Hyper-converged Infrastructure Device Replacement via Workload Hashing

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Hyper-converged infrastructure environments face challenges in ensuring network bandwidth, leading to overburdened switches that drop data packets, affecting overall data center performance due to the pre-configured and highly integrated nature of resources.

Innovation Solution

The implementation of a hyper-converged management service that generates workload and device hashes for fast comparisons, allowing for the reassignment of workloads from failing nodes to available nodes, and configures new devices with exact settings of failing devices for seamless replacement, maintaining performance thresholds and resource allocation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If resources are pre-configured and highly integrated in hyper-converged infrastructure, then system integration and modularity are improved, but network bandwidth management and switch capacity become problematic

Engineering Contradiction:

Improvemodular and expandable compute, storage, and network resourcesVSAvoidnetwork bandwidth assurance and switch capacity

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The patent introduces a software-defined networking (SDN) controller as an intermediary between the hyper-converged infrastructure components and the physical network switches. This controller manages network traffic flow, allocates bandwidth dynamically, and prevents switch overload by intelligent routing decisions, thus resolving the contradiction between integrated resource management and network capacity constraints

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system implements dynamic network configuration where virtual network functions and traffic routing are adjusted in real-time based on current workload demands. This allows the pre-configured integrated infrastructure to adapt dynamically to varying network conditions, ensuring reliable bandwidth delivery without requiring over-provisioned switch capacity

Inventive Principle:

Principle #15Dynamics

2Device complexity

If switches handle all hardware connections in hyper-converged infrastructure, then system integration is improved, but switch overload and data packet loss increase

Engineering Contradiction:

Improveintegrated hardware architectureVSAvoiddata packet delivery and network performance

Core Design Contradiction:

Device complexityVSProductivity

Solution Approach 1:

The patent segments network traffic into different virtual networks and traffic classes, routing them through different paths and switches when possible. This segmentation distributes the processing load across multiple switches rather than concentrating all traffic through a single bottleneck, maintaining integrated architecture while improving packet delivery performance

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system adds a virtualization dimension to the physical network architecture, creating multiple virtual network layers that operate over the same physical infrastructure. This allows traffic to be routed through different virtual paths and logical switches, effectively increasing network capacity without adding physical switch hardware

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

3Ease of repair

If device replacement is performed in hyper-converged infrastructure, then system maintenance is enabled, but workload reassignment and configuration management become complex

Engineering Contradiction:

Improvedevice replacement capabilityVSAvoidworkload reassignment and configuration management

Core Design Contradiction:

Ease of repairVSDevice complexity

Solution Approach 1:

The patent implements automated configuration copying where the replacement device automatically receives and inherits the configuration, workload assignments, and network settings from the failed device through software-defined management. This eliminates manual configuration complexity and enables seamless device replacement while maintaining system integrity

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The system employs automated feedback mechanisms where the management platform continuously monitors device health, automatically triggers replacement workflows when failures are detected, and verifies successful workload migration. This closed-loop control simplifies the replacement process by eliminating manual intervention and ensuring configuration consistency

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS10445143B2Device replacement for hyper-converged infrastructure computing environments

Publication Date: 2019.10.15 VMWARE INC

AI summary

Disclosed are various examples for device replacement for device replacement in hyper-converged infrastructure environments. A workload hash for a workload to be processed by a hyper-converged compacting device can be compared to a device hash for the hyper-converged computing device to determine whether to assign the workload to the hyper-converged computing device for processing. When a new device is added to a hyper-converged infrastructure environment, a device has for the new device can be generated and workloads having low resources can be identified for reassignment to the new device.