Private Cloud vs. Public Cloud
Utilities are faced with some formidable challenges – aging infrastructure and applications, supported by an aging workforce; a new generation of consumers with higher than ever expectations for self-service and innovative product offerings; new and increased regulatory pressures, and pressure to increase operational efficiency while reducing O&M costs. Combine that with increased pressure to maintain revenues when big box and online retailers are selling energy generation technology – like solar panels – straight to consumers and it’s easy to see why utilities must begin to take a different and more aggressive approach with modern platforms, capabilities and services. The rise in cloud capability, with multiple deployment and service models, make these challenges well within the range of manageable, even for mission critical applications like CIS and ERP. An updated business and technology strategy, thoughtfully including cloud, is a meaningful first step.
Increase in cloud adoptionIncrease in cloud adoption
Utilities have seen a significant increase in cloud adoption in just the past 3 years.
TMG Consulting, a well-respected thought leader focused on CIS solutions in the utilities industry, commissioned their research arm to explore cloud trends in the utility industry and found that executives are serious about their cloud strategy – 77% of utility executives surveyed in 2019 are open to using a cloud-based CIS solution or hosting it in the cloud.
Other utilities have already adopted a cloud first strategy and some have boldly stated that they will be 100% in the cloud within a few years.
The case for cloud is clear and the industry is headed that way. But what does that really mean? Solution providers are ready to respond – they’re rapidly adopting a cloud delivery approach for even the most mission-critical utility applications like CIS and all claim to have cloud offerings ready to consume. Ready or not, the cloud is here and it’s crucial for you to develop a cloud strategy with an understanding of the benefits and implications of the various cloud models. Going to the cloud means new ways to contract and manage your technology solutions and requires an intentional plan to transition your organization to your chosen deployment and service model.
The first step in developing a sound cloud strategy is having a solid understanding of what cloud is. Some of you already have a strong footing in this understanding. If so, I encourage you to come back for future installments of this series, where we’ll dig deeper into topics like accounting capitalization and rate recovery – two very different things sometimes classified together; security and operations in detail; and cloud maturity. If you’re still still scratching your head about how the various cloud deployment and service models match your specific business needs, or simply want a refresher, read on.
What is cloud?
In it’s simplest terms, cloud is a multi-tenant computing environment with shared infrastructure – things like servers, storage, networking, applications, and analytics – delivered as a service. While the concept is simple, thoughtful application of these concepts to an overall cloud strategy for your utility will help identify how best to reap the touted benefits of cloud – all within your defined risk tolerance. There are a few concepts to keep in mind that makes cloud unique from other outsourcing or data center hosting arrangements. Specifically, the National Institute for Standards in Technology (NIST), says it’s cloud when:
Resources are available on demand, via self-service. This means an IT contact or business owner within the utility can provision the computing environment automatically.
Capabilities are accessed over the network. Your computing capabilities can be accessed over a network or the internet with any number of platforms – desktops, laptops, mobile phones, or tablets.
Resources are pooled. The computing resources serve multiple tenants or applications and are dynamically assigned and reassigned across available resources as demand fluctuates.
Resources are elastic. Resource elasticity simply means an ability to provision and release computing resources as demand fluctuates – giving the appearance of unlimited, on demand capacity.
Services are measurable. Cloud capability is offered as a service, and thus, must be accompanied by an appropriate metric by which to monitor, control, and report usage.
Let’s begin by taking a look at private versus public cloud.
Many people think of private cloud as a virtualized on-premise computing environment. Others think of it as simply data center hosting of a virtualized computing environment. While this is the beginning of the story, it doesn’t end there. Private cloud computing takes virtualization a step further, to include the key capabilities discussed above that earn a cloud designation.
The National Institute for Standards and Technology (NIST), defines private cloud computing in this way: “The cloud infrastructure is provisioned for exclusive use by a single organization comprising multiple consumers. It may be owned, managed, and operated by the organization, a third party, or some combination of them, and it may exist on or off premises.” You may hear private cloud referred to as enterprise cloud or corporate cloud.
In utility industry specifics, the private cloud would be provisioned for use by a single utility corporation or municipality, with multiple tenants. Those tenants could be wholly owned subsidiaries, business units, different operating companies or service/regulatory jurisdictions. This deployment model allows the utility to pool resources (compute, storage, network bandwidth) and create a consumption-based offering specifically and exclusively for the utility. Private cloud offers a consolidation of hardware and potentially even data centers, resulting in reduced total cost of ownership for infrastructure. The organization may opt to locate hardware off-premises, with infrastructure maintained on a private network, further reducing cost of ownership.
Operating costs are reduced as well, although not eliminated. A private cloud computing model still requires the organization to provide for the planning, deployment, and management of the environment – either with internal resources or by procuring the services of a managed services organization with the skills to support computing environments matched to the utilities’ portfolio of applications.
