Independently Coherent Compute Nodes With Pooled NIC I/O

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Data centers face inefficiencies due to resource stranding, where some compute nodes are idle while others are resource-constrained, leading to suboptimal utilization of hardware resources.

Innovation Solution

Implementing a system-on-a-chip architecture with a network interface controller hosting virtual and physical functions, allowing compute nodes to share pooled resources like memory and I/O devices through a fabric manager, enabling flexible access and independent operation modes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If compute nodes operate independently with dedicated resources, then each node has sufficient resources, but resource utilization is low and redundancy is high

Engineering Contradiction:

Improveresource availabilityVSAvoidresource utilization

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The patent merges resources across multiple compute nodes into shared pools. The fabric manager consolidates memory and I/O devices into common resources that can be dynamically allocated to any node, eliminating the need for each node to have dedicated copies of these resources.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The fabric manager creates a universal resource management system where memory and I/O devices serve multiple nodes simultaneously. Any compute node can access any pooled resource through the fabric manager, making resources multi-functional across the entire system.

Inventive Principle:

Principle #6Universality (Multi-functionality)

2Length of moving object

If compute nodes are brought together into a single package, then inter-node distances are reduced, but resource sharing efficiency decreases

Engineering Contradiction:

Improveinter-node distanceVSAvoidresource sharing efficiency

Core Design Contradiction:

Length of moving objectVSProductivity

Solution Approach 1:

The fabric manager acts as an intermediary between compute nodes and pooled resources. It manages the allocation and access rights, enabling efficient resource sharing even when nodes are physically close but logically distributed across different packages or modules.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Productivity

If a fabric manager pools resources across nodes, then resource utilization improves, but system complexity increases

Engineering Contradiction:

Improveresource utilizationVSAvoidsystem architecture

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent segments the system into compute nodes and a separate fabric manager component. This segmentation allows the fabric manager to handle all resource management complexity independently, while compute nodes maintain simpler, standardized interfaces for accessing pooled resources.

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS20260029918A1Computing device with independently coherent nodes

Publication Date: 2026.01.29 MICROSOFT TECHNOLOGY LICENSING LLC

AI summary

A computing device includes a system-on-a-chip. The computing device comprises a network interface controller (NIC) that hosts a plurality of virtual functions and physical functions. Two or more compute nodes are coupled to the NIC. Each compute node is configured to operate a plurality of Virtual Machines (VMs). Each VM is configured to operate in conjunction with a virtual function via a virtual function driver. A dedicated VM operates in conjunction with a virtual NIC using a physical function hosted by the NIC via a physical function driver hosted by the compute node. The computing device further comprises a fabric manager configured to own a physical function of the NIC, to bind virtual functions hosted by the NIC to individual compute nodes, and to pool I/O devices across the two or more compute nodes.