8 steps to effective multi-cloud cost management
Strategic thinking and careful planning can help you squeeze the maximum value out of your multi-cloud environment. Here are eight tips to help you get started.
The financial and productivity benefits associated with embracing a multi-cloud environment are well-known. But multi-cloud infrastructures are complex, with many different providers and service terms. When working with multiple clouds, it's easy to waste money without even realizing it.
Multi-cloud cost management should be a priority from the start. While the cloud itself has relatively low costs initially, moving cloud workloads and dealing with multiple cloud providers can be tricky. Budgeting too severely is not the way to go with multiple clouds either, as vital features like management, monitoring and security should not be sidestepped because of their price tags.
Luckily, there are practical steps that you can take to avoid additional cloud costs. Here are eight ways to ensure that your organization gets the full value from its multi-cloud investment:
1. Investigate all available options
Research is the best and most reliable tool for multi-cloud cost management.
"Businesses need to understand and evaluate their needs and which providers offer the best pricing," said Paul Sussex, a financial services principal at professional services firm Ernst & Young.
It's essential to understand workload demand patterns to arbitrate costs across providers and leverage on-demand services effectively. "In the cloud, elastic storage may be less expensive with one provider than another that might have cheaper CPU pricing," he noted.
It's also important to understand what capabilities, such as third-party services or tools, are needed to accommodate workloads.
"In some cases, it can be more effective to choose a basic infrastructure provider or even keep an existing internal environment," Sussex concluded.
2. Map out your strategy
Careful upfront planning is always necessary for effective multi-cloud cost management. Evaluate all of the available options and pick the one with a cloud-native architecture that's most suitable for the workload, said Mike Fitzgerald, global solutions director at SoftwareOne, a software asset management and cloud optimization software and services provider.
"Evaluate all your on-premises workloads and prioritize and migrate the workloads that are going to get the biggest benefits," he said. Do not blindly lift and shift workloads to the cloud -- some workloads may just have to be retired instead of sent into the cloud. "The planning and the initial assessment will assure that you are cost-optimized from day one," Fitzgerald said.
3. Fully analyze business requirements
"Organizations often deploy application components to multiple cloud environments as per business requirements," Sussex said. Yet this approach, when made without fully considering its larger, real-world implications, can lead to higher costs.
"Using application workflows and relationships in conjunction with business requirements can reduce data hops across cloud boundaries," he said.
4. Use insightful, reliable monitoring
Continuously monitoring infrastructure use is a highly effective way of ensuring that resources and money aren't being wasted in a multi-cloud environment.
"We've seen lots of examples where customers inadvertently deploy significant additional resources within Azure or AWS and then [get] walloped with a significant increase in costs," said Julian Boneham, a director at N4Stack, a firm that helps customers manage and optimize their data and cloud platforms. "We all know that public cloud platforms give the opportunity to ramp resources up and down on demand, but this only provides a cost benefit if the utilization is managed effectively and underutilized resources are avoided."
5. Ensure proper data classification and tiering
Tiering and classifying data is essential to multi-cloud cost management. Skipping this step can lead to expensive data transfers that require the movement of data from one platform to another.
"Segregation of data across multi-cloud platforms to serve local applications and systems is favorable over data replication across cloud platforms," Sussex observed.
6. Don't be short-sighted
Multi-cloud cost management needs to be treated as a continuous, ongoing initiative and not a one-time activity.
"Don't instantly assume there will be cost-savings," declared Joe Kinsella, CTO and founder of VMware subsidiary CloudHealth Technologies, a cloud optimization platform provider. "The simple truth is that managing multi-cloud effectively takes time and proper planning to secure buy-in across business units and teams."
Cost-savvy organizations make the effort to align cloud costs to business key performance indicators. "For example, cost per user per month, cost per document per month," he said.
7. Exercise strong management
A multi-cloud environment gives an organization tremendous flexibility in deploying and supporting mission-critical applications.
"However, if multi-cloud environments are left unmanaged, the associated costs can spin out of control rather quickly because of additional layers of complexity," Sussex said. "Companies often struggle with hidden costs associated with unnecessary cloud data migrations [and] compute and storage resources, ultimately failing to deploy cloud resources at a cost that best fits business needs."
8. Don't scrimp on security
The biggest multi-cloud cost management mistake adopters make is failing to understand that the cloud provider isn't responsible for data and other stored assets, said Debbie Zaller, privacy leader at Schellman & Company, an independent security and privacy compliance assessor. Data breaches occur often, and organizations that mistakenly place their trust in their cloud providers are often left holding the bag after an attack. "This not only increases costs to add in security controls after the breach, but the breach itself is very costly," she explained.