BrainBlog by Jason Bloomberg
In this series of articles, my colleague Eric Newcomer and I have written about leveraging VergeIO to kick the hyperconverged infrastructure habit and cost-effectively regain control over infrastructure and sovereignty.
Perhaps the most immediate business benefit from deploying VergeIO, however, is migrating off VMware.
Most enterprise deployments of VMware have been in place for years, growing in complexity over that time. The result (as VMware parent Broadcom fervently hopes) is permanent vendor lock-in.
VergeIO changes this equation. Instead of prohibitively expensive and time-consuming VMware migrations, enterprises can leverage VergeIO to migrate off legacy VMware to the more efficient, AI-ready VergeOS Private Cloud Operating System with a final cutover that only takes a matter of minutes – an interval short enough to fit into a modest, planned maintenance window.
Sound too good to be true? Let’s look under the covers and see how VergeIO manages this remarkable feat.
How VergeIO leverages VMware APIs to work its magic
VergeIO’s VMware migration tool, ioMigrate, leverages two essential VMware capabilities to help operators migrate off the legacy virtualization platform: VMware’s vStorage APIs for Data Protection (VADP) and the associated Changed Block Tracking (CBT) capability.
VADP are VMware’s backup API framework. VergeIO leverages these APIs to establish a data transfer path for migration, rather than relying on traditional VM import and export capabilities.
By leveraging these APIs, VMware migration is a straightforward, five-step process:
Step one: connect – Operators connect to VMware vSphere and use VMware’s own VADP to access source VMs and initiate snapshots without the need to export VMs as offline images.
Step two: initial copy – Next, operators perform the initial copy to the destination running VergeOS while the source VMs continue running on VMware. This initial copy can take as much time as necessary but takes place out of band and thus has no effect on the production environment.
Step three: migrate changed blocks with CBT – The third step in the migration process leverages CBT to identify and track which disk blocks have changed since an earlier backup or synchronization point.
With this information, ioMigrate can copy only the blocks that have changed since the initial migration, dramatically reducing both the time of migration as well as network traffic. Operators can migrate changed blocks as necessary until it’s time for the final cutover.
Step four: final cutover during planned maintenance window – At the designated time, the operator shuts down the production VMs, performs the final synchronization, and starts up the destination VMs on VergeOS – a process that may only take minutes because of the use of CBT as well as the fact that ioMigrate can perform this final synchronization by migrating blocks in parallel.
Step five: final test – Before the maintenance window closes, operators test the destination environment to ensure it is fully operational. If a problem crops up that they can’t solve during the window, then they simply roll back to the source VMs, fix the issue, and repeat the final migration steps as needed.
Only after all tests pass do the operators switch live traffic to the new environment.
How VergeIO lowers VMware migration risk
Most examples of failed VMware migrations result from an insufficient equivalence between the source environment and the destination.
A migration that is 99% successful is in reality a failure, as the 1% that failed can torpedo the entire effort. Given organizations can’t simply turn off their infrastructure – and thus, their business – for an extended period of time, most VMware migrations are either rushed or incomplete, or both.
VergeIO addresses these limitations via its pre-copy, continuous synchronization, and rapid cutover processes. Not only is the final cutover quick and relatively painless, but it also allows for sufficient testing to ensure the destination environment is fully operational.
The fact that VergeIO leverages the VADP and its CBT capability from VMware are important enablers of this migration process by giving operators API-level control.
The Intellyx take
Given Broadcom’s draconian VMware licensing changes, simply moving off VMware can be a cost-effective, strategic move for any organization struggling with vendor lock-in.
VergeIO, however, offers more than simply a new port in the storm. VergeIO built VergeOS with a single, unified code base that integrates the hypervisor, storage, and networking to abstract all the elements of the private cloud infrastructure.
As a result, VergeIO not only moves organizations off VMware, but also provides a more efficient environment that requires fewer servers, is easier to manage, and is overall more AI-ready than VMware’s offering.
The AI readiness benefit, in fact, highlights an important fact about how VergeIO handles VMware migration: such migration is not just about the VMs, but also the workloads that run on them.
For enterprises struggling to deploy sufficient AI infrastructure, VergeIO gives them more than a new place to run old workloads. VergeIO also offers a new place to run new workloads – in particular, AI workloads that place new burdens on virtualization infrastructure that the VergeOS Private Cloud Operating System is well-positioned to support.
Copyright © Intellyx BV. VergeIO and Broadcom are Intellyx customers. All of Intellyx’s content is 100% human generated, including this article. Intellyx retains final editorial control of this article. Image credit: Craiyon.
