4 Key Frameworks for Optimizing Cloud Strategy and Deployment

“40% of all enterprise workloads will be deployed in CIPS [cloud infrastructure and platform services] by 2023, up from only 20% in 2020.”^[1] As the cloud permeates every aspect of business, decision-makers must make critical choices regarding infrastructure at every turn. Their answers will ultimately determine if every part of an organization is empowered to move forward in a cohesive way to reach business outcomes.

In the HSIF session How To Rise Above the Fog and Soar Into the Cloud, Hitachi Vantara experts synthesize perspectives on cloud strategies from both IT and line of business. The goal is to help decision-makers align the considerations of the technical organization with the business drivers propelling the need to make these choices. I, along with Hitachi Vantara colleagues Senior Vice President and Head of Technology and Solutions Office, Premkumar Balasubramanian, and Chief Information Officer Tim Langley-Hawthorne propose four frameworks to help decision-makers organize their cloud considerations.

The Convergence of IT and Business Decisions

In the past, IT and line of business decision-makers were seen as distinct entities with separate values. Today’s CIOs, however, are challenged by business stakeholders to deliver more capability, deliver it faster, and contribute to business growth with new revenue-generating, technology-based products and services.

What is needed is a strategic approach that leverages the right mix of cloud technologies. The options for deploying cloud workloads may include any combination of:

• The private cloud, which has been applied in scenarios where hardening security, keeping pace with rising regulatory requirements, or standardizing and enhancing internal service delivery are priorities.
• The public cloud is famously favored by those seeking the flexibility to optimize their IT resource utilization and IT budgets, as workload needs change with time.
• Hybrid-cloud models that aim to let decision-makers choose among the best of both private and public clouds. This is generally the way to go for most businesses today.

An effective strategy will simultaneously meet the organization’s needs to provide access to and run workloads where you want them. At the same time, it will protect critical data, drive business outcomes, achieve sustainable return on investment (ROI) and deliver value to customers.

Aligning Around a Single Cloud Strategy

There is no single strategy that suits every enterprise, but there is a universal guiding principle. Line-of-business and IT decision-makers within a given enterprise have to agree on a single cloud strategy. That is the only way to effectively maintain and manage a distributed, hybrid cloud environment that can deliver fast results and increase business value. The following four frameworks will help give shape to your organization’s unique strategy.

Framework 1: Technology Drivers

Technology should be treated as an enabler for your future data center and the business applications it supports. Three primary drivers guide successful cloud journeys:

• Think simplification. An increasingly distributed IT environment can quickly become complex, resulting in further downstream technology complexity. It’s imperative to develop a technical and business strategy, and cloud road map. Breaking the complexity must be a top priority, and tools and technology should follow. Make tools and technology choices that help simplify things and enable the overall environment to operate efficiently.
• Maximize value. Not all clouds are equal and different cloud applications may have unique requirements that a particular cloud model best serves. It is essential to take an application-specific approach, determining the best cloud for any given application, based on that application’s architecture and the workloads it supports.
• Unify operations. As the organization settles on different clouds for its many workloads, it is critical that, together, they function as a seamless, unified environment.

Framework 2: The Three Cs of Cloud Mistakes

So how can IT avoid mistakes when it makes cloud choices? The following three areas require particular attention:

• Capabilities. Be sure that the cloud option chosen is suitable for the sought-after business capabilities. For example, the public cloud typically serves agile application and infrastructure capabilities especially well. On the other hand, where critical regulatory requirements or sensitive data are considerations, a private cloud may be more suitable. Forcing a workload where it doesn’t fit may lead to frustration at the very least. It is best to take a portfolio “application-first” approach to your cloud strategy.
• Cybersecurity. Consider the criticality of the individual workflows and data pipelines in terms of risk and compliance as it relates to data moving in and out of the cloud. Use that as a constant reference when evaluating your cloud options and where your applications need to live.
• Cost. A deep, sophisticated, and long-term view of cost is important. Consider characteristics such as usage, data retention timeline and data retrieval needs. Contrary to popular belief, the public cloud may not always be the most cost-effective answer. For example, some providers, such as Hitachi, offer consumption-based pricing. This approach allows an organization to pay only for the services it uses at predetermined, published prices and with elastic availability to meet evolving needs now and in the future.

Framework 3: Optimize Business Impact

Regardless of where one constructs future “data centers” using cloud services, pay careful attention to how individual applications fit the cloud and affect business.

• Take an application-specific view. To determine the best cloud to suit a given application, consider the nature of the application first. How complex is it? How business critical? Does it undergo frequent modification? What are its needs concerning performance, scalability, fault tolerance, and so forth? Only then is it time to look at cloud options. Think of the future as being made of “floating data centers” that can be anywhere. What will your applications require to provide a seamless view of that setting? Needs change, so having the ability for applications to rapidly onboard and exit cloud platforms may be important.
• Lift and shift isn’t the only way. The notion of moving capital and operational expenditures (capex and opex) by just shifting applications from on-premises locations to a cloud is proving to be more difficult than first thought. One recent survey found that 43% of IT decision-makers  learned that the cloud is more costly than they thought. Refactoring applications first by making them container-friendly, for example, might go a long way toward achieving those cost benefits.
• Aim for a single, coherent management plane. The journey to the cloud must be part of a larger plan that also includes better cybersecurity, enhanced governance and adopting modern engineering principles that support agility, reliability, DevSecOps and closely managed spending (also known as FinOps). Even with the best efforts to simplify things, running ever more workloads across multiple clouds means that a robust RunOps capability will be increasingly important.
• Increase automation. Again, managing complexity is critical. A robust management plan should provide 360-degree observability, as well as extensive automation to help manage potentially thousands upon thousands of virtual machines, containers and serverless functions.

Framework 4: Plan for Long-Term Business Goals

• Develop a FinOps mentality. As organizations democratize access to the public cloud and decentralize their IT, it’s easy for spending on public cloud services to increase quickly and unexpectedly. It’s critical to have multicloud cost management and an optimization platform built into the DNA of the entire cloud deployment to manage costs effectively.
• Compliance. Depending on your company’s industry, geographic location and business function, there could be a range of compliance regulations that can impact your company. Having a tool that allows you to build, manage, modify and ensure compliance across private and public cloud infrastructure is essential. It is the only realistic way engineers can deploy and manage infrastructure code regardless of a workload’s destination.
• Train, attract and manage talent. Invest in learning and development to keep cloud-native employees engaged. Retrain and develop valued staff not “born” to the cloud-based world. Embrace a product engineering culture that brings business, IT and engineering personnel together as a single DevOps organization.
• Prioritize resilience. Operations alone can’t guarantee resilience. It must be built into applications through intelligence that makes the applications fault-aware, fault-tolerant and self-healing. These criteria are often treated as secondary to performance. Smart, self-protective applications are essential to business in a highly distributed, always-on setting.

Putting Frameworks Into Action

Ultimately, every enterprise has to construct its own strategy. Maintaining and managing distributed hybrid-cloud environments and generating tangible business value from them will require thoughtfully operationalized processes. It may be beneficial to work with a partner with business knowledge within your specific industry and broad experience of hybrid-cloud deployment to arrive at the best fit. The frameworks provided in this article are the first step. Use them to help align thinking throughout your organization as decision-makers evaluate cloud solutions for their suitability to meet your business outcomes.

^[1] Smarter with Gartner,  Gartner Predicts the Future of Cloud and Edge Infrastructure, Katie Costello, February 8, 2021.