Managing cloud costs is challenging, and it's only getting more difficult with each passing year. Costs continue to spiral out of control. And getting the balance right cost, utility, and flexibility? Nearly impossible.
So, how exactly did we get here?
In this blog, I will share my insights and recommendations based on my experience working with enterprises successfully managing cloud costs. I'll also specifically dive into what you can do when it comes to your data environments.
How Managing Cloud Costs Spun Out of Control
We have come a long way since AWS launched Elastic Compute Cloud (EC2) and Simple Storage Service (S3) in 2006. In doing so, they heralded a new era of agile, cost-effective, and highly available compute and storage solutions. But I wonder if we ever anticipated where we would be in 20 years?
Companies like Facebook and Google set a trend in the early 2000s of gathering more and more data about their customers to gain more insight into their needs and preferences, digitizing ever more services as they grew. Every other business was forced to follow to remain relevant.
Then came the COVID-19 pandemic and lockdowns that forced every business to accelerate their digital transformations and journeys to the cloud. And so we find ourselves in 2025, where 90% of the data stored today was created in the last two years, with data volumes growing at a compound annual growth rate of 19.2% (Statistica, 2024).
As a result, predictions that a majority of IT budget will be spent on cloud infrastructure and services looks set to become true. Canalys estimates global cloud infrastructure services spending rose 20% in 2024, up from $267.7 billion in 2023 to $321.3 billion in 2024. This was driven in part by the growth in compute- and data-intensive AI use cases.
Today, it feels like a perfect storm is brewing. Cloud costs can no longer be ignored. You need to think about deploying innovative solutions and adopting new practices to keep costs under control and achieve business outcomes.
Back to topWhy Should You Care About Managing Cloud Costs Now?
CFOs and financial controllers have always cared about metrics like cost to serve and cost per unit. However, in the higher interest rate world we now find ourselves in they have been brought to the forefront. There is now a significant capital cost, and everyone is focused on cash flow. Business growth at any cost is no longer an option.
In my day-to-day interactions with customers, I see more teams either migrating existing services to or building new services on the public cloud. However, a recurring theme over the past year is that it costs more than everyone was expecting. The costs have become so great that they are now causing many to either scale back their initiatives and curb innovation or completely rethink their plans.
We all got so excited about the agility and the ability to switch from lumpy capital expenditure (CapEx) investments into smooth operating expense (OpEx) investments. Everyone agreed that moving to the cloud was a great idea because of the flexibility it promised. However, as architectures and data flows have become more complex, organizations have struggled to model the associated costs upfront. This is in part contributing to a new trend of cloud repatriation.
Overspending Is Easy
It is far too easy to overspend on cloud — without recognizing the value that they should by managing resources in the cloud.
According to Gartner® , “Organizations with little or no cloud cost optimization plans rush into cloud technology investments. They end up overspending on cloud services by up to 70% without deriving the expected value from it. Executive leaders must use this research to realize cost savings and avoid cloud overspend.” 1
Overspending is exacerbated by the underutilization of cloud resources. Many enterprises I work with have found themselves unable to switch off idle resources and recognize the benefits of cloud utility cost models. Easy access to “unlimited" resources has caused sprawl and duplication across multiple accounts.
Forbes also found that 32% of a company’s cloud budget is squandered, a majority of which results from too much money allocated to applications and storage that go unused. CxOs and heads of procurement are now demanding cost-savings.
Costs aren't the only metrics that matter. If you're measuring DevOps success, consider DORA metrics.
Back to topThe Potential Pitfalls of Managing Cloud Costs
Enterprises have recognized for some time that cloud spend is accelerating at an unhealthy rate. This has driven many to adopt a range of strategies, some of which I would argue may lead to even greater cloud spending.
Shifting from Infrastructure as a Service (IaaS) to Platform as a Service (PaaS)
In an attempt to reduce administrative costs and technical debt, many solutions have started to make use of PaaS offerings. There are potential benefits of evergreen compute and data storage in production environments. But the lack of data management capabilities and the always-on nature of the solutions means costs spiral. This is especially true when customers build out multiple separate environments for development, testing, quality assurance, user acceptance testing, and training purposes.
Standardizing on a Single Public Cloud Offering
To get the best possible prices many enterprises have also consolidated their operations onto a single cloud. This has also helped reduce the number of skills needed by development and operational teams.
However, these same businesses, as a result of legislation like the EU Digital Operational Resilience Act, also known as the DORA regulation, now face the prospect of having to at least partially reverse this decision to demonstrate they are not overly dependent on a single cloud services provider.
All in Favor of Multicloud?
As a result, multicloud and hybrid cloud strategies are seeing a resurgence in popularity. I work with several Perforce Delphix customers whose primary cloud is either an on-premises private cloud or hosted with one of the three hyperscalers. Alongside that, they will utilize at least one other major cloud provider and can demonstrate to regulators their ability to spin up services on the alternate cloud at will.
While a multicloud architecture helps avoid lock-in and keeps regulators happy it does add additional cost and complexity. It requires separate technologies and skill sets, which can be challenging. Moving data between clouds also becomes more of a challenge, especially if PaaS solutions have been employed.
