How to Crack the Code: Minimising Cloud Waste in a Hybrid World

More and more, businesses are turning to the cloud to support remote work, manage cost requirements and improve productivity. Cloud adoption continues to grow at exponential rates – with one figure recently reported by Microsoft of a 775% increase in use in Azure in certain locations. It can be very easy to overshoot your planned investment or even not take full advantage services you may already be paying for. Ensuring that you’re optimising your cloud spend is critical to supporting long-term distributed work environments, continued organisational agility and eventually, managing overall costs.

  

The complexity of cloud costs

The cloud is a strategic resource for many businesses, offering greater organisational agility, adaptability, improved efficiency and cost reduction. Yet, a common challenge that many enterprises face when beginning their cloud journey is unexpected bills. It’s not hard to understand why though. The cloud is billed in a very different manner than traditional on-premises systems.

It does not help that cloud cost management is more complicated than many anticipate. Driven by consumption, cloud licensing encompasses a bewildering array of options, from enterprise-wide to token-based models, that requires careful planning. The licensing rules not only vary considerably between vendors – but they also regularly change and get updated, causing more headaches for IT managers handling hybrid cloud environments. While individual public cloud vendors provide tools to manage their cloud services, they may not offer a comprehensive overview across all cloud deployments including other public providers, private instances and virtual machines.

On demand but hard to spot

It’s easy to spin up a cloud instance on demand, which means there could be activity above and beyond what IT teams are aware of (or have contracted for), creating spikes in the organisation’s overall bill. The same ease of use that differentiates the cloud also means that on-demand instances may not be managed as strategically as they should be, if IT teams lack the visibility into these deployments. For example, if your development team needs to spin up cloud instances for specific projects, these deployments could be scaled down when the team is offline and/or when the project is complete.

IT teams have an inherently more difficult task understanding who is using what, when, and why across a hybrid cloud environment. Add the increasing instances of individual departments who purchase and deploy cloud solutions on their own, without the knowledge of the IT team, and it is easy to see why businesses have difficulty just wrapping their arms around cloud management.

The potential ripple effect of blind spots

Visibility into how the cloud, whether that is private, public, or virtual machines, is being used is crucial.

Just last year, Gartner estimated 'cloud waste' would reach $14.1 billion in 2019, up from $12.9 billion in 2018, showing just how much of a challenge organisations have when it comes to attempting to gain visibility into and control of their cloud investments.

Cloud costs can also add up in other ways beyond the bill itself. It’s common for individuals to have many certifications in one cloud service, but it can be challenging to find staff that has training from various providers. That means that some companies may need to employ extra staff with the required background and certifications in a particular public cloud to add the appropriate expertise to the IT team.

To manage cloud costs effectively and avoid potentially unsustainable solutions, IT decision-makers should ensure they set expectations, understand the needs of their workforce and adjust policies about the use of the cloud accordingly to avoid short-term pitfalls.

How can you take action?

Businesses must get the right tools in place to manage complex cloud environments.

·         Establish organisational expectations and policies. While IT may not be able to completely control spend by business units, there should still be a standard set of policies in place that clearly outlines on what acceptable cloud usage looks like.

·         Ensure common cloud cost optimisation strategies are in place. When implementing the cloud, every organisation needs to regularly check that they have rightsized (the appropriate allocation of resources is in place for their workload), created power schedules, and considered reserved instances for AWS, among other things.

·         Consider whether tools are needed to provide additional visibility. Cloud management platforms and cloud cost optimisation resources can provide a comprehensive views into the enterprise’s cloud footprint and offer suggestions to better optimise usage. The best tools combine insight from on-premises and cloud workloads to inform usage and spend forecasting, as well as provide actionable or automated capabilities to address these suggestions quickly.

·         Set up robust automation and governance capabilities. A high level of visibility into cloud requirements and usage also enables process automation to eliminate cloud waste. By identifying the main sources of cloud usage, you can increase or reduce the available resource to ensure the most efficient cloud use. For some businesses, this could mean giving users self-service portals to request cloud resources and turn it off as they need to. Either way, given that businesses may be using anywhere between five or ten thousand machines in the cloud, automation is essential.

As businesses look to the cloud, IT leaders need to be aware of the broader requirements for cloud management prior to implementation to avoid spiralling costs. Once organisations decide to move a significant volume of their legacy on-premises technology to the cloud, the ongoing management following migration must be high on the company’s agenda. No matter where an enterprise is on their cloud journey, visibility is key to reducing the hybrid cloud challenge.