Companies are generating and contending with more data than ever before. Unsurprisingly, cloud costs have become challenging to manage due, in part, to this massive influx of data. According to our Enterprise Cloud Index, the vast majority (85 percent) of organizations report cloud cost control as a challenge within their company. Further, 30 percent say they’re “very concerned” about cloud costs in relation to their IT budget for the coming year.
When done right, cloud can provide significant benefits such as on-demand capacity, simple scalability, and agility. For applications and workloads that have high variability in their resource needs over time, cloud can deliver significant business value. On the other hand, the cost of that flexibility generally gets harder to justify for the vast majority of workloads that change at a slow, predictable pace. The problem is that many companies have been so hyper-focused on getting to the cloud—at any cost—that they’ve often neglected to consider how best to optimize their investments.
In a mad dash to get to the cloud and figure out the details later, organizations have found themselves in a tough position: they’re not experiencing all the benefits they expected from cloud (conversely, they’re oftentimes incurring additional costs and complexity) and then have to start making reactive decisions to right the ship. To avoid this, it’s critical that companies start seeing the cloud as an operating model versus simply a destination.
Every cloud-related decision a company makes should be focused on addressing the business's needs and ensuring long-term success, not jumping on the "IT du jour." Most often, this means carefully selecting and leveraging multiple cloud environments (architecting on-premises as one of these cloud options) based on each application's specific requirements; the end goal being to gain more control over both cloud operations and spend, while also having the flexibility to pivot (i.e., change clouds or strategy) as necessary.
Here are the top considerations organizations should take into account to optimize cloud costs:
Selecting the right cloud for each workload or application is vital for controlling cloud costs. Certain use cases are better suited to public clouds, and others may perform more effectively on-premises. Of course, deciding which workloads to run and where to run them requires careful consideration, and what’s right for the business one day might not be right the next, so this needs to be a long-term strategy rather than a quick solution. To do so efficiently, companies should implement a hybrid multi-cloud strategy, leveraging one platform to run all applications and data across different clouds, on-premises, and at the edge.
For instance, if an organization is building a new application and wants a low barrier of entry and accelerated speed of experimentation, using a public cloud to quickly stand up resources and assess performance can be the best choice. But down the line, when costs grow, traffic increases, and the application reaches a stable state, it might be better suited for on-premises or a different public cloud provider to optimize costs. An increasing number of companies even opt to run customer-facing applications in the cloud for scale and geo accessibility reasons but conduct all testing and development on-premises to control costs.
Choosing the right cloud at the right time requires some degree of experimentation and regular assessment of one’s infrastructure, which leads into the next point.
Knowing when to move across cloud offerings is integral to optimizing cost and performance. But how can organizations know when it makes sense to change clouds or perhaps move away from the cloud entirely (as in the example below)? The key is establishing a routine evaluation process so that decisions are made based on business needs versus fads that might not provide adequate value. A good place to start is by implementing tooling that monitors the cost of applications so that companies can evaluate whether said applications would run more cost-effectively in a different cloud location.
As a best practice, businesses should also keep a close eye on application performance and set thresholds that alert them to when it’s time to evaluate whether a move is necessary. Additionally, companies should stay abreast of the latest cloud offerings so they can pick and choose cost-optimized solutions. Be discerning here: if switching clouds will ultimately save a few pennies, but the move itself will cost a dollar, the move will not make sense.
In some instances, moving on-premises makes the most sense—such was the case for email service company HEY. The company ran exclusively in both Amazon's and Google's clouds since its inception, but eventually concluded that, “Renting computers is (mostly) a bad deal for medium-sized companies…with stable growth,” and “The savings promised in reduced complexity never materialized.” By regularly evaluating its cloud strategy, the company realized that the cloud was not the most cost-effective or streamlined option and pivoted accordingly.
Organizations need to be wary of hidden fees and factor them into their overall cloud spend. One example of hidden fees is egress costs, for which NASA was notably fined a whopping $30 million by AWS. High-performance applications can also incur unexpected costs because they commonly require specialized, expensive public cloud instances and configurations to avoid throttling. Accordingly, organizations should ensure the need for these specialized cloud instances is factored in when estimating cost for high-performance applications.
Whether it’s egress costs or other miscellaneous fees associated with running high-performance applications, companies need to have a predetermined strategy in place so they don’t become derailed by unexpected expenses.
Getting the most value from your cloud investments is a continuous process. Organizations benefit from a deliberate hybrid multi-cloud approach that delivers a consistent cloud operating model and keeps them in control. By choosing the right cloud at the right time, adjusting one’s cloud strategy as needed, and factoring in hidden fees, businesses have the best chance at optimizing cloud costs effectively.
Thomas Cornely is the Senior Vice President of Product Management at Nutanix.
Related articles:
With many enterprises new to NaaS, service providers can differentiate themselves from competitors by prioritizing superior customer service and experiences.
Unlocking the full potential of telecom in the SaaS era relies on comprehensive strategies and a constant commitment to evolving security practices.
Eliminating complexity by standardizing with a platform approach is the way to put the brakes on the complexity cycle.
