vNIC Failover State Migration Between DPUs in Pass-Through Mode

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In servers with multiple DPUs, vNICs in pass-through mode become non-functional when an active DPU fails, as they are not synchronized with the standby DPU, leading to communication disruptions.

Innovation Solution

A method for migrating the state of vNICs from a failing DPU to a standby DPU, involving the hypervisor acquiring and transmitting state information to ensure seamless communication, including resetting state information in cases where direct communication is impossible.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If vNICs use pass-through mode for direct communication with DPU, then data path latency is reduced, but reliability deteriorates when DPU fails because vNICs are not synchronized with standby DPU

Engineering Contradiction:
Improvedata path latencyVSAvoidcommunication continuity
Core Design Contradiction:
SpeedVSReliability

Solution Approach 1:

The system performs preliminary actions by maintaining synchronized state information between active and standby DPUs before failure occurs. The standby DPU is pre-configured with vNIC state information including memory locations and processing parameters, enabling immediate takeover without loss of functionality when the active DPU fails.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system creates and maintains copies of vNIC state information in both the active DPU and standby DPU. This copying mechanism ensures that when the active DPU fails, the standby DPU already possesses the necessary state information to continue processing network packets without interruption, thus maintaining reliability while using pass-through mode.

Inventive Principle:
Principle #26Copying

2Productivity

If vNICs communicate directly with DPU in pass-through mode, then processing efficiency is improved, but adaptability deteriorates when switching between DPUs

Engineering Contradiction:
Improveprocessing efficiencyVSAvoidDPU switchability
Core Design Contradiction:
ProductivityVSAdaptability or versatility

Solution Approach 1:

The system implements universality by designing the DPU state information structure to be standardized and interchangeable between different DPUs. The standby DPU is configured with universal state information that allows it to assume the role of the active DPU, enabling seamless switching while maintaining direct communication efficiency.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system enables adaptability through parameter changes by updating the DPU state information parameters when switching between active and standby DPUs. The state information includes configurable parameters such as memory locations and processing rules that can be dynamically adjusted to match the new active DPU's configuration, maintaining processing efficiency across DPU transitions.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP4647907A1Failover migration for virtual network interface controllers between data processing units
Publication Date: 2025.11.12 VMWARE INC
  • EP4647907A1 patent drawingFigure 1
  • EP4647907A1 patent drawingFigure 2A
  • EP4647907A1 patent drawingFigure 2B

AI summary

A method of migrating states of virtual network interface controllers (vNICs) of virtual computing instances between data processing units (DPUs), includes the steps of: in response to detecting a failure in a first DPU, transmitting a request to the first DPU for a state of a first vNIC of a first virtual computing instance, wherein the state of the first vNIC includes memory locations at which network packets of the first vNIC are to be stored by one of the first vNIC and a first virtual function (VF) of the first DPU for further processing by the other; and transmitting the state of the first vNIC, received from the first DPU in response to the request, to a second DPU, and instructing a second VF of the second DPU to store and process network packets of the first vNIC based on the state of the first vNIC.