Cutting Cloud Costs: Key Strategies to Keep Budgets in Check

Hariprabu Sengoden Kandasamy
Published 03/13/2024
Share this on:

Cutting cloud costsCloud services are an integral part of operations for many organizations. Over the past decade, the rush to use cloud services instead of or in addition to in-house server farms for data storage and IT application operations has intensified. The marketing campaigns for cloud services by AWS, Microsoft, and Google make it seem like a “must-have” migration imperative. Yet, keeping cloud-related costs in check requires ongoing monitoring and management. Accordingly, it is time for companies to examine how they use the cloud to ensure they get the biggest bang for their buck.


Getting a handle on cloud expenses

A recent survey based on the experiences of 750 executives found that 30-35 percent of cloud expenditures may be wasted. Many organizations fail to take full advantage of the discounts offered by cloud providers. Redundant subscriptions for underutilized services (as when using a multi-cloud approach) and failing to monitor those costs can be problematic. More than half of those surveyed admitted that reigning in the costs for software licensing in the cloud has been a challenge. The State of the Cloud report by Flexera found some encouraging news, with survey respondents stating they are using automation to lower cloud costs, such as reducing workloads after hours and “rightsizing” regarding capacity.


Cloud services are not a fad

Employing various public and private clouds as a hybrid approach is not a fad. Gartner predicts that by 2025, enterprise IT spending on public cloud computing will overtake traditional IT spending as on-premise server farms become more of an exception and not a rule. Gartner warns that technology and service providers must adapt to that shift in real-time or risk becoming obsolete. As such, it’s critical to adopt a strategic plan to take optimum advantage of cloud services while minimizing negative impacts on the bottom line. The key components of a successful plan include:

  • Rightsized capacity. Avoid oversizing the cloud storage needed when moving away from an on-premise server farm. It is better to be conservative and pay less upfront, as cloud services can scale up as necessary to address an unanticipated demand for online transactions.
  • Avoidance of duplicate services. Organizations can lose track of cloud spending when using various suppliers and vendors. In the rush to embrace a still-emerging technology, different employees may purchase cloud services on behalf of their departments without coordinating those initiatives with others doing the same.
  • Optimized workloads. Reduce the cloud capacity required during downtime, such as when the business is idle on weekends and holidays. This is more difficult to predict for retail-oriented enterprises, but accurate forecasting can help lower on-demand cloud costs versus infrastructure reserved for predicted usage. It’s a tough task for IT teams: capacity planning requires that unexpected traffic spikes are covered – without overspending on resources that aren’t necessary.
  • Clean-up of inactive resources. Eliminate unused static IP addresses, unattached elastic block storage (EBS) volumes, and other resources that do not provide value for extended periods. This could include idle load balancers, underutilized databases, and orphaned snapshots.
  • A culture that minimizes cloud expenditures. The perception that the cloud is “cheap” and the ease of setting up a cloud service sometimes makes tracking usage and unneeded expansion challenging to monitor. Empower teams to manage costs and cloud workload. Invest in the right talent and retrain valuable employees who entered the IT environment when it was geared towards on-premise server farms. Use all available discounts from the onset of cloud migration and rewrite programs that move legacy workloads to the cloud as efficiently as possible, with improvements to existing architecture where needed. Avoid automatic subscription renewals where it makes more sense to review those cloud vendor services and their value to the company first.
  • A FinOps approach. Financial operations (FinOps) help align IT spending with business objectives. Establishing a cloud center of excellence (CCoE) staffed by IT professionals can reduce cloud costs and growth by 10 to 30 percent.
  • Discounts. Surveys show that less than half of cloud users take full advantage of the price breaks offered by the growing number of cloud providers. One example is reserved instances, where organizations commit to use a certain amount of capacity compared to on-demand instances. According to one account, this results in savings of up to 80 percent and is best suited where workloads are consistent and more predictable. On the other hand, spot instances can be considered clearance sale prices—where unused capacity from a cloud vendor is snapped up at discounted rates. This type of spot pricing may be subject to abrupt cutoffs if a surge in usage from other clients happens and may be best utilized for tasks that can weather any disruptions. All major cloud players offer this type of discounting. Working with a vendor who understands how a potential client operates and its load fluctuations is crucial for cost optimization.
  • Address technical debts. Code or technical debt is a term widely used in the IT industry. Technical debt occurs when a product or development team decides on a quick solution to expedite a project, prioritizing speed over perfection. This often means additional costs linked to refactoring, where internal source codes must be overhauled to address decisions made in haste.
  • Legacy systems. One way to cut operational costs when committing to the cloud is to have a plan to handle older legacy systems (either by retiring and/or rewriting). Outdated hardware and software, license and warranty expirations, potential security breaches, and a lack of scalability for increased workloads may provide financial justification to move away from legacy systems, allocating more company resources for ramping up cloud services.



Want More Tech News? Subscribe to ComputingEdge Newsletter Today!



Clearly define the cloud journey

As with other nascent industries or technologies and the rush to have the newest, shiniest toy, some organizations are still determining where cloud services fit within their operations and where it is a cost-effective move that enhances daily operations. The pressure to embrace the latest technologies and keep up with competitors that incorporate cloud services may lead to paying for excess capacity and underutilized cloud apps. Organizations with little or no history of using the cloud as part of standard operation procedures don’t always have a track record of where and how tweaks can be made to optimize cloud usage. There may also be hidden costs for certain transactions that weren’t realized when signing on initially.

If possible, avoid locking into a single cloud vendor. Creating the IT architecture in-house specific to one provider can be costly and makes it difficult to justify migrating to another cloud service that may be a better fit. On the other hand, employing a multi-cloud approach, using a mix of public and private clouds with on-premise IT infrastructure, can make managing costs very complex.

The relative newness of the cloud also means there is a learning curve for in-house IT analysts and managers as they strive to fully understand where more services are needed and when they might be overpaying for what they are currently using. As the cloud industry evolves and end users become savvier, providers may align more with client needs and work more closely with those who pay too much for underutilized services.


About the Author

Headshot of Hariprabu Sengoden KandasamyHariprabu Sengoden Kandasamy has more than 20 years of experience in the tech industry with a focus on cloud optimization, software performance engineering, and tools development. A mechanical engineering graduate, Hariprabu has lent his expertise to numerous Fortune 500 companies, driving system optimization projects that have consistently enhanced the performance and efficiency of complex systems and technologies. For more information, contact


Disclaimer: The author is completely responsible for the content of this article. The opinions expressed are their own and do not represent IEEE’s position nor that of the Computer Society nor its Leadership.