Smart cloud consumers must consider a number of factors when calculating the total cost of ownership (TCO) of their environment. Here’s how to cut through the complexity.
Historically, private clouds have been labeled as the premium choice—an enterprise grade solution for mission-critical applications—that comes at a premium price. Total control, data protection, and performance are commonly cited as the best arguments for going private versus public. When considering total cost of ownership, some cloud users are quick to believe that public clouds are more affordable than their private counterparts. However, there is a tipping point where private clouds are a less expensive option for enterprises and offer additional benefits that add value to your operations.
In fact, a 451 Research study conducted earlier this year with 150 IT decision makers revealed that 41% of organizations were operating their VMware private clouds at lower unit costs versus public cloud. Another HPE report claimed that 40% of organizations were operating at a lower cost within the private cloud using OpenStack.
“If your organization is focused on efficiency, then private cloud might be the lower TCO option. Centralized applications, just-in-time hardware provisioning and prioritized workloads can drive high utilization, while automation, homogeneous workloads and outsourcing can drive labor efficiencies,” says Own Rogers, a research director for 451 Research.
It’s important to determine your needs not only today, but also how your business will change in terms of labor efficiencies and resource usage.
Image Source: “Cloud Computing” at Datamation
Most organizations should, but don’t, quantify the resources, assets, and utility charges it takes to run their cloud environment. Measuring resource consumption specific to an application or service requires cross-departmental coordination. If you’re able, track factors such as utilities, bandwidth, storage, software licenses, human labor, support costs, maintenance, auditing and compliance, and downtime or outages. As you can see, getting an accurate measurement on the TCO per application can become very complex. If you were to outsource some of your staff, or purchase new tools for automation and self-service, how would that affect your overall costs? If you aren’t already, measure your estimated cloud usage versus actual usage. Cloud waste could make the difference between using one model over another. Let’s dig into other financial considerations.
When investing in infrastructure, you’re getting a combination of resources in the form of CPU, memory, storage, and IOPS. Infrastructure from a public cloud comes in predefined bundles in what are known as flavors. Your application must fit those flavors, and you’ll want to choose the one that’s least restrictive. The spare or unused resources in these categories go back into the shared pool of resources and are consumed by your team or another company. In reality, you are paying to reserve resources, not for what you use. The term pay-per-use if often used to describe public cloud models, but that doesn’t quite illustrate how the financial model works.
Public cloud environments require careful management. It’s not unusual for cloud users to forget to shut down VMs, especially as projects build up and work becomes chaotic—but these VMs continue to accrue costs. You can either use third-party tools to help control costs and cloud resource waste, or create internal processes and reporting measures—both require time or additional money.
When it comes to private cloud, you have several options. You can pay a fixed monthly price for a dedicated private cloud that only your company has access to, or you can use a pay-as-you-go virtual private cloud that offers you a logically isolated environment. In a managed environment, you can offload the burden of managing and optimizing your cloud environment and gain some value-added services.
There are four basic types of workloads: linear, seasonal, stable, and unpredictable. The larger the number of virtual machines running, the more predictable your workload. (This is consistent with the probability theory, the law of large numbers.) The consensus among consumption studies concludes that stable, flat workloads are cheaper on a private cloud over a fixed term, while variable workloads are likely to reduce costs if in a public cloud or on-demand cloud model.
According to a 451 Research survey for 2018 IT planning, both public and private cloud growth will continue next year, with 14.8% of enterprises stating they plan to grow their private clouds, compared to the 7.2% who plan to expand their public cloud.
Other Cloud Drivers
Cost is not the only driving force behind adopting one cloud model over another. For some, the assurance of having dedicated infrastructure for their company is worth it.
Data protection, security and compliance, control, customization, as well as integration with business processes are of equal or higher importance than price for some cloud adopters. Those industries subject to HIPAA or PCI laws have to keep security and compliance as a top priority—they don’t have a choice. Plus, the cost of non-compliance far exceeds the price of maintaining a compliant cloud, as the average consolidated cost of a data breach is now $3.62 million, according to a recent IBM Security Report.
For other cloud users, there’s a fine line between cloud models: A 2017 research study conducted by VMware earlier this year measured the exact tipping point at which organizations would switch from private to public based on price. The majority of respondents said they would consider migrating approximately 25% of their workloads at a 10% price decrease.
Find the Right Solution
A Voice of The Enterprise (VotE) Organizational Dynamics 2017 report that 34% of 259 IT decision-makers surveyed migrated data from public to private cloud in the past 12 months. So what does all this evidence suggest? Your organization must develop a strategy that addresses your unique needs—there’s no such thing as one-size-fits all for cloud strategy, and you shouldn’t rely on what others’ are doing to guide you.
Your cloud budget is based on assumptions of estimated usage and labor efficiency. Shifting demands often make it difficult for your businesses to anticipate true cloud costs, but using our tips, you can make an educated guess. While the lower initial costs of public cloud hosting might seem like a compelling reason to choose a public cloud solution, if you have regulatory, security, and/or performance concerns, private hosting may be a better choice and allow you to remain both cost-effective and secure.
OnRamp cloud specialists can help you evaluate options and find the best fit for your company. Contact us today.
Additional Resources on This Topic: