Cloud Customers Must Adopt FinOps to Control Costs
As cloud computing has become mainstream, gaining visibility into and controlling cloud costs has become an ongoing issue for businesses. And the cost management tools offered by cloud providers don’t do the job.
Cloud costs continue to rise and the amount of unnecessary spending remains high, according to the Flexera 2022 State of the Cloud Report survey published in March. For the sixth year in a row, optimizing their use of the existing cloud for cost savings – sometimes referred to as FinOps – was the number one cloud priority this year for enterprise technical and business professionals. Improving financial reporting on cloud costs was also a major initiative.
“FinOps and all of this optimization and cost management is a critical part of the cloud operating model that customers need to put in place when migrating to cloud domains,” said Chris Wegmann, chief technology and practice officer. Accenture’s AWS Business Group. “[It’s] usually a function that is ignored or missed or not invested as much.
Deploying FinOps encourages data-driven management of cloud spend while simultaneously increasing efficiency and typically reduces cloud spend by up to 20% to 30%, according to Accenture.
While cloud providers provide some tools and support through technical account managers, customers must play a very active role, according to Wegmann. This means knowing what their workloads need and making the necessary changes to optimize them, or risk paying more than they should.
“This is a shared responsibility because…AWS cannot make these changes on your behalf,” Wegmann said. “Many customers in their on-premises environments don’t have that visibility, so they don’t know where the spend is. But that’s a big advantage in the cloud.
It’s all about data
Cloud providers offer basic cost management services, but they usually need to be supplemented with third-party software and require cloud users to do some of the heavy lifting.
“Data is absolutely essential for cloud cost management, optimization, and it usually comes down to markup,” Wegmann said. “This means that you label each element of your cloud domain and then [make] sure that the data model brings it all together. The tools themselves don’t automatically create this data model or mapping, and that’s something that needs to be done.
AWS’ cloud cost management product suite includes AWS Cost Explorer to help customers view and manage their AWS costs and usage across all accounts; AWS Budgets to improve planning and cost control with budgeting and forecasting; and AWS Cost Anomaly Detection, a machine learning model that learns customers’ historical spending patterns to detect one-time cost spikes and ongoing cost increases. AWS Trusted Advisor inspects a customer’s AWS infrastructure and provides best practice recommendations when there are opportunities to save money and optimize system availability, performance, and security.
If you know what kind of computing performance your application requires, AWS offers more than 475 instance types for customers to choose from to give them the best price-performance scenario, noted Chris Grusz, AWS business development director, AWS Marketplace, Service Catalog, and Control Tower.
“We have over 15 different database types…so you can choose the exact database that will be what you’re really looking for, then run it on the exact instance type you need,” said Grusz. “And then, of course, we’re always looking to lower that price wherever we can. [AWS] Marketplace also integrates with a number of our partner solutions, so if a customer wants to view third-party spend for SaaS applications alongside their AWS consumption, we also have ways to power many of the management products. popular costs that are on the market.”
Accenture itself has a very large AWS presence that it constantly seeks to optimize, and it uses the same internal tool it uses with its clients – the Accenture myNav platform – to optimize its cloud architecture, track and manage consumption and adjust spending accordingly, according to Wegman.
“We have tens of millions of lines of billing data coming in; that means tens of millions of different services are running,” Wegmann said. “We aggregate that and bring that to about 1,000 different teams. Through this, we are constantly looking to optimize, save money, disable things when not in use, switch expensive instance types to lower costs. Sometimes it’s cheaper storage or just a change… to more serverless, which is usually cheaper than having servers sitting around to process things.
Accenture and nearly all of its AWS customers use a combination of native AWS and third-party tools such as CloudHealth by VMware or Cloudability to help monitor and control their cloud costs because no one tool can do it all. do, Wegmann said.
