On-node DHCP Agent for Virtual Machine IP Assignment

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Centralized DHCP servers in distributed computing environments face reliability, scalability, and debugging challenges, particularly with increased server cluster sizes and virtual machine densities, as they can become overwhelmed and require complex physical network configurations.

Innovation Solution

Implementing on-node DHCP support by caching networking state on a network node, intercepting DHCP packets, and generating responses locally, thereby distributing DHCP load and improving reliability and troubleshooting capabilities.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If a centralized DHCP server is used to assign IP addresses to hardware servers and virtual machines, then IP address assignment can be managed centrally, but the DHCP server becomes overwhelmed with DHCP traffic and processing requirements as server cluster sizes and virtual machine densities increase

Engineering Contradiction:
Improvecentralized managementVSAvoidDHCP server capacity
Core Design Contradiction:
Ease of operationVSProductivity

Solution Approach 1:

The centralized DHCP server is segmented into distributed DHCP agents deployed on individual hardware servers. Each DHCP agent handles DHCP traffic locally for its host server and virtual machines, dividing the centralized workload into multiple distributed units that can process requests independently, thereby scaling with increasing server cluster sizes and virtual machine densities

Inventive Principle:
Principle #1Segmentation

2Ease of operation

If a centralized DHCP server is used, then IP address assignment is centralized, but reliability decreases because if the DHCP server malfunctions or is unavailable, all hardware servers and virtual machines served by this DHCP server are impacted

Engineering Contradiction:
Improvecentralized managementVSAvoidDHCP service availability
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The DHCP service is segmented from a single centralized server into multiple distributed DHCP agents deployed across hardware servers. Each agent provides DHCP functionality locally, creating redundant service points. If one DHCP agent or the centralized server fails, other agents continue to provide DHCP services to their respective hosts and virtual machines, maintaining overall service availability

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The DHCP agent acts as an intermediary between the centralized DHCP server and local virtual machines. It caches IP address assignments and networking state locally, allowing it to respond to DHCP requests even when the centralized server is unavailable, thereby improving reliability while maintaining centralized management capabilities

Inventive Principle:
Principle #24Intermediary (Mediator)

3Reliability

If DHCP packets are transmitted over the physical network with broadcast functionality, then DHCP requests can be reached, but physical network configuration becomes complicated requiring DHCP relay and end-to-end debugging of the DHCP channel

Engineering Contradiction:
ImproveDHCP packet deliveryVSAvoidphysical network configuration
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

DHCP traffic is extracted from the physical network broadcast domain and handled locally by DHCP agents on each hardware server. The agents intercept DHCP packets before they enter the physical network, process them using cached networking state, and respond locally. This eliminates the need for DHCP relay configuration and physical network broadcast functionality while maintaining reliable DHCP packet delivery

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

Each hardware server with its DHCP agent becomes self-sufficient in handling DHCP requests for its virtual machines. The DHCP agent caches networking state locally and uses this cached information to respond to DHCP requests without requiring external DHCP relay infrastructure or physical network broadcast capabilities, simplifying the overall network configuration

Inventive Principle:
Principle #25Self-service

Data Source

PatentEP3788772B1On-node DHCP implementation for virtual machines
Publication Date: 2023.08.16 MICROSOFT TECHNOLOGY LICENSING LLC
  • EP3788772B1 patent drawingFigure 1
  • EP3788772B1 patent drawingFigure 2
  • EP3788772B1 patent drawingFigure 3

AI summary

A DHCP server implementation includes transmission of a DHCP packet from a virtual machine executing on a server node to a node agent executing on the server node, generation, by the node agent, of a DHCP response packet based on the DHCP packet and on DHCP information previously stored in a local memory of the server node, and transmission of the DHCP response packet from the node agent to the virtual machine. Neither the DHCP packet transmitted by the virtual machine nor the DHCP response packet are transmitted out of the server node.