Virtual Machine Seed Image Replication via Parallel Deployment

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current disaster recovery methods for virtual machines are inefficient, particularly in cloud environments, as they require extensive time and bandwidth to replicate primary virtual machine images to secondary sites, or are cumbersome and costly due to physical transportation of data.

Innovation Solution

A method for generating secondary virtual machine seed images by deploying primary and secondary virtual machines from a golden image, suspending the secondary virtual machine, and replicating state data updates to create an up-to-date seed image, eliminating the need for large image replication and physical transport.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If primary virtual machine images are replicated to secondary sites using current disaster recovery methods, then disaster recovery capability is achieved, but extensive time and bandwidth are required

Engineering Contradiction:
Improvedisaster recovery capabilityVSAvoidtime required for image replication
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent segments the virtual machine image into two parts: a compressed base image stored at the secondary site, and incremental state data updates transmitted during operation. This segmentation allows the majority of the image to be prepared in advance while only small updates need to be transferred in real-time, significantly reducing bandwidth requirements and replication time.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary actions by compressing the base virtual machine image and storing it at the secondary site before any failure occurs. This pre-preparation eliminates the need to transfer large image files during disaster recovery, as only small incremental updates need to be synchronized during normal operation.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If physical transportation of data is used for disaster recovery, then data can be transferred, but the process is cumbersome and costly

Engineering Contradiction:
Improvedisaster recovery capabilityVSAvoidease of data transfer
Core Design Contradiction:
ReliabilityVSEase of manufacture

Solution Approach 1:

The patent replaces the mechanical process of physical data transportation with an electronic data transmission system. Instead of physically moving storage media, the system uses network-based transmission of compressed base images and incremental updates, eliminating the cumbersome and costly physical transportation process while maintaining disaster recovery capability.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

3Reliability

If large virtual machine images are replicated, then complete disaster recovery is achieved, but bandwidth consumption increases

Engineering Contradiction:
Improvedisaster recovery completenessVSAvoidbandwidth consumption
Core Design Contradiction:
ReliabilityVSLoss of energy

Solution Approach 1:

The patent extracts and separates the base virtual machine image from incremental state data. The base image, which constitutes the majority of data, is compressed and stored locally at the secondary site. Only small incremental updates are transmitted over the network during operation, dramatically reducing bandwidth consumption while ensuring complete disaster recovery capability.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent applies compression to the base virtual machine image, changing its size parameter from large to small. This compression allows the majority of the image data to be stored efficiently at the secondary site with minimal bandwidth consumption during updates, while maintaining complete disaster recovery capability.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS10768961B2Virtual machine seed image replication through parallel deployment
Publication Date: 2020.09.08 INTERNATIONAL BUSINESS MACHINE CORPORATION
  • US10768961B2 patent drawing
  • US10768961B2 patent drawing
  • US10768961B2 patent drawing

AI summary

Generating secondary virtual machine seed image storage is provided. An input is received to deploy a primary virtual machine and a secondary virtual machine based on a golden virtual machine image. In response, the primary virtual machine from the golden virtual machine image on a primary data processing site and the secondary virtual machine from the golden virtual machine image on a secondary data processing site are deployed. Execution of the secondary virtual machine is suspended on the secondary data processing site. Using the golden virtual machine image, a seed image corresponding to the secondary virtual machine is generated that is up-to-date at that point in time in storage at the secondary data processing site to form the secondary virtual machine seed image storage. The secondary virtual machine seed image storage is enabled to receive state data updates from the primary virtual machine on the primary data processing site.