Keeping control of your cloud spend

By Andrew Logan, Head of ECS’s Cost Optimisation Practice

While cloud adoption is proven to have a positive impact on banks’agility, there is always the risk that it will result in an excess of consumption that results in massive over spending.

But with cloud projects popping up across your business, those planned-for operational cost savings can vanish in a heartbeat. Plus the initial calculations around cloud provider pricing often ignore the full extent of on-premise support required, exacerbating the issue.

Analyst firm Gartner states that it is not unusual for public cloud bills to be 2x to 3x higher than expected. And it is a growing phenomenon, with Gartner anticipating that, “Through 2020, 80% of organisations will overshoot their cloud IaaS budgets due to lack of cost optimisation approaches.”

The need for restraint

We are already seeing this first hand. A growing number of banks are looking for help with their cloud cost optimisation as the realisation dawns that the unrestrained cloud spending cloud must stop.

No financial services organisation would consider running a data centre without adequate cost controls, and the same should apply in the cloud to avoid expenditure surprises, waste, and general inefficiency.

Understanding your cloud consumption habits

Understanding the costs and drivers of technology at a service level is essential for an efficient, sustainable cloud journey.

There are a number of cloud business management tools, including those from Apptio, that will help you to manage and optimise your IaaS and PaaS investments.If you don’t have the in-house resources and expertise to handle the entire cloud cost optimisation lifecycle process then you will likely need to engage an enterprise cloud consultancy.

If, like most banks, you are at the start of your official company-wide cloud journey, not counting any rogue, ad hoc cloud environments already in use, then a high-level migration and business case analysis is the first step towards justifying your cloud spend. This evaluates the total cost of ownership (TCO) of your on-premise, hybrid and cloud choices – and the potential savings.

The result is a detailed technical assessment with immediate recommendations, including governance and operating models. By mapping workloads to the most appropriate cloud provider and service, you can begin optimising costs, and making decisions based on how fast you can migrate specific workloads and applications.

Once the cloud migration is underway, using a tool to track and manage your hybrid IT environment will allow you to plan and forecast spend, utilisation and capacity,and justify future migration decisions.

One of the toughest challenges is managing the costs associated with working with multiple cloud providers. This is where cloud management tools really come into their own, as they give you a single view and full transparency of your cloud spend and usage across all providers.

Tracking out of control cloud users

If your organisation has struggled to suppress the voracious cloud appetite of business teams, then a policy of shared accountability is necessary to achieve any cloud cost optimisation.

By tagging workloads you can allocate cloud costs to the applications and business units that consume them.Once internal customers have direct visibility of their cloud costs, they are more likely to turn off unwanted services and stay within budgets.

Based on our experience working with large enterprises,the only way to avoid cloud budget overruns is by taking a granular look at costs and usage on a daily basis. This has to include associated employee, networking and security costs to get a complete picture.

While some cloud budgets may be stretched to the maximum, studies show that cloud utilisation can sometimes be as low as 10-20% of the provisioned capacity. The machine learning algorithms built into most cloud management tools make it easy to radically reduce waste from these under-utilised and idle instances.

To avoid the chaos that results from finding cloud resources that do not seem to have an identifiable owner, it’s worth standardising on a tagging strategy to understand the applications and business units that drive usage.

Once you have nailed the governance element and your bank’s cloud cost optimisation process is looking shipshape, you can turn your attention to breaking down the silos between public cloud and on-premise and start managing 
all of your IT costs in one place.  To align these with a standard cost model you have to ensure you are using apples to apples comparisons. This will then accelerate future migration and optimisation decisions.

In conclusion,applying cost optimisation best practice to your bank’s public cloud consumption will help you to avoid that bloated post-holiday feeling that comes from over indulging on those unnecessary extras. By keeping close tabs on public cloud usage, avoiding cost overruns and analysing the data to make more informed decisions, you can take full advantage of the agility the cloud has to offer and ultimately deliver a greater return on investment.