As enterprises look to capitalize on the scalability and agility of cloud computing without completely moving outside the firewall, hybrid cloud has become an increasingly popular option. And because hybrid cloud combines private cloud and public cloud services, public cloud providers are vital to any hybrid cloud strategy. Public clouds run business applications, store business data, enable cloud bursting and scale on-demand when computing needs change. But all public cloud providers are not created equal. Each offers unique services at different price points and performance levels.
For example, the same workload won't receive the same virtual machine (VM) instance, cost or performance level from Amazon Web Services (AWS), Microsoft Azure, IBM Cloud or Google Compute Engine. Performance levels and services also vary based on a provider's location. Selecting a provider in a local region tends to yield better performance and reliability than a public cloud provider in a distant location. To understand downtime's impact on hybrid cloud workloads, pay close attention to service-level agreements.
Other considerations, such as workload compatibility and migration, influence provider selection. For example, IBM shops usually opt for IBM Cloud to streamline workload migration. Similarly, Microsoft Azure typically caters to Windows environments better than Google Compute Engine.
The risk of tying workloads to a single public cloud
Ideally, public clouds are interchangeable utilities; if one public cloud provider falters or another offers a better deal, users should be able to seamlessly migrate workloads from one public cloud to another. But public clouds are hardly ubiquitous, and workloads are typically coded -- or at least optimized -- to use a specific provider's APIs and stack configuration. Meanwhile, some public cloud tools only work with specific VM file formats.
Unfortunately, these limited coding and tool choices can make it difficult to migrate a workload between providers without new tools or additional software development. Hybrid cloud operators face an arduous choice: Find another compatible public cloud; invest more resources to accommodate a different public cloud; or bring workloads back on-premises -- an undesirable option, since private cloud resources tend to be thin.
To avoid vendor lock-in, businesses can develop workloads that are compatible with multiple APIs or public cloud environments. Additionally, businesses can develop workloads around commonly-accepted cloud software frameworks, such as OpenStack and its RESTful API. Using cloud providers that adhere to Open Virtualization Format 2.0, which enables translation between proprietary VM formats, also offers flexibility.
Stephen J. Bigelow is the senior technology editor of the Data Ceter and Virtualization Media Group. He can be reached at [email protected].
Hybrid cloud market still in IBM's sights
How to create a strong hybrid cloud governance plan
VMware, AWS battle over hybrid cloud management
Dig Deeper on Building and maintaining a hybrid cloud
Related Q&A from Stephen J. Bigelow
Learn how load balancing in the cloud differs from a traditional network traffic distribution, and explore services available from AWS, Google and ... Continue Reading
Access management is critical to securing the cloud. Understand the differences between AWS IAM roles and users to properly restrict access to AWS ... Continue Reading
Containers have rapidly come into focus as a popular option for deploying applications, but they have limitations and are fundamentally different ... Continue Reading