In this deployment model, the client retains responsibility for securing the environment up to their point of ownership. In all cases, data protection and the supporting policies and procedures remain the responsibility of the customer. If the infrastructure is on-premise, physical security and access to the data center will be the responsibility of the utility. Locating the infrastructure off-premise provides the benefit of shared physical security and access control for the provider data center. Considerations for off-premise deployment include network connectivity to the off-premise data center as well as defining access policies and procedures for support resources provided either internally or via 3rd party managed service providers.
According to Forrester Consulting, a private cloud can offer impressive returns on investment. Their Total Economic Impact (TEI) study examined the ROI results of multiple organizations that adopted a private cloud. The composite result was a 111% ROI, with a 13.2-month Payback Period. Dell EMC, a provider of private cloud infrastructure and services, has also conducted ROI analyses across dozens of organizations in multiple industries, including energy. They evaluated different types of cloud transformations – moving to private cloud, adopting public cloud, and moving to hybrid cloud – within organizations of varying sizes and maturity levels. Their findings align with Forrester, with savings of 18-26% on average and as high as 45% in some cases.
The private cloud deployment model is ideal for utilities who are eager to take advantage of the benefits of cloud computing and who prefer to retain more control over the planning, deployment, and operations of the environment. This model offers reduced total cost of ownership (TCO) and decreased technical complexity, while still providing flexibility for customization of infrastructure to meet specific requirements for performance and compliance.
In a private cloud environment, infrastructure is provisioned for the exclusive use of a single organization. In contrast, multiple organizations – or tenants – share computing resources in a public cloud. Let’s start with the NIST definition:
“The cloud infrastructure is provisioned for open use by the general public. It maybe owned, managed, and operated by a business, academic, or government organization, or some combination of them. It exists on the premises of the cloud provider.”
A more commonly and easily understood definition of public cloud is cloud infrastructure that is made available to multiple consumers (i.e. tenants) – utilities and other types of businesses – over the public internet. Resources such as storage, virtual machines, or applications are shared and available at a scale not typically possible for a single utility in an on-premise or even private cloud scenario. You might think of iCloud or Microsoft365 when someone says public cloud. While these types of public cloud models are well-known, there is also a variation of public cloud that can work even for mission-critical applications, and at a price point that can fit almost any budget.
In the public cloud, the computing environment is highly standardized and automated. While this makes scale almost infinite, with little upfront cost to the utility, customization is limited. From an operational perspective, the infrastructure is managed by the cloud service provider, relieving the burden of finding resources or hiring a managed service provider for the care and feeding of the computing environment. There are a number of service models (e.g., IaaS, PaaS, or SaaS) that determine ownership and cost of operations for databases and applications. These concepts will be discussed in detail in later installments.
Security is often cited as a concern for the public cloud. Public cloud providers take security seriously and hold a number of certifications like SOC-2, COBIT, HIPPA, and PCI, which can help utilities feel a level of assurance that proper care is taken to protect proprietary and customer data and ease the burden of compliance through inherited controls. Still, security in the cloud is a partnership between the utility and the cloud service provider. The shared responsibility model is dictated by the service delivery model that we will explore in greater detail later in the series.
Virtual private cloud (VPC) is a concept within public cloud computing that brings the best of both private and public cloud. A VPC allows your utility to take advantage of the benefits of being in a multi-tenant environment – lower costs and near-instant scalability via shared infrastructure – with the safety of operating on a private network that isolates your utility’s virtual resources from other tenants. This makes running mission-critical workloads in the public cloud a more viable option, but there are still a few considerations to think through. While the public cloud offers highly resilient, highly scalable infrastructure solutions, which makes availability with most public cloud providers a non-issue, oversubscription in a multi-tenant environment could still pose a concern for performance. A couple of key considerations when contemplating public cloud and VPC for your organization will be the availability service level and what, if any, service level is proposed for application performance.
From an overall ROI, it’s easy to assume returns equally as impressive as those cited above for private cloud. One thing to be cautious of when moving enterprise systems to public cloud is attempting to duplicate on-premise or private cloud architecture in the cloud. Many companies have done this as a protective measure to ensure adequate sizing for performance purposes. This is certain death to your ROI. Work together with your cloud architects to right size your environment to achieve optimal benefits from both a cost and performance perspective.
Both private and public cloud offer benefits to utilities and each comes with considerations that must be evaluated in the context of an overall cloud strategy for your utility. This installment just scratches the surface of the differences between private and public clouds, and the potential implications for utilities considering cloud for mission-critical enterprise applications.
In future installments, we’ll continue to explore these concepts, in the context of multi-cloud and hybrid cloud. We’ll also discuss the concept of community clouds and how that may be an option for utilities. We’ll also dig deeper into security and operations in detail. These topics further expand on the topics above and are important for building a well-rounded cloud strategy for your utility.
Kim Schafer is a founding partner and CEO at Validos. She brings a passion to helping organizations, especially Utilities, understand the dramatic and ever-evolving cloud deployment options and their impacts. Questions, comments, feedback? Email Kim@validos.com or send a note via LinkedIn.