3 reasons why 2017 will see massive cloud migration

Enterprises have spent two years putting the pieces in place to move applications to PaaS and IaaS; they're now ready to go full speed ahead

3 reasons why 2017 will see massive cloud migration
Thinkstock

How many Global 2000 enterprise applications are on the public cloud right now? You’ll see estimates of 20 to 30 percent, but that overstates the reality.

When all outsourced hosting is taken into account—which includes SaaS, IaaS, and PaaS—many analysts estimate that 20 to 30 percent of workloads are currently on the cloud. But a better metric is to look at what enterprise applications have migrated to an IaaS or a PaaS platform, which is how most enterprises measure their presence in the cloud. Although SaaS is certainly an option for replacing on-premises applications, its usage tends to be for new, often off-the-shelf software, not existing software as in the case of IaaS and PaaS.

Using that IaaS- and PaaS-only scenario, Global 2000 enterprises have migrated about 5 to 7 percent of their on-premises applications. That’s up from my estimate of 1 percent in 2013, a figure that aligns well with the revenue growth of the major public cloud providers.

What will the number be at the end of 2017? Brace yourself: I figure it will be about 18 to 20 percent. Why that huge jump? I see three reasons:

  1. In 2015 and 2016 many Global 2000 companies set up migration factories in their IT groups. These factories include methodologies, tools, approaches, and even the use of devops practices to speed cloud migrations. Moreover, these companies have been training and hiring cloud architects, consultants, and developers, then putting in place other new skills they will need. Thus, the troops are aligned and ready for battle.
  2. The projects in 2015 and 2016 that got the Global 2000 to 5 to 7 percent were pioneer projects that provided enough experience for enterprises to become confident that they can now hit the accelerator.
  3. Enterprises have picked their low-hanging fruit for the first wave of migrations. In other words, they are going to pick the easy stuff first, which willed them migrate a large volume of applications fast.

If you’re surprised by my migration projection of 18 to 20 percent this year, I’ll surprise you again: I believe we’ll see a similar jump in 2018, too—for the same reasons.

Of course, at some point, we’ll reach a saturation point where the workloads left to migrate won’t prove cost-effective, and many enterprises will elect to leave them where they are, move them to a managed server provider or colocation provider, retire them (with—where possible—a SaaS replacement if the functionality is needed).

Anyway, that’s my story and I’m sticking to it. We’ll see if I’m right this time next year.