Cloud computing technology has evolved rapidly over time, and so too has the vocabulary we use to describe it. At Equinix, our How to Converse in Cloud series aims to help business and IT professionals keep up with the latest developments in cloud, including the changing terminology that goes along with them.
In the early days, a “cloud” was simply any compute instance running in an off-premises location. The early cloud was also subject to the capacity limitations of the underlying hardware. Now, things have changed: Hyperscalers offer customers flexibility to scale up capacity on demand, along with a multitude of integrated cloud-based services to supercharge digital transformation.
In this new reality, the possibilities of what enterprises can achieve with cloud services are practically endless. However, today’s cloud users also face new challenges. Notably, while the hyperscale cloud is flexible and scalable, it’s definitely not cheap.
Further, the idea of hyperscaler-led vendor lock-in is a well-known and much-discussed phenomenon. Getting the full benefits of cloud—security, high availability, the ability to adapt to an ever-changing regulatory landscape, the agility to incorporate new technologies and services as they come to market, and more—requires a cloud-agnostic approach, allowing users to draw intelligently, efficiently and adaptably from best-of-breed services across different clouds. This means it’s more important than ever for enterprises to avoid vendor lock-in and minimize the impact of allowing data gravity—the tendency for large data sets to draw in workloads—to anchor them to a single hyperscale vendor for some or all of their business.
Enterprises can avoid these downsides to the cloud by deploying and leveraging dedicated cloud as a component of their cloud architecture. Simply put, dedicated cloud is the next evolution of private cloud.
Both “private cloud” and “dedicated cloud” are terms used to describe single-tenant compute, storage and analytics instances deployed outside traditional on-premises environments. However, unlike private cloud, dedicated cloud combines single tenancy with access to services from multiple cloud providers. Said another way, a dedicated cloud environment provides single-tenant compute resources in a neutral “staging ground” that allows businesses to move data and workloads between clouds quickly, efficiently and cost effectively.
Digital transformation trends highlight the growing importance of dedicated cloud
The shift from private cloud to dedicated cloud is part of a wider trend of digital transformation. Businesses are deemphasizing capital assets (CAPEX) in favor of as-a-service consumption models (OPEX). This shift makes it easier for businesses to access the services they need with decreased time to value, while also allowing them to use their resources more efficiently.
To minimize application latency and costs in this new model, enterprises would ideally store as little of their data in the hyperscale cloud as possible. Cloud vendors typically make it very easy to move data to their clouds, and very difficult and expensive to move it away. This is because they know that once data is stored with them, data gravity will inevitably drive greater consumption of their surrounding compute and application services. Combined with punitive data egress fees and other hidden costs of cloud, this is what drives vendor lock-in and leaves customers saddled with astronomical cloud bills, limited flexibility and choice, and few options for how to address those issues.
In contrast, many enterprises are now using cloud-adjacent infrastructure to help them get as close as possible to all the clouds without committing their entire data estate to a particular provider or providers. Instead, they can move data freely to those clouds or any other destination as needed to best support their workloads and take advantage of specific cloud services.
Dedicated cloud provides that neutral staging ground that makes meaningful multicloud possible. The value in doing so is reinforced by the Equinix analysis published in the Leaders’ Guide to Digital Infrastructure e-book, which showed that dedicated cloud can drive operational savings of 30-40% by eliminating cloud connectivity overhead costs and accelerating the shift from CAPEX to OPEX.
How to simplify dedicated cloud deployment
It is possible to take a do-it-yourself approach to dedicated cloud by building the staging ground you need and hosting it with a colocation provider such as Equinix. Businesses looking to get started with dedicated cloud typically follow these five basic steps:
- Deploy dedicated compute in a colocation center.
- Integrate that compute into a hybrid infrastructure environment.
- Run cloud-agnostic application performance and availability management.
- Orchestrate resources with a DevOps toolchain integration.
- Scale hybrid workloads across multiple clouds.
However, there’s another way. Digital services from Equinix can help customers deploy dedicated cloud in a more efficient, “cloud-like” manner. For instance, Equinix Metal®—which already offered the ability to consume digital infrastructure at Equinix in minutes, not months—has expanded beyond its roots as a Bare Metal as a Service (BMaaS) solution to offer more cloud-like functionality. This includes integration with Equinix Fabric® for software-defined interconnection, thus helping simplify hybrid multicloud networking.
Equinix Metal gives you the flexibility to deploy compute capacity in distributed locations throughout the world, while also integrating that compute with interconnection services for on-demand multicloud access. In short, the Equinix Digital Services portfolio can help you get both elements of dedicated cloud—the single-tenant compute and the multicloud access—with greater efficiency and agility.
The diagram below illustrates a cloud-adjacent infrastructure. At the heart is the “authoritative data core”, which stores copies of data sets aggregated from across your distributed operations. Interconnection services help move those data copies wherever they’re needed, providing connectivity both to the single-tenant compute and to the clouds of your choice.
Learn what’s next for cloud
Just as dedicated cloud describes the latest evolution in private cloud, “supercloud” is a new term that describes the next step in the evolution of hybrid multicloud. While hybrid multicloud means simply running workloads on different clouds, the idea behind supercloud is that what customers really want is a cloud abstraction layer that can determine which workloads should run where, and quickly and seamlessly move those workloads between the various hyperscale clouds and dedicated cloud environments as the need arises.
Today, supercloud is still more ideal than reality, as it requires software solution builders to engineer that abstraction layer for their applications, taking the complexity of reengineering workloads out of customers’ hands. Check back soon for our follow-up blog where we’ll provide a deeper look at the concept of supercloud, including further contemplation and evaluation around how dedicated cloud is sure to play an essential role in making the dream of supercloud a reality.
See how digital leaders are thriving in a changing landscape
Dedicated cloud is only one example of how digital leaders are adapting to maximize agility and succeed in the modern economy. For a look at some of the other common patterns they’re following, read the Leaders’ Guide to Digital Infrastructure e-book.