Cost management in the age of multi-cloud
and cloud native

Multi-cloud adoption has made it more of a challenge for enterprises to predict their cloud spend. Cloud native will make it even more difficult and even hinder your business success, unless you fundamentally change your mindset on how you integrate cost analytics, adopt the principles of FinOps and create value for your end
customers in a fundamentally different way.

Multi-cloud has become a must have across industries and continues to be one of the fastest growing areas of IT spend. Gartner’s latest IT investment forecast, for example, expects data center systems spend to drop from $195 billion in 2019 to $190 billion through 2022. Conversely, spending on cloud infrastructure services (IaaS) is expected to ramp up dramatically from $39.5 billion in 2019 to $63 billion through next year.

Predicting the cost of cloud

With this increased budget, CIOs must deliver on the promise of cloud to create an agile and scalable environment that can satisfy the needs of the low-touch, high-tech economy. Cost analytics with continuous optimization is an important part of this scenario.

Cloud cost analytics solutions are designed to give insight to continuously optimize overall cloud spend. This is done by pinpointing areas of potential improvement, eliminating areas of cloud waste, reserving some capacity for dynamic workloads and providing the ability to proactively forecast cost based on predictive analysis of your environments.

Mandatory work-from-home programs and an increase in digital interactions has seen the cloud usage of most enterprises skyrocket this year. According to Flexera’s State of the Cloud 2020 report, organizations are over budget for cloud spend by an average of 23% and forecast that cloud spend will increase by 47% in 2021. As a result, the report maintains that there will need to be a greater focus on optimizing migrating workloads in addition to cost management and governance for maximum operational efficiency.

More efficient cloud cost control

Through 2024, most legacy applications migrated to public cloud IaaS will require optimization to be more cost effective, according to Gartner. This will see cloud providers building up their native optimization capabilities to provide enterprises with cloud choice in cost-effective architectures that will satisfy their varying performance demands.

Cost optimization is fundamental to any cloud migration strategy. It is essential to identify from the onset the right applications to migrate, for example. Also, cloud-native, microservice-based applications, which are created for distributed multi-cloud deployments, may perform better than traditional applications but could also lead to extreme complexity over time. This can make it more difficult to predict your cloud costs unless you’ve adopted modern tools to connect these continuous variations in cost to business outcomes.

Asset management system for traditional workloads?

If you are moving traditional applications to the cloud, which can’t necessarily take full advantage of cloud-native tools for analytics or consumption models, you should still implement a strategy to identify unused and unattached resources so they can be removed – manually or automatically – based on policy. All too often, services are spun up for a project and forgotten. At the same time, ensure that teams have visibility on what they are spending.
Deploying an asset management system could potentially detect a new service as soon as it goes live, making services and cloud costs easier to track. Tagging resources with user ownership is a way of tracking spend and where it originates from.

Also, know where your data is, its value and its legal ownership. Some enterprises have a habit of storing everything in the cloud, even worthless data, which could prove expensive unless you have associated a data strategy linked to business outcomes.
Another way of reducing operating costs and increasing flexibility is adopting containers designed to package applications, making it easier to deploy and manage them. Note, however, that they need to be built in the right way or they can end up increasing complexity and costs.

Additional considerations for cloud native

Cloud native is an approach to creating and running applications in the cloud that takes advantage of cloud computing’s distributed delivery model. This increases speed, agility and scalability, all essential ingredients to creating applications and services for the digital economy.

However, cloud-native projects usually require additional commitment, in both financial terms and from human resources. It is a continuously evolving landscape as new tools and practices are being continually developed. Introducing cloud native principles and DevOps best practices into an enterprise is dependent on adopting cross-functional methods and ways of working that address business goals. This may require executive sponsorship, culture training and mentoring budgets.

Furthermore, cloud native requires a different perspective on costs. If enterprises start to look at the costs before the business outcomes, they will not get the outcomes or agility they expect. While centralized mechanisms need to be in place to monitor spend, locking down spend prematurely will inevitably choke the speed of development.

This is what FinOps aims to address with cloud financial management. Nonprofit organization The Linux Foundation has linked up with the FinOps Foundation to advance this discipline through best practices, education and standards. FinOps will help ensure that enterprises have insight into their cloud spend and drive efficiencies for DevOps.

Cloud transformation must be treated differently

A massive cultural change may be necessary in moving to cloud and, in particular, cloud native that includes cost analytics and FinOps. To put it into perspective, the jump to cloud native is a far bigger step than moving from mainframe to open source, since it could fundamentally change the way you create and bring value to your next-generation customers.

Costing is one of the major pain points that enterprises face in their moves to the cloud. But enterprises have to remember that cloud transformation isn’t just about lifting and shifting data – it is also about re-inventing themselves. Costs are dynamic, and everything should be linked to business outcomes. The revenue will follow.