Dual-Rail Fabric Routing for VM Migration in HPC Networks

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing high performance computing environments face challenges in efficiently managing network traffic and addressing issues related to live migration of virtual machines due to complex addressing and routing schemes, particularly in InfiniBand networks, which hinder scalability and flexibility.

Innovation Solution

Implementing a heterogeneous and asymmetric dual-rail fabric configuration that utilizes InfiniBand and RoCE technologies, along with dynamic reconfiguration mechanisms and efficient routing algorithms, to support virtual machine migration and optimize resource utilization in cloud computing environments.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If traditional network addressing and routing schemes are used in InfiniBand networks, then virtual machine migration can be supported, but network reconfiguration overhead increases and scalability decreases

Engineering Contradiction:

Improvevirtual machine migration capabilityVSAvoidnetwork reconfiguration overhead

Core Design Contradiction:

Adaptability or versatilityVSLoss of time

Solution Approach 1:

The network fabric is divided into multiple independent rails (e.g., rail 0, rail 1) that can operate separately. Each rail maintains its own addressing scheme, allowing virtual machines to migrate between hosts within the same rail without requiring global network reconfiguration. This segmentation isolates reconfiguration overhead to individual rails rather than the entire network.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent implements asymmetric dual-rail configurations where different rails can have different characteristics (e.g., one rail for management traffic, another for data traffic). This asymmetry allows optimization of each rail for specific purposes while maintaining overall network flexibility for virtual machine migration without uniform reconfiguration requirements.

Inventive Principle:

Principle #4Asymmetry

2Adaptability or versatility

If complex addressing and routing schemes are implemented in InfiniBand networks, then network functionality is enhanced, but system complexity increases

Engineering Contradiction:

Improvenetwork functionalityVSAvoidaddressing and routing scheme complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent introduces rail identifiers as an intermediary layer between the virtual machine migration functionality and the underlying complex InfiniBand addressing system. The rail ID provides a simplified abstraction that masks the complexity of global InfiniBand addressing and routing, allowing virtual machines to migrate while the complex addressing schemes remain hidden in the background.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The addressing space is segmented into rail-specific addressing domains. Each rail has its own addressing scheme independent of others, which simplifies the addressing complexity within each rail compared to a single global complex addressing system. This segmentation makes the overall system more manageable despite supporting multiple complex addressing schemes simultaneously.

Inventive Principle:

Principle #1Segmentation

3Ease of operation

If homogeneous fabric configurations are used, then network management is simplified, but flexibility and adaptability decrease

Engineering Contradiction:

Improvenetwork management simplicityVSAvoidfabric configuration flexibility

Core Design Contradiction:

Ease of operationVSAdaptability or versatility

Solution Approach 1:

The network fabric is designed with universal rail interfaces that can accommodate different rail types and configurations. Each rail follows a common interface standard allowing for homogeneous management procedures, while the underlying implementation can be heterogeneous to provide flexibility for different traffic types and migration scenarios. This multi-functionality enables both simplified management and increased adaptability.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS12470623B2System and method for supporting heterogeneous and asymmetric dual rail fabric configurations in a high performance computing environment

Publication Date: 2025.11.11 ORACLE INT CORP

AI summary

Systems and methods for supporting heterogeneous and asymmetric dual rail fabric configurations in a high performance computing environment. A method can provide, comprising at one or more computers each including one or more microprocessors, a plurality hosts, each of the plurality of hosts comprising at least one dual port adapter, a private fabric, the private fabric comprising two or more switches, and a public fabric, the public fabric comprising a cloud fabric. A workload can be provisioned at a host of the plurality of hosts. A placement policy can be assigned to the provisioned workload. Then, network traffic between peer nodes of the provisioned workload can be assigned to one or more of the private fabric and the public fabric in accordance with the placement policy.