7 truths about data in the cloud

Fresh from the IBM acquisition, Cloudant CEO Derek Schoettle offers his take on how the migration of data to the cloud is changing IT

These are heady times for both cloud service providers and their customers. Never before has it been so easy to spin up large-scale applications and databases, fueling remarkably agile manuevers by companies of almost any size.

In this week's New Tech Forum, Derek Schoettle, CEO of Cloudant -- bought by IBM just two weeks ago -- walks us through the rapidly evolving world of cloud data management and where the cloud transition will be taking IT over the next few years. --Paul Venezia

Big bets in cloud data management

I'm not going to bore you with terms like "big data" and the capabilities of various strains of NoSQL databases. What I'd rather discuss is something that is changing drastically: how we store and access our data. As the cloud evolves beyond cheap infrastructure, the real innovation in 2014 will come from how we manipulate data stored on that infrastructure.

1. Multicloud use cases will become the norm, not the exception

Whereas Amazon Web Services used to be the only choice for developers, the market now has several viable cloud options, such as SoftLayer, Rackspace, and Windows Azure that work well with one another. Developers will look for the best service and expect providers to be compatible with one another.

Take Green Man Gaming as an example of an effective multicloud application topology. The digital e-commerce site uses Cloudant database as a service (DBaaS) on IBM SoftLayer infrastructure to ensure that its back end is always available during the large traffic spikes generated when the company runs special promotions on its website. With Cloudant managing the data layer, Green Man Gaming can focus on scaling its application tier, which the company manages itself on AWS. Green Man Gaming uses a multicloud strategy to get the best of both worlds for its site: elastic virtualized application servers, and bare-metal cloud infrastructure tuned for database performance.

No one wants all of their eggs in one basket, and developers are no different. With more choice will come the birth of the heterogeneous cloud experience.

2. Open source technologies will continue to drive out proprietary options
Keeping up with the pace of technology innovation is nearly impossible for one company to do on its own. Part of this is because the larger a company gets, the more risk averse it becomes, and pushing technology forward requires taking risks. The power of open source development in cloud data management and its exportability will allow companies like Cloudant to innovate at a rate as fast as -- if not faster than -- the largest software companies in the world, as open source allows a company to select the best existing solutions or platforms and deliver a tailored final service to its customers. This reality drives up the value of innovating on top of open source projects.

3. Customers do not benefit from vertical integration in the cloud economy
While at first vertical integration can drive cost efficiency, ultimately, it compromises the sum value of the parts. For example, having all of your data in one Oracle or AWS stack may save your company money up front, but once you're locked in, there is little need for that provider to innovate, given the lack of competition. At this point, it becomes a game of price increases and big support contracts.

That's not to say there is no benefit to vertical integration when it comes to hardware. Wikinomics made the point long ago that it's good for everyone when Oracle can create a new business ecosystem by acquiring Sun, or when Google launches its own handsets so that it can control the firmware and deliver better mobile apps to users. The supply chains involved in producing hardware harken back to the industrial age. Efforts like the Ubuntu Edge show how difficult it is to manage the hardware supply chain at a smaller scale.

The cloud economy allows companies to leverage the good of vertical integration while avoiding stagnation. The whole purpose of the cloud is to make hardware irrelevant. Many new companies launch today never having touched the physical hardware that runs their applications. When cloud infrastructure is this commoditized, there's no reason to get locked into the software and services layered on top of it. That's why we're seeing the current revolution in open source information management software. As 2014 progresses, we'll see the smart companies hopping from cloud provider to cloud provider, taking the best open source software along with them for the ride.

1 2 Page 1
Page 1 of 2
How to choose a low-code development platform