network requirements and physical network capabilities, thereby resolving the contradiction between compatibility and performance

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent changes the parameter of packet handling mode from treating multicast as unicast/broadcast to implementing true multicast forwarding. By modifying how packets are processed (using multicast group addresses, membership tracking, and selective forwarding), the system achieves both compatibility and improved network performance

Inventive Principle:
Principle #35Parameter changes

2Device complexity

If multicast packets are forwarded to all hosts in the network, then simplicity of implementation is maintained, but resource wastage increases due to unnecessary traffic delivery to non-subscribers

Engineering Contradiction:
Improveforwarding complexityVSAvoidresource wastage
Core Design Contradiction:
Device complexityVSLoss of energy

Solution Approach 1:

The patent implements preliminary action by establishing multicast group membership relationships before multicast traffic is generated. The hypervisor tracks which virtual machines are members of which multicast groups in advance, enabling selective packet forwarding only to interested hosts, thereby reducing resource wastage without significantly increasing complexity

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent segments the network traffic handling by creating separate multicast forwarding paths for different multicast groups. Instead of a single broadcast mechanism, the system divides traffic into multiple targeted multicast streams, each delivered only to relevant subscribers, reducing overall resource consumption

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS10904134B2Multicast packet handling in logical networks
Publication Date: 2021.01.26 VMWARE INC
  • US10904134B2 patent drawing
  • US10904134B2 patent drawing
  • US10904134B2 patent drawing

AI summary

Example methods are provided for a host to perform multicast packet handling in a logical network. The method comprise in response to detecting a request to join a multicast group address, a hypervisor modifying the request by replacing a first address associated with a virtualized computing instance with a second address associated with the hypervisor; and sending the modified request to join the multicast group address on behalf of the virtualized computing instance. The method may also comprise: in response to detecting an egress multicast packet, the hypervisor encapsulating the egress multicast packet with an outer header that is addressed from the second address to the multicast group address; and sending the encapsulated egress multicast packet via one or more multicast-enabled network devices that are capable of forwarding, based on the outer header, the encapsulated egress multicast packet to one or more destinations that have joined the multicast group address.