Remember when enterprises were skittish about the cloud? Security was one of the top concerns blocking developers from embracing the public cloud. That may be changing.Or is it? In a new Evans Data survey of 1,441 developers, 81.3% of respondents said they plan to build applications in the cloud within the next 24 months, but 51.9% said they'd jump to a rival cloud if better security were available. In other words, developers are diving into public cloud, in particular, due to its convenience, but they'll feel a lot better about the decision once better security comes along.
Have cloud, need security
A surprisingly low 27.4% of developers surveyed by Evans Data admit to building cloud applications today. But, if we look at their plans, another 53.9% expect to build cloud applications within the next six to 24 months.
SEE IaaS demand soars: Why this is great news for Amazon and IoT startups (TechRepublic)
Even that low 27.4% results in billions of dollars in revenue for Amazon Web Services, Microsoft Azure, and others. But, there's even more money waiting for the clouds that can entice users with a few key benefits.
Topping that list is security, at 52%, but reliability (42%), user experience (41%), and breadth of services (40%) aren't far behind:
These responses suggest that developers are actively embracing the cloud, and expect to continue, but additional work on the part of the clouds will fuel even greater adoption.
Multiple clouds, multiple headaches
This is not to suggest that enterprises are likely to shop around ad nauseum. While RightScale (which offers multi-cloud management tools) found that enterprises use an average of three public clouds and three private clouds, this multi-cloud dalliance is unlikely to stick.
SEE Security's future is the cloud, as enterprise trust in Amazon grows (TechRepublic)
As AWS cloud chief Andy Jassy told an audience at the Public Sector Summit, while a multi-cloud strategy sounds good in theory, in practice it results in higher service fees and increased complexity. Managing separate cloud stacks that don't really talk to each other does little to mitigate lock-in, but does much to increase frustration.
If you want to go multi-cloud what you have to do is to standardize on the lowest common denominator, and these platforms are nowhere close to the same. There is so much more capability and maturity in AWS than the other providers, and also iterating at a much faster clip.
The second thing is, for anyone who's had to maintain multiple stacks, it's a pain in the butt. It's hard, it's resource intensive, it's costly, and development teams hate it.
The third reason is that you give up some of your buying leverage, because all the providers have volume discounts, and if you split it over more entities, you don't have the same buying leverage.
So, most companies eventually settle into one preferred provider, usually AWS. Though Jassy is correct on the current feature gap between AWS and its nearest competitors, developers still want more, not to mention that holy grail of security. This means that while AWS has a huge lead, the cloud game is still very much on, with Microsoft Azure making great strides to catch up.
- IaaS demand soars: Why this is great news for Amazon and IoT startups (TechRepublic)
- Security's future is the cloud, as enterprise trust in Amazon grows (TechRepublic)
- Microsoft Azure doubles its lead over Oracle, IBM (TechRepublic)
- The cloud is a two-horse race between AWS and Microsoft (TechRepublic)
- Cloud Computing: Moving to IaaS (ZDNet special feature)
Matt Asay is a veteran technology columnist who has written for CNET, ReadWrite, and other tech media. Asay has also held a variety of executive roles with leading mobile and big data software companies.