What is distributed cloud, and how does it facilitate edge computing?
Earlier this year, we added distributed cloud as another cloud deployment option on our cloud architecture heatmap slide. When it all started, we had two cloud deployment models: public and private. Then there were three: public, private and hybrid. Now we have five cloud deployments models: public, private, hybrid, multicloud and distributed (see Figure 1 below):
Given that we know it is a new cloud deployment model, what exactly is distributed cloud? Is it yet another combination of the public and private? What makes this cloud deployment model so different? And how does it relate to edge computing?
This blog post will address these and other related questions by looking at specific use cases.
Please make sure to check out all the installments in this series of blog posts on edge computing:
- Part 1: “Cloud at the edge”
- Part 2: “Rounding out the edges”
- Part 3: “Architecting at the edge”
- Part 4: “DevOps at the edge”
- Part 5: “Policies at the edge”
- Part 6: “Models deployed at the edge”
- Part 7: “Security at the edge”
- Part 8: “Analytics at the edge”
- Part 9: “5G at the edge”
- Part 10: “Clusters at the edge”
- Part 11: “Automation at the edge”
- Part 12: “Network slicing at the edge”
- Part 13: “Data at the edge”
- Part 14: “Architectural decisions at the edge”
- Part 15: “GitOps at the edge”
- Part 17: “Storage services at the edge”
- Part 18: “Cloud services at the edge”
- Part 19: “Distributed cloud: Empowerment at the edge”
- Part 20: “Data sovereignty at the edge”
- Part 21: “Solutioning at the edge”
- Part 22: “Connected products at the edge”
- Part 23: “Foundational models at the edge”
What is distributed cloud?
In a distributed cloud, public cloud services are made available to consumers at different physical locations outside of the cloud provider’s facilities, also known as satellite locations. The public cloud provider is responsible for the operation, governance and updates of the services. Distributed cloud computing extends the range of cloud use cases all the way to the edge. Thus, an enterprise using distributed cloud computing can store and process its data in different data centers that may be physically located in different remote locations.
To paraphrase from an IBM paper on distributed cloud, it is public cloud computing that lets you run public cloud infrastructure in multiple different locations — not only on your cloud provider’s infrastructure, but on-premises, in other cloud providers’ data centers and in third-party data centers or colocation centers — and manage everything with a single control plane.
The promise of managing everything from a single control plane is what makes distributed cloud compelling. Remote locations, a single control plane and a secure network tunnel are the key components in a distributed cloud offering.
Distributed cloud empowering the edge
In this blog series, we have discussed nuances of edge computing. In a previous blog called “Cloud Services at the Edge,” we alluded to distributed cloud or a “cloud-out” notion wherein the computing resources of the data center and cloud services are now available at the edge. That allows for processing and analysis of data at the source where the data is generated. Distributed cloud provides credence to that “cloud-out” thinking by bringing those remote locations in focus.
Distributed computing empowers edge computing by bringing the power to process large amounts of data due to the physical proximity of the data sources, while maintaining data security and compliance aspects. Edge computing can be — and has been — implemented without a distributed cloud architecture, but distributed cloud makes edge application deployment and management a lot easier, especially when we deal with the telco edge or the enterprise edge.
Distributed cloud use cases
It is easy to envision many use cases, but we highlight two in this section that we think are rather unique given the complementary nature of distributed cloud and edge computing. Figure 2 highlights the areas that befit this paradigm.
We will use the IBM Cloud Satellite product to showcase these solutions. The remote locations we alluded to — Satellite locations — can be managed from a common control plane. IBM Cloud Satellite does this by attaching remote infrastructure in the Satellite location to the IBM Cloud Satellite control plane. The Satellite location can then be used by IBM Cloud services as deployment targets. Remember, a Satellite location can even be an environment hosted on another hyperscaler like AWS, Azure or Google Cloud:
Meeting industry or local data residency requirements
Also known as data sovereignty, many industries and countries have regulations that specify that a user’s personal information (PI) cannot leave the user’s region/country. Distributed cloud infrastructure makes it easier for an organization to process PI in the user’s country of residence. This is especially useful in regulated industries like healthcare and in many European countries.
Figure 3 depicts one such distributed cloud architecture involving a hypothetical healthcare provider. The healthcare provider has multiple clinics, hospitals and diagnostic centers. Patient data confidentiality and data residency regulations dictate that the data cannot be moved out of the provider’s individual locations. The data is accessed by the medical staff comprised of doctors, nurses, research teams, diagnostic teams and the patient themself.
