Modular Node Card Management for Scalable Server Configurations

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The increasing diversity of applications in scalable servers makes it difficult to predict and configure compute resources, as traditional server systems offer fixed options that do not adapt well to varying needs between compute-focused and storage-focused setups.

Innovation Solution

A modular compute provisioning system that includes a system board and node cards, allowing for flexible configuration and scaling by providing modular compute resources, with features like hierarchical hardware partitioning, Ethernet connectivity, internal fabric connections, and dynamic power management, enabling diverse system configurations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If traditional fixed server systems are used, then system stability is maintained, but adaptability to diverse applications deteriorates

Engineering Contradiction:
Improveadaptability to diverse applicationsVSAvoidsystem configuration complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The server system is segmented into independent node cards that can be individually configured and installed. Each node card represents a modular unit with standardized interfaces, allowing the system to be customized by selecting and combining different node cards rather than configuring a fixed entire system.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system transitions from a static fixed configuration to a dynamic reconfigurable architecture. Node cards can be added, removed, or replaced based on application requirements, enabling the system to adapt its compute resources dynamically to match diverse workloads.

Inventive Principle:
Principle #15Dynamics

2Ease of operation

If fixed server options are provided, then manufacturing simplicity is maintained, but ease of operation for diverse configurations deteriorates

Engineering Contradiction:
Improveease of configuring compute resourcesVSAvoidmanufacturing complexity
Core Design Contradiction:
Ease of operationVSEase of manufacture

Solution Approach 1:

A universal node card design with standardized interfaces and protocols is implemented. The same physical platform supports multiple application scenarios through software configuration rather than hardware customization, making the system easy to operate across diverse applications while maintaining manufacturing simplicity.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system allows operational parameters such as compute node count, memory allocation, and storage configuration to be changed through software rather than requiring physical reconfiguration. This enables easy adaptation to diverse applications while the underlying hardware manufacturing remains standardized.

Inventive Principle:
Principle #35Parameter changes

3Adaptability or versatility

If incremental compute node installation is enabled, then adaptability improves, but system complexity increases

Engineering Contradiction:
Improveincremental scaling capabilityVSAvoidmodular configuration management
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

Node cards are designed with a nested modular structure where compute resources, storage, and networking are hierarchically organized. This nesting allows incremental installation and configuration while managing complexity through standardized hierarchical interfaces at each level.

Inventive Principle:
Principle #7Nested doll (Nesting)

Data Source

PatentUS9965442B2Node card management in a modular and large scalable server system
Publication Date: 2018.05.08 III HOLDINGS 2 LLC
  • US9965442B2 patent drawing
  • US9965442B2 patent drawing
  • US9965442B2 patent drawing

AI summary

A system for a system and method for provisioning of modular compute resources within a system design are provided.