It seems like there is no way to win.
Back to topHow to Successfully Manage Cloud Costs in Your Data Environments
When embarking on cloud projects, you need to have technologies and processes that allow you to control and manage cloud costs. Just like security, cost control should not be an afterthought. Consider solutions that help mitigate the growth in consumption driven by the ease of access to cloud resources.
You must strike the right balance between flexibility and cost efficiency to drive better outcomes. To ensure new solutions are adopted, they need to deliver more than cost savings. They need to bring benefits to analyst, development, and testing teams. New solutions need to integrate seamlessly with existing toolchains. Any perceived friction will slow uptake. Efficiencies need to be delivered at the same time as improvements in quality and development velocity.
Succeeding with Multicloud
It looks like the regulators will require a multicloud strategy, at least for companies operating in Europe. For the customers I work with, having Delphix as a DevOps data platform creates a unified data layer across their multicloud environment, covering both private and public clouds.
The presence of Delphix simplifies access to data, especially for developers, while helping ensure regulatory compliance. It provides the necessary flexibility and governance without sacrificing operational efficiency. Multiple, independent writeable thin copies of data can be shared with different teams, reducing storage costs by as much as 90%, while integrated data replication technologies limit the cost of intra-cloud and inter-cloud data transfers.
Data provisioning is not just self-service, lowering wait times for developer and testing teams it can also be integrated into existing development pipelines using the APIs available in the centralized management plane, Data Control Tower. The ability to automate data provisioning allows for ephemeral development and test environments built on top of traditional IaaS solutions or new Kubernetes-based PaaS solutions like AKS, EKS, and GKE.
In this video, my colleague Grant Ward shares how Delphix cuts cloud costs through ephemeral data environments. See it in action.
I have customers who have seen total infrastructure costs drop by more than 80% as a result of leveraging this data agility.
Analysts agree. In a recent IDC study — IDC: The Business Value of Delphix — the infrastructure savings that Delphix enables are clear.
IDC: The Business Value of Delphix
Analyst study of customers shows that the Delphix DevOps Data Platform delivers a 408% 3-year ROI. Discover the tangible benefits of Delphix, validated by IDC analysts.
**Analyst data gathered from the IDC White Paper, sponsored by Delphix, The Business Value of Delphix, doc #US52560824, December 2024.
For the average customer studied, they achieved $1.9 million in three-year infrastructure cost savings. This was based on the number of full-time equivalent (FTEs) required for equivalent environments. Without Delphix, it would require 58.6 FTEs. With Delphix, that number reduced to 43.1. Plus, with Delphix, the cost of the infrastructure itself was reduced by an average of $707,600. This all adds up to an average 408% 3-year ROI.
This is just a small sample size of Delphix customers. The infrastructure savings with Delphix can depend on the scale of the organization. Many large enterprises that I work with have achieved even bigger savings in the multi-millions.
Additional Examples
Technology
Delphix test data automation enabled Dell Technologies to flip developer productivity from 20/80 to 80/20.
Healthcare
Delphix has enabled Molina Healthcare to reduce storage requirements from 4PB to 200TB, resulting in an estimated $6 million to $10 million savings in storage costs over three years.
Financial Services
Gain Capital achieved faster data delivery with Delphix, which increased environment utilization. This allowed Gain Capital to remove QA environments and lower infrastructure costs.
Telecommunications
Sky Italia improved test data management practices by choosing Delphix. As a result, they condensed the infrastructure footprint by over 90% and slashed operational costs by 30%.
Back to topOptimize Your Data Environments With Delphix
Legacy approaches to data delivery often involve manual processes that take days or weeks, creating bottlenecks in development, testing, analytics, or AI projects. Delphix automates these processes and delivers application data in just minutes.
Related blog >> What is Delphix?
Data Virtualization Transforms Environments
Delphix syncs with production data sources then instantly provisions space-efficient virtual data copies for non-production use cases. Rather than making and moving new data blocks, Delphix intelligently shares common data blocks across downstream environments over the network.
The result? Delphix accelerates provisioning times by 100x while reducing storage footprints by 10x.
Your Data Has Never Done This Before
Virtual data copies function like physical ones but can be delivered and refreshed from production on demand. Delphix also provides a comprehensive set of data APIs that enables DevOps teams to refresh data after test runs, synchronize multiple data copies for integration testing, or version data copies like code.
Scalable Data Management
Delphix also offers Elastic Data to enable cost-efficient, scalable data management across the application lifecycle. With Elastic Data, Delphix delivers 2x less expensive storage to help optimize IT costs.
The result? Delphix reduces storage costs by 50% in the cloud.
Get Started With Delphix
Contact us to learn how Delphix can help you virtualize databases and scale data management, so you can better manage your cloud infrastructure costs.
1Gartner, Realize Cost Savings After Migrating to the Cloud, By Finance Research Team, 17 August 2022. GARTNER is a registered trademark and service mark of Gartner, Inc. and/or its affiliates in the U.S. and internationally and is used herein with permission. All rights reserved.