BMC Host Network Traffic Failover Routing

Solution Overview

Problem

Conventional fault-tolerant Information Handling Systems (IHSs) require a dedicated secondary Network Interface Card (NIC) for redundancy, which increases hardware costs and is not feasible for large-scale deployments, as they do not offer failover capabilities for Ethernet controller failures.

Innovation Solution

The system utilizes a Baseboard Management Controller (BMC) to route host network traffic through a network failover path, leveraging existing BMC resources, such as a secondary Reduced Gigabit Media-Independent Interface (RGMII) and standby Ethernet Media Access Control (MAC) addresses, without the need for an additional NIC, using a Field-Programmable Gate Array (FPGA) for monitoring and switching host network traffic.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a dedicated secondary NIC is used for network failover, then network reliability is improved, but hardware cost increases

Engineering Contradiction:

Improvenetwork reliabilityVSAvoidhardware quantity

Core Design Contradiction:

ReliabilityVSQuantity of substance

Solution Approach 1:

The BMC's network interface is designed to serve dual purposes: managing baseboard operations and providing network failover for the host processor. By making the BMC's network path universally usable for both management traffic and host network traffic, the system achieves redundancy without adding dedicated failover hardware.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The BMC uses its own existing network infrastructure (RGMII interface and Ethernet MAC address) to provide failover services for the host processor. Instead of requiring the host to have dedicated backup NICs, the BMC's self-contained network capabilities are leveraged to serve the host's failover needs, eliminating redundant hardware.

Inventive Principle:

Principle #25Self-service

2Reliability

If a dedicated secondary NIC is used for network failover, then fault tolerance is improved, but device complexity increases

Engineering Contradiction:

Improvefault toleranceVSAvoidsystem complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent combines the management network path and host network failover path into a single shared infrastructure through the BMC. By merging these functions into one unified network path using the BMC's RGMII interface and Ethernet MAC, the system reduces complexity compared to maintaining separate dedicated NICs for each function.

Inventive Principle:

Principle #5Merging (Combining)

3Reliability

If a dedicated secondary NIC is used for network failover, then network availability is improved, but cost burden increases

Engineering Contradiction:

Improvenetwork availabilityVSAvoidhardware cost

Core Design Contradiction:

ReliabilityVSQuantity of substance

Solution Approach 1:

The BMC's network interface is designed to serve dual purposes: managing baseboard operations and providing network failover for the host processor. By making the BMC's network path universally usable for both management traffic and host network traffic, the system achieves redundancy without adding dedicated failover hardware.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The BMC uses its own existing network infrastructure (RGMII interface and Ethernet MAC address) to provide failover services for the host processor. Instead of requiring the host to have dedicated backup NICs, the BMC's self-contained network capabilities are leveraged to serve the host's failover needs, eliminating redundant hardware.

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS10069718B2Switching of host network traffic through baseboard management controller (BMC)

Publication Date: 2018.09.04 DELL PROD LP

AI summary

Systems and methods for switching host network traffic through a Baseboard Management Controller (BMC) are described herein. In an illustrative, non-limiting embodiment, an Information Handling System (IHS) may include a host processor; a Baseband Management Controller (BMC) coupled to the host processor; and a memory coupled to the host processor, the memory having program instructions stored thereon that, upon execution, cause the IHS to: determine that a network path dedicated to the host processor has failed; and route host network traffic to and from the host processor through a network failover path provided by the BMC.