Patients can visit any facility operated by the healthcare provider. This introduces the need for a fast and secure communication channel for the doctors and research partners to allow them to provide timely medical care and up-to-date information to facility staff and patients.
The location staff expect a secure solution that allows control of the data within their private cloud, while providing cloud-native capabilities locally. The solution also complies with data security and local data privacy requirements. The solution also needs to be operated by the provider’s staff within their own data centers.
The distributed cloud instances use a Red Hat OpenShift-based Kubernetes platform and a secure cloud-native messaging application based on event-driven architecture provided the low-latency message platform:
Solution highlights
- The health provider designates a hospital data center as the on-premises Satellite location.
- Telemetry data from various devices monitoring the patient is fed to the Internet of Things (IoT)/edge application running at the Satellite location.
- Data is analyzed at the Satellite location in close proximity to patient and medical staff.
- Hospital medical and patient records are stored on-site and cannot be moved out.
- Hospital staff leverage a fully managed OpenShift platform’s capabilities, including end-to-end encryption.
- The health provider has plans to add another hospital as the next Satellite location and will use the same public cloud services.
Multi-access edge computing (MEC)
Telecoms or communications service providers (CSPs) have been looking at ways to monetize 5G technology by providing better multi-access edge solutions. A distributed cloud topology enables this offering by allowing CSPs to offer their customers single tenancy in the Satellite locations and bring computing power and MEC services to those premises. This, in turn, helps address data security concerns while delivering the edge applications that demand low latency. In summary, the telco operator uses its 5G network and a distributed cloud to provide MEC services.
Consider the scenario where a CSP is offering its services to an auto manufacturer that has multiple plants in a 100-mile radius. Parts and other assembly-line data is shared between those plants. The CSP can provide a distributed cloud solution where MEC services are provided at each plant, acting as a Satellite location and providing a secure and consistent view of all the relevant manufacturing data to the apps running in those plants. These could be visual inspection apps that inspect painting or welding of parts or analytical apps that provide analysis in real-time or near-real time with the help of 5G technology.
Another use case along those lines is the CSP servicing two different auto manufacturers in an industrial city. A distributed cloud solution using IBM Cloud Satellite, shown in Figure 4, allows the CSP to offer its customers single tenancy in remote locations. Each customer can run workloads where they want with complete observability, and all their data is totally secure and isolated:
Solution highlights
- Deploy edge applications at multiple Satellite locations.
- CSPs can monetize existing networks at edge locations and offer new services based on 5G.
- Leverage fully managed platform capabilities, easing the burden on telecom providers.
- Customers get identity and key management services
- With a single pane of glass, customers also get observability that includes central logging and monitoring for apps and the platform.
Wrap up
Distributed cloud enables public cloud providers to offer an entire set of services wherever a customer might need it — on-premises in the customer’s own data center, in a private cloud or off-premises in one or more public cloud data centers that may or may not belong to the cloud provider.
The use cases we described show how distributed cloud and edge computing complement each other. Among other benefits, solutions using these combined technologies provide low latency access to on-premises systems, local data processing, and even local data storage which is especially useful for running AI workloads at the edge.
Do let us know what you think. Special thanks to Joe Pearson and Gerald Coon for reviewing the article.
Please make sure to check out all the installments in this series of blog posts on edge computing:
- Part 1: “Cloud at the edge”
- Part 2: “Rounding out the edges”
- Part 3: “Architecting at the edge”
- Part 4: “DevOps at the edge”
- Part 5: “Policies at the edge”
- Part 6: “Models deployed at the edge”
- Part 7: “Security at the edge”
- Part 8: “Analytics at the edge”
- Part 9: “5G at the edge”
- Part 10: “Clusters at the edge”
- Part 11: “Automation at the edge”
- Part 12: “Network slicing at the edge”
- Part 13: “Data at the edge”
- Part 14: “Architectural decisions at the edge”
- Part 15: “GitOps at the edge”
- Part 17: “Storage services at the edge”
- Part 18: “Cloud services at the edge”
- Part 19: “Distributed cloud: Empowerment at the edge”
- Part 20: “Data sovereignty at the edge”
- Part 21: “Solutioning at the edge”
- Part 22: “Connected products at the edge”
- Part 23: “Foundational models at the edge”
Learn more
- What is Distributed Cloud?
- IBM Cloud Satellite
- IBM Edge Application Manager
- IBM Cloud Pak for Network Automation
Related articles
Was this article helpful?
YesNo