HCI Node Management via DHCP Static IP Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Hyper-converged infrastructure (HCI) systems face complexity in managing node configurations, especially in distinguishing between cluster and free nodes, and determining the need for re-imaging, which hinders efficient cluster management and resource allocation.

Innovation Solution

Assigning two static IP addresses, one out-of-band and one internal management IP, to each HCI node, allowing access to the baseboard management controller to retrieve cluster configuration information, and generating a dashboard interface to display node and cluster details, enabling administrators to manage clusters, create new clusters, and reimage nodes as needed.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If HCI systems use virtualized compute, storage, and network resources with centralized management, then resource efficiency and flexibility are improved, but configuration complexity and difficulty in distinguishing node types increase

Engineering Contradiction:
Improveresource flexibilityVSAvoidconfiguration complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent segments node identification by assigning distinct IP address types (OOB and internal management IPs) to different node categories (cluster nodes vs. free nodes, fresh nodes vs. dirty nodes). This segmentation allows administrators to clearly distinguish between node types through the dashboard interface, reducing configuration complexity while maintaining the flexibility of virtualized resources.

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If administrators manually manage each node configuration, then control precision is improved, but time consumption and operational burden increase

Engineering Contradiction:
Improveconfiguration accuracyVSAvoidmanagement time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system performs self-identification and self-display of node configuration status through the automated dashboard. Each node automatically provides its own configuration information (cluster assignment, fresh/dirty status) which is then displayed in the dashboard, eliminating the need for administrators to manually query and manage each node while maintaining precise configuration awareness.

Inventive Principle:
Principle #25Self-service

3Loss of information

If the system displays detailed node configuration information, then information completeness is improved, but interface complexity and ease of use decrease

Engineering Contradiction:
Improveconfiguration information completenessVSAvoidinterface usability
Core Design Contradiction:
Loss of informationVSEase of operation

Solution Approach 1:

The dashboard interface presents configuration information with differentiated local quality - cluster nodes are highlighted with distinct visual indicators compared to free nodes, and fresh nodes are distinguished from dirty nodes. This localized visual differentiation provides complete configuration information while maintaining interface simplicity and ease of use through intuitive visual coding.

Inventive Principle:
Principle #3Local quality

Data Source

PatentUS12141598B2Managing hyper-converged infrastructure (HCI) appliance nodes and clusters with DHCP-based resource
Publication Date: 2024.11.12 DELL PROD LP
  • US12141598B2 patent drawing
  • US12141598B2 patent drawing
  • US12141598B2 patent drawing

AI summary

A disclosed method may assign, with a DHCP module, two static IP addresses to each of one or more hyper-converged infrastructures nodes. The two static IP addresses may include an out of band IP address for a baseboard management controller (BMC) network interface to an OOB management network and an internal management IP address for a network interface associated with an internal management network. Disclosed methods may then access, for each of the one or more nodes, a BMC at the OOB IP address, to invoke a BMC tool to retrieve cluster configuration information for each node. A dashboard user interface may then be generated, based on the configuration retrieved for each node, and displayed. The cluster configuration information may indicate whether the node is a cluster node, comprising a node assigned to a cluster, or a free node, comprising a node not assigned to a cluster.