“Unfortunately, it takes two or three different tools, as well as what’s from Amazon, to do this,” Wegmann said. “We spend a lot of time with our clients, helping them make those decisions. Some of them do reviews or do free trials [of third-party software] or things like that to understand capabilities and understand where it fits into their shortcomings.
help me help you
According to Amit Zavery, Managing Director and Head of Platform for Google Cloud, helping customers control their cloud costs is all about maintaining long-term relationships.
“If they’re successful, we’re successful,” said Zavery, who manages Google Cloud’s commerce, billing and monetization product team. “If they have budget constraints or budgetary requirements, or if they want to have a future plan around their use, we give them control. Otherwise, it would be very difficult for them to scale and ultimately use many more services. »
The Zavery team expands the cost control and management capabilities provided to customers through Google Cloud Console, which integrates cloud usage and historical customer data.
“We’ve tried to ensure that we provide as many tools and capabilities so customers can review their usage, manage costs, and optimize their capability utilization so they can achieve better results using our products from the point from an economic point of view,” Zavery said. “That data is also available to them to extract and put into any kind of analytical tool. We also have value advisors within Google Cloud who can work closely with clients.”
Google Cloud announced a series of price increases for its infrastructure services in March that will take effect on October 1. It adds new data replication and network egress charges, and doubles the prices of some services, such as cold-line storage operations.
“We just brought our pricing somewhat up to industry standards compared to anything else,” Zavery said. “We have of course worked closely with customers whenever we make price changes, and we also try to provide our customers with information on how their invoices will look.”
Microsoft declined to discuss how it helps customers monitor and control their cloud costs.
“Tracking cloud usage and associated costs should be a transparent process for any customer using the cloud,” a Microsoft spokesperson said in a statement. “That’s why Microsoft offers a suite of services, like Azure cost management and billingto ensure customers have the tools to accurately forecast and manage their cloud usage. »
While the big three cloud providers say they help customers or make it easy for them to monitor and control their costs, Corey Quinn begs to disagree.
“All three will tell you they absolutely do, but they don’t,” said Quinn, chief cloud economist for The Duckbill Group, which helps AWS customers manage their cloud costs. “They offer different tools, different points of visibility and so on, but part of the problem is that they’re far enough away from their customers that they don’t fully understand the scope of the problem. And on some level, let’s be clear, it’s an unsolvable problem to solve.
The convoluted pricing structures of cloud providers have effectively become impenetrable, according to Quinn.
“For virtually all customers, the monthly cloud bill is what the provider tells them, because who will even have the energy to dissect it, let alone discuss it,” he said.
The complexity of understanding and controlling cloud costs depends on what an organization looks like (as it is, how it interfaces between functions) and its technology stack.
“The more elastic, cloudy, and higher-tier a cloud provider’s services, the less predictable the spend becomes,” Quinn said. “Yes, it becomes much more efficient, but also much more difficult to predict.”
Quinn typically starts customer conversations about cloud costs with a series of questions: Is their cloud bill accurate? Are there a bunch of things that should be disabled? Are there any configuration errors in the technical sense that relatively minor changes would have a big impact on? Do they have contract negotiations coming up?
“All vendors offer deep discounts for incurred spend, but that commitment is measured over multiple years to a point where it’s not just ‘predicting your cloud bill next month,’ but ‘predicting what it will look like. your spending on what curve for the next half-decade,” Quinn said. “It’s an incredibly daunting task because you’re about to sign on the dotted line, but it determines what it’s going to look like.”
“One thing that’s often a duck is people thinking, ‘Ah, I’m going to move to the cloud, because it will save me money,'” Quinn said. “I did the TCO in depth [total cost of ownership] analysis, and I’m left with the conclusion that if you’re taking this route to the cloud from your data center to save money, you’re almost certainly wrong.
After all, cloud computing is not just a server rental market; it’s a whole new way to build and maintain apps.
“You won’t see any appreciable savings for at least five years — probably never, depending on what you’re doing,” Quinn said. “So the reason to do that instead is because of a capability story – it allows you to move faster as a business.”