UX Design - V2V Feature


#1

V2V Feature

This topic includes high level design concepts for the initial implementation (POC) of the V2V feature. Note that these are low fidelity mockups, and haven't been reviewed at a detailed level. This UI will be aligned with PatternFly.

Goals

  • Support large scale VM transformation between providers
  • Initial POC will target migration from VMware to oVirt
  • There is a longer term goal to support multiple providers including VMware, oVirt, OpenStack and KubeVirt

POC

The initial POC will demonstrate the ability to migrate VMs from VMware to oVirt

Navigation

The V2V features will be accessed in the Migration area, found under the Compute menu.

Overview

The overview page will provide access to both the Infrastructure Mapping and Migration Plan wizards. It will also include dedicated areas to show migrations currently in progress as well as completed migrations.

Infrastructure Mapping Wizard

Clicking on the Create Infrastructure Mapping button from the Overview page will invoke the wizard. The target oVirt environment will be utilized as the conversion host. This wizard allows the user to map source & target clusters, datastores and networks to produce an infrastructure mapping.

This wizard has 5 steps:

1 - General

The user enters the name of the infrastructure mapping and optionally includes a description.

2 - Clusters

In this step, the user is able to add cluster mappings, including 1+ source clusters and a target cluster.

3 - Datastores

In this step, the user is able to add datastore mappings, including 1+ source datastores and a target datastore.

4 - Networks

In this step, the user is able to add network mappings, including 1+ source datastores and a target datastore.

Once network mappings have been added, the user clicks Create to create the Infrastructure Mapping.

5 - Results

The results are shown in this page. A link is provided to allow the user to continue to the Migration Plan wizard in context.

Migration Plan Wizard

This wizard is either accessed by clicking the the link in the Infrastructure Mapping wizard or by clicking on the Create Migration Plan button from the Overview page.

This wizard allows the user to select an infrastructure mapping, import a file including the VMs to be migrated from the source, create a migration plan, and begin to execute it.

This wizard has 3 steps:

1 - General

The user selects the infrastructure mapping and enters the name of the migration plan and optionally includes a description.

2 - VMs

The user imports a CSV file, which includes the VMs to be migrated.

Once the file is imported, the user clicks Migrate to create the plan and begin to execute it.

3 - Results

The results page explains that the plan has been created and is in progress. Upon closing the wizard, the progress of the migration can be shown in the overview.

Viewing Migrations In Progress

The initial state of migrations is "Queued". As VMs begin to migrate, a progress bar will be shown to indicate the number of VMs migrated.

View Completed Migrations

Completed Migrations can be viewed in the Overview page.

Detailed design is in progress to show migration status as well as to allow the user to access those VMs which have failed to migrate, along with associated error messages.


#2

The ‘bring-your-own-CSV’ step sounds a little bit weird to me. Wouldn’t it be better to have some selection pane where multiple VMs can be chosen from? (ie, Ticked). This would allow users to unambiguously select the right VM(s). Also the source hypervisors can be queried so VMs are presented in any hierarchical organization they’re into (eg: Vmware Folders). Clicking on a folder would select all included VMs, etc.


#3

@frangdlt great point, you are correct! In the long term we will allow a different mechanism for selection of VMs. For this initial milestone, we will start with the Import via a CSV.