Jump to Content
Storage & Data Transfer

5 cloud migration tasks you might be worried about (but don’t need to be)

August 8, 2017
Tom Nikl

Senior Product Marketing Manager, Cloud Migration

More and more organizations are looking to migrate their on-premises applications (and their data) into the public cloud, and an important early step is often reading stories from other enterprises that have successfully made the jump. But in reading these stories, you might hear about specific places where traditional cloud migration strategies left gaps in the process. Consequently, you might think you can plan hacks and workarounds ahead of time to address these gaps. But these planning challenges can still lead to tension between the IT and business teams and result in a never-ending debate over what will or won’t work—and, ultimately, a cloud migration project that never actually begins.

Many of the cautionary tales that exist stem from using cloud migration tools or strategies that were originally designed for backup and/or disaster recovery. They weren’t built for cloud migration—hence the gaps. Choosing a purpose-built cloud migration solution can help you avoid these gaps.

Here are five common challenges that can often be non-starters for cloud migration initiatives, and some ideas on how to avoid them outright.

1. How can we migrate all of our data (even with a slower uplink)?

The bulk of what you migrate will probably be the data supporting your applications, and this can easily be in the realm of terabytes. As a result, you probably need an efficient, optimized way to move it all into the cloud. If your uplink is on the slower side, bandwidth efficiency becomes an even more important component of your cloud migration.

You’ll need to take advantage of WAN and storage optimization technologies to move a lot of data through a small pipe. Deduplication, for example, makes sure that you only transfer data from on-prem and into the cloud once (after that first transfer, if it’s needed again, it’s copied from the cloud—not from on-prem). Additional WAN optimizations (like advanced compression) also aid efforts to migrate all that data to the cloud efficiently. We see Velostrata users able to reduce migration bandwidth usage by up to 75 percent.

2. How can we run applications in cloud fast (even before the data is there)?

As we mentioned above, moving the data is often the biggest part of a cloud migration (in terms of storage size). Many migration strategies make you wait for that data transfer to conclude before you’re able to run an associated application in the cloud. That means potentially waiting days before you can even give an application a test run in the cloud.

Velostrata removes this requirement by starting applications in the cloud first and then transferring the rest of the data seamlessly in the background afterward. Applications begin running in the cloud within a matter of minutes. This accelerated path to the cloud means migration projects can progress significantly faster, eliminating the staggered wait times associated with each application’s data transfer.

3. How can we adapt our workloads to run in the cloud?

There is more to getting applications running in the cloud than just copying and pasting all the bits. Workloads must be adapted to run in the new environment, which requires changing drivers, updating DNS and IP configurations, and more. You’ll also want to avoid application downtime—especially for production applications. A seamless cutover between on-prem and the cloud is also crucial to avoiding data loss.

We designed Velostrata to do all of these adaptations to cut out these common challenges. As soon as you initiate migration (via the vCenter management console, our REST API, or our Automation Runbook tool), Velostrata will install the required drivers, handle the DNS and IP information swap, and more. And because the actual instance of your application is migrated immediately, you can avoid complex cutovers.

4. How can we migrate legacy applications to the cloud?

Even in the best of circumstances, many enterprise data centers still have at least a few systems running legacy operating systems. These operating systems may no longer be supported by their manufacturers, but they may still be vital to your business. As a result, many businesses leave these legacy systems as-is when they’re moving to cloud, despite the potential for security vulnerabilities.

We think cloud migration should recognize the importance of these legacy systems, but also recognize that simply leaving them as-is isn’t the right solution. One of our goals when building Velostrata was to provide the same versatile migration path for legacy systems as any other system. This lets you migrate these legacy systems into the cloud quickly and easily, and then figure out the most ideal upgrade path. And because these systems are running in the cloud, they’ll benefit from enhanced security as a result.

5. What should we do if we still have physical servers?

Many organizations still have physical servers running bare metal operating systems. By moving them to the cloud, you can decommission on-prem hardware faster, giving you the power to right-size the cloud instance types whenever you need to (instead of being locked into the existing hardware specs).

What all of this shows is that a good up-front migration plan will go a long way toward successful cloud adoption--both the technology and the human aspects. And asking the right planning questions at the start can help keep migration plans on track and able to succeed. Find out more here about cloud migration with Velostrata.

Posted in