Rolling Update Mechanism for Networked Virtualization Storage

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

In networked virtualization environments for storage management, rolling updates are challenging due to data unavailability during node shutdowns or restarts, which can lead to service disruptions and data loss, necessitating a mechanism to optimize resource availability and minimize these impacts.

Innovation Solution

A mechanism for performing rolling updates that integrates across multiple types of upgrades, utilizing a distributed architecture with Controller VMs to manage storage resources, including local and networked storage, and employing a master update module to coordinate data unavailability checks and token-based approval for node updates, ensuring data availability and minimizing downtime.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a node is shut down or restarted to complete an update, then the node receives necessary bug fixes and security updates, but data residing at the node becomes unavailable during the update process

Engineering Contradiction:

Improvesystem reliabilityVSAvoiddata unavailability time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system divides the node into multiple components that can be updated independently. Specifically, the node is segmented into a first virtual machine, a second virtual machine, and a management agent, allowing updates to be applied to individual components rather than requiring complete node shutdown. This enables continuous availability of other components during the update process.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The management agent performs preliminary actions by detecting available updates and coordinating the update process before actual updates are applied. The agent prepares the node by identifying update candidates and orchestrating the sequential updating of virtual machines, ensuring that updates are applied in a controlled manner that maintains system availability.

Inventive Principle:

Principle #10Preliminary action

2Adaptability or versatility

If updates are performed periodically at nodes, then the most current version of storage management functionality is available to user virtual machines, but the node must be shut down or restarted for a period of time

Engineering Contradiction:

Improvestorage management functionalityVSAvoidsystem productivity

Core Design Contradiction:

Adaptability or versatilityVSProductivity

Solution Approach 1:

The system segments the node's virtualization components into independently updatable units. Virtual machines and management agents can be updated separately and sequentially, allowing the node to maintain productivity while storage management functionality is modernized. This segmentation enables updates without complete system shutdown.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The update mechanism ensures continuous useful action by maintaining system operation during updates. The management agent coordinates updates so that while one virtual machine is being updated, others continue to operate, ensuring that storage management functionality evolves without interrupting overall system productivity.

Inventive Principle:

Principle #20Continuity of useful action

3Reliability

If the hypervisor is upgraded to install necessary security updates, then the virtualization system receives critical security patches, but the node becomes unavailable during the upgrade process

Engineering Contradiction:

Improvesecurity reliabilityVSAvoidnode unavailability time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system separates the hypervisor update process from individual virtual machine updates. The management agent detects hypervisor updates independently and coordinates their application, allowing security updates to be applied to the hypervisor layer without requiring complete node shutdown. Other virtual machines can continue operating during hypervisor updates.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The management agent acts as an intermediary between the hypervisor and virtual machines during updates. It detects hypervisor updates, coordinates the update process, and manages the transition, ensuring that security updates are applied while minimizing disruption to virtual machine operations and maintaining node availability.

Inventive Principle:

Principle #24Intermediary (Mediator)

4Reliability

If storage devices are taken offline for firmware upgrades, then necessary bug fixes and security updates are installed, but the storage device remains unavailable until the upgrade is complete

Engineering Contradiction:

Improvefirmware reliabilityVSAvoidstorage device unavailability time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system segments storage devices into independently manageable units with separate firmware update processes. Each storage device can be updated individually through coordinated offline/online cycling, allowing firmware updates without taking the entire storage system offline. This enables bug fixes and security updates while maintaining overall storage availability.

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS9740472B1Mechanism for performing rolling upgrades in a networked virtualization environment

Publication Date: 2017.08.22 NUTANIX INC

AI summary

Disclosed is a mechanism for performing rolling updates in a networked virtualization environment for storage management. The approach is applicable to any resource in the system, including controller VMs, hypervisors, and storage devices. Integrated processing may be performed across multiple types of upgrades.