The key questions you need to ask before throwing your data into the cloud

Cloud has been an industry buzzword for some time now, but has continued to evolve at quite a pace. Despite this, one of the predominant themes of discussion that remains is how to take advantage/avoid the risks associated with the technology.

The hype surrounding software as a service (SaaS) combined with cloud computing has become almost deafening in recent times. Although it's great for exposure and recognition, hype can also be a detriment to successful adoption of a solution or a technology. Prospective users can often find this is paired with unrealistic expectations, misunderstandings, and even complete disillusionment.

Read more: The five latest trends in cloud computing

No doubt, on paper (or should that be virtual desktop, these days?), the proposition on offer from service providers and hyperscale companies, mainly public cloud providers, is tempting. Benefits include limitless compute power, lower expenses, simple pricing, a pay as you use model, global access and end-user transparency. It does indeed sound like a fabulous club to join.

However, there seem to be quite a few catches that need careful consideration before diving head first the cloud. And most of them concern your data. I recommend using the following checklist:

  • Who in your business owns overall data governance and what are their thoughts?
  • What happens if the NSA or GCHQ want to access your data? Is that a concern?
  • What happens if you want to change suppliers – either in an emergency (think of what happened with Nirvanix or 2E2) or simply because a newer service might be better for your needs (e.g. the recently announced Google Compute Engine or VMware Hybrid Cloud Service)?
  • What data structures will you use? Open standards or otherwise?
  • What level of performance guarantee and quality of service do you have? Like broadband, is your service level dependent on how others choose to use the same cloud?
  • Have you considered any network distances involving the speed of light and latency?
  • Are you going to be breaking any laws by moving your data to the cloud?
  • How will you audit your data in the cloud, or across multiple clouds, if you need to? If you have a method, what does your regulator think?
  • Even if the cost for storing it is next to nothing, how much does it cost to retrieve your data when you need it? How long will this process take?

How to counteract industry concerns

Despite all of this, for many workloads - especially those with fluctuating and/or very high scale resource requirements - the speed and cost benefits will no doubt outweigh any of these risks. And, unlike traditional IT, cloud services don't have to be purchased by the IT department. So, there is no real way of stopping eager employees even if the IT or security departments have restrictions in place.

Read more: How the cloud supports green IT computing

Most organisations will suit a hybrid or private cloud offered by cloud service providers who can offer bespoke IT and applications run "as a service" in their data centres. Alternatively, the hyperscale players who can offer global economies of scale for IT infrastructure, that are going to be difficult to compete against and replication in any other way is not an option. Obviously, price point has been a huge part of technology investment considerations but as we move out of austerity, the flexibility and agility of elastic computing must be the key considerations for infrastructure. The key to this will be building a data platform that allows the IT department (led by the CIO) to offer users the choice of services they have come to expect, but with the appropriate long-term risks and the questions above taken into account.

One unique and invaluable aspect of cloud computing is that you can enlist vendors to help you convince management that the ROI/business value potential is there. By testing the concept first, you can help alleviate fears and hesitation before signing a contract. This can eliminate any questions you might have before putting your data into any cloud.

Specifically, ask about the ability to pilot the solution. You may still need to pay for implementation services associated with the pilot, but in the new world of cloud computing, look for proof points and results before you make a large investment.

Read more: 10 major cloud computing terms you need to know

Over the past six months, the industry has heard very few senior IT people recently say they are "all-in" when it comes to the cloud. Although slightly ambiguous, one would hope they have thought about what they would do if they decide they need to be "all-out" at some point in the future. The cloud should be an extension of your IT strategy, not a replacement for one.

Laurence James is the product alliances and solutions marketing manager at NetApp