Rethinking the Complex Economics of Cloud
This blog was originally published by Booz Allen here.
Written by Dan Tucker, Booz Allen.
How to optimize cloud for enterprise value
IT and mission leaders often spend numerous budget planning cycles weighing the economics of on-premises versus cloud operations, and that’s completely understandable—the math is anything but straightforward. Simply comparing before and after expenditures won’t get you very far. It’s really apples to oranges when you start to consider the costs of labor, licenses, equipment, and myriad other line items.
That’s not to say that cloud won’t bring immediate operational value. It’s just that organizations are sometimes caught off guard by some of the expenses that come with the transition, or unsure how to go about measuring and characterizing the financial benefits.
With that in mind, let’s set the record straight about a few key issues related to cloud economics:
ROI is tied to much more than swapping capital expenditures (CapEx) for operating expenditures (OpEx).
It’s taken for granted that with cloud, you only pay for what you need, and that organizations should see a reduction in spending on data center leases, hardware, and licensing. But the real value comes from driving down the total cost of ownership for developing and maintaining applications: from operations and maintenance, to decreased cost per deployment, to leveraging as-a-service platforms and software. Patching, backups, and disaster recoveries are less labor-intensive when workloads reside in a well-architected cloud environment—and that’s on top of the value that comes with highly flexible cloud-native capabilities.
Costs can be deceptive without optimization.
Provisioning added capacity in the cloud takes not months, not weeks, but minutes—making it possible to optimize for cost efficiency in ways that simply can’t be done in a CapEx environment. Such speed and elasticity are some of cloud’s major advantages, but they can create an environment of overspending and over-provisioning if an organization isn’t savvy about the tradeoff between architectural decisions and cloud spend. Put simply, it’s critical to make sure that the infrastructure is optimized for the precise needs it’s intended to meet. This requires upfront work to nail down architectural decisions like how much virtual space to get, which instances should be reserved versus “on-demand,” how much data to archive versus store in a readily available state, what needs to be backed up, and what reporting mechanisms should be put in place to project costs into the future. These decisions are more than financial management 101 and they make a material difference on the monthly bill.
To truly scale value across an enterprise, automation is key.
The emergence and maturation of infrastructure-as-code and DevSecOps patterns resulted in significant cost savings in the development and maintenance of cloud platforms and applications. Whether standing up new environments, building out or modifying security architecture, creating a pipeline, or refining environment-wide observability and logging, organizations should take advantage of automation opportunities wherever possible, and account for the long-term savings in their ROI calculus. The cloud service providers (CSP) are bringing more automation and orchestration services to market every day, and are achieving accreditation for these services for use at the highest levels of security. Beyond the basic benefits that come with as-a-service cloud products, the integration of automated services removes otherwise manual and repetitive processes to streamline—and govern—capability deployment while improving security posture and platform stability.
“It’s taken for granted that with cloud, you only pay for what you need, and that organizations should see a reduction in spending on data center leases, hardware, and licensing. But the real value comes from driving down the total cost of ownership for developing and maintaining applications.”
4 Critical Steps to Drive Long-Term ROI
With out-of-the-gates elasticity, the power of scale, modern architectures, and more, it’s hard to overlook the value organizations can achieve in the cloud. But realizing that value requires a dedicated focus on enterprise optimization. Based on our experience helping hundreds of clients along the cloud value chain, here are four critical steps to drive long term ROI:
1. Get a Handle on the Current State of Your Enterprise Cloud Operations
Step one toward getting the most out of enterprise cloud is a thorough cataloging of what’s being used and who’s using it. What products, capabilities, and services are being billed by what programs, offices, and projects? How is storage and compute optimized against organizational needs? Are there opportunities to consolidate toward a core set of shared services?
Having a precise, up-to-date picture of your entire cloud presence helps identify areas where there’s considerable fat to trim and opportunities to standardize foundational solutions to serve the needs of teams and programs across the enterprise.
2. Deploy Advanced Financial Management Tools
Armed with a fully transparent view into its cloud universe, an organization can then optimize each environment with the most appropriate architecture. This upfront optimization has a dramatic impact on cost over time, since programs often err on the side of overprovisioning, and may not be taking advantage of flexible cost models.
With that in place, leadership still needs the forecasting to continue making informed decisions and adjustments over time. Every CSP provides basic reporting tools nowadays. They’ll let you set service alerts for when you hit a certain percentage of your budget allocation, for example. These are a good start, but to get to where you can forecast the weather rather than just observing it as it happens, you need to deploy more mature tools that provide enterprises with the ability to iteratively project cloud spend—a power that’s all the more important in multi-CSP ecosystems. Properly assessing the data against best practices can reveal both the forest and the trees, providing a granular look across an entire multicloud environment. This allows organizations to identify patterns, architect with precision, identify opportunities for less expensive service options (such as serverless functions), and stay ahead of any budget surprises.
3. Explore How Other Agencies Are Innovating with Shared Services
Across the federal landscape, more agencies are starting to centralize and standardize their cloud offerings to streamline process, build in technical governance, and provide valuable resources to teams across the enterprise. By giving everyone access to the same set of shared services, with the same standards and guardrails, innovation can happen with confidence in quality and security. For those researching the benefits of a consolidated cloud approach, there are now some standout examples to look to for guidance.
The U.S. Department of Treasury offers a shared services model with their Workplace Community Cloud (WC2) and Workplace Community Cloud-High (WC2-H). An early cloud adopter in government, Treasury established WC2 in 2010 as a shared service hosting platform for critical assets with moderate security needs. Building on learnings from WC2, the agency then expanded its offerings to include a FISMA-High-certified shared service platform, WC2-H, which can securely host Treasury’s most sensitive, mission-critical assets.
WC2-H offers shared services across all three of the major cloud service models: Infrastructure as a Service, Platform as a Service, and Software as a Service. By its own estimation, WC2-H’s first customer tenant garnered $3.7 million in annual cost savings after transitioning from a legacy on-premises application to a refactored data lake environment within WC2-H. It’s an impressive achievement and a model across government.
4. Map Out What Has Common Value Across Mission Teams
Moving toward a shared service model and eventually reaping the value that comes with consolidation requires a balanced consideration of technology and mission. What are the services that would really add value for people across the organization? What are the common denominators among projects and programs (e.g., identity management, content management, logging) that don’t need to be custom built?
It takes a deliberate approach to head off tech sprawl and meet the most needs with the fewest number of solutions and services. Leaders can get ahead of the curve by looking across the enterprise and engaging with individual departments to understand the common needs of major programs. It’s important to communicate that teams will still have the ability to customize and innovate, even as they leverage a core set of services and solutions.
The economics of cloud are complex, and they reward those willing to get into the weeds and untangle them. As much as we’d love it to be the case, there isn’t always an immediate financial windfall from cloud migration. But ultimately, there’s long-term and transformative value to claim from getting off premises, deploying scalable cloud-native capabilities, adding rigor to financial management, and gradually working towards a shared services approach that’s optimized for mission and maximizes value.
Sign up to receive CSA's latest blogs
This list receives 1-2 emails a month.