If you build a hybrid cloud using infrastructure in colocation data centers, the most obvious use cases for it center on basic IaaS services, like compute and storage. Those hybrid cloud services are certainly important, enabling you to burst capacity temporarily, for example, or fail over to the cloud. But to get the most out of a hybrid cloud -- especially one that takes advantage of colocation -- it may be useful to think beyond simple virtual machines and object storage services.
Here’s a look at other types of hybrid cloud services that may offer special benefits in the context of colocation.
What Is a Hybrid Cloud Service?
In this context, hybrid cloud services refer to the types of cloud services you run in a hybrid environment.
Again, the most basic type of cloud services you can run in a hybrid cloud include compute services that let you deploy virtual machines and storage services for your data. These are almost certainly the most widely used types of cloud services today, but they represent only a fraction of the types of cloud services you can deploy in a hybrid environment.
Hybrid Cloud Services for Colocation
Other examples of cloud services that you might consider using include:
Desktop-as-a-Service
Desktop-as-a-Service, or DaaS, is a cloud service that delivers access to desktop environments hosted in the cloud. It is not a new technology, but it has enjoyed a surge in usage during the pandemic, which has led more companies to place a premium on cloud-based desktop environments that can be accessed from anywhere, without being tethered to specific physical PCs.
Most DaaS solutions today rely on public cloud services like Amazon WorkSpaces and Azure Virtual Desktop. A major drawback of these offerings is that the servers hosting desktop environments are in far-off public cloud data centers, which can produce latency issues.
A colocated hybrid cloud environment can address this challenge by enabling the deployment of cloud desktop servers closer to end users while still keeping the cloud desktop infrastructure in the cloud (which is important because locally hosted virtual desktop infrastructure can be less reliable).
For now, running DaaS in a hybrid cloud environment is somewhat complicated, because DaaS services are not yet part of the major hybrid cloud frameworks. You’d therefore most likely need to rely on a bespoke solution. However, Micorosft says that support for Azure’s cloud desktop solution within Azure Stack is in the works, which would make it easy to host Windows cloud desktop environments inside a colocated hybrid cloud.
Backup-as-a-Service
Cloud backup services are another example of a workload that you don’t want to run locally but want to keep close to your end users to ensure high performance of the network on which your backup routines rely. That makes Backup-as-a-Service another good candidate for a hybrid cloud hosted in a colocation data center.
Here, again, data backup services are not part of the service offerings of the major hybrid cloud frameworks, at least for now. But it’s simple enough to build a backup solution by pairing a hybrid cloud storage service with a backup and disaster recovery software platform.
Bare-Metal Compute
Typically, compute services that run in both public clouds and hybrid clouds focus on virtual machines. Those are fine for most situations where you need to deploy a server in the cloud.
Sometimes, however, you want to be able to access a bare-metal server environment as a cloud service because you need to run workloads that don’t work well on top of a hypervisor, or you want to avoid the performance drawbacks of virtualization.
You can use bare metal in one of the hyperscale public clouds, but those come with a price premium. It will likely be more cost-effective to deploy bare-metal services inside a hybrid cloud environment that exposes bare-metal servers as compute instances. And if you host the servers in a colocation facility, you gain the reliability benefits of hosting them off site.
Cloud Data Analytics
All of the major public clouds now offer sophisticated cloud-based data analytics services. These services typically make it easier to run data analytics than it would be if you had to set up your own big data environment on premises. The drawback is that cloud-based data analytics can be quite expensive, especially when you factor in all of the data storage and egress charges that cloud vendors pile on top of the costs for the analytics services themselves.
This means that moving data analytics to a colocated hybrid cloud could be a great way to reduce costs while still enjoying the scalability and simplicity of cloud-based analytics. It’s also easy to do: Hybrid cloud frameworks already enable certain types of analytics services. AWS Outposts supports Amazon EMR, for example, and Google’s Anthos-based BigQuery Omni platform supports hybrid architectures. And some colocation providers, such as Digital Realty, have pre-designed products for such use cases.
Conclusion
As hybrid cloud platforms continue to evolve, it’s becoming easier to deploy more diverse types of cloud services on them. That’s great for hybrid cloud users, who can take advantage of a greater array of solutions to build more flexible, feature-rich hybrid environments. At the same time, it may offer opportunities for colocation providers to provide or support more types of hybrid cloud services within their facilities, making colo that much more attractive for organizations with a hybrid cloud commitment.
