Cloud sprawl is defined as an uncontrolled growth of computing resources that underlay cloud-based IT services and exceed an organisation's resources.
Cloud has proven to be a democratic force for many organisations, as people no longer have to wait for the IT department to create a new service, bring a new offering to market, or launch an application.
However, according to Empired, this sense of democracy is a key reason for the growing phenomenon of cloud sprawl.
Andrew Burke, Empired chief technology officer of Infrastructure Services, says, “The adoption of best practices and methodologies relating to IT service management have helped the IT department ensure change occurs in a planned, risk-averse way.
“The trouble is, sometimes these checks and balances can choke innovation and responsiveness from a business perspective.
“This has made the cloud a perfect workaround for business users who crave innovation and feel constrained by their IT department.”
“Organisations are increasingly subject to ‘grey IT’, where business users consume cloud services directly from the public domain without the express knowledge of the IT department,” he says.
Burke says this brings a number of challenges including:
Burke says, “The sprawl of cloud services is very similar to the growth and spread of server virtualisation in the early to mid-2000’s, when VMware ESX really hit things off.
“Back then we were seeing huge growth in server numbers for all sorts of purposes, many without clear definition. The outcome then, as it is today, is that without planning, proper execution becomes very difficult.
“To find an answer to cloud sprawl, we can look back upon some of the lessons we learn from the virtualisation sprawl.”
How to control cloud sprawl, according to Empired:
1. Build cloud expertise and an application services model based on the cloud.
Empired says, familiarise yourself with the market place and work with the business to define suitable use cases and product alignments.
If there is a strong alignment to one vendor or a select number of vendors, get some staff certified or engage a trusted partner to help.
Once you have an idea of what the future state looks like, create a gap analysis to help figure out how to get there.
2. Understand your applications and data.
Know which applications contain your confidential, proprietary, or otherwise sensitive data, and build a policy that defines where these data types can be stored and where they cannot. Use this knowledge when defining your end state, Empired says.
3. Implement management tools.
An effective management toolset will help aggregate use from internal, hybrid, and public cloud systems.
This use, when combined with some form of chargeback mechanism, will provide insight into the wider IT platform and who is consuming it. This may require some ongoing updates as new cloud vendors appear, but the effort will be worth it, according to Empired.
4. Review your Information Technology Infrastructure Library (ITIL) and provisioning processes.
Strong controls within an IT organisation should not be discounted. It is likely that your existing processes can be streamlined or presented in a more business-friendly way, Empired says.
Similarly, it is quite simple to provide your users with a private cloud-style, front-end to your internal systems.
This step is all about making the IT department easier to deal with because, if you’re easier to deal with than a cloud vendor’s website, your users will probably come to you first, says the company.
5. Begin or alter the course of your cloud journey with the new endgame in mind.
Once you have a view of where the IT future needs to be you can begin to 'trim the edges' and get your IT footprint back under control, says Empired.