It may feel like a betrayal, but public cloud providers do need to sunset services at times. Here’s how to plan for it and even find an opportunity to improve. Credit: Life of Pix Over the years I’ve had a few friends and clients reach out to me, unhappy about some public cloud services being removed, either from a name-brand cloud provider or secondary players. At times, entire clouds were being shut down. The cloud providers typically give plenty of notice (sometimes years), calling the service “legacy” or “classic” for a time. They’ll have a migration tool and procedures to move to other similar services, sometimes to competitors. In some cases, they will pay for consultants to do it for you. As a tech CTO for many years, I also had to sunset parts or all of technologies we sold. This meant removing support and eventually making the technology no longer viable for the customer. Again, this was done with plenty of notice, providing migration tools and even funding to make the move to more modern and likely better solutions. Customers closely watched the way this was done and the number of times I sunsetted things in a year. Indeed, technology providers often screwed this up and made quick enemies out of customers, and I saw this ignite PR problems that often led to a quick sale of the company. Cloud computing, while providing a different consumption model, deals with the same issues. However, cloud technology does not exist in your data center, and when it’s removed, you don’t have the option to keep it running without support. During the normalization of the cloud computing market several years ago, services and entire cloud providers were removed or vastly changed often—sometimes with little or no notice from the provider—and applications stopped dead that had dependencies in those services. Fortunately, most enterprises had nonessential workloads on the cloud in those days, but today that’s no longer the case. There is a shared responsibility here. The cloud providers are obligated to provide cloud services in a reliable way as well as improve those services as required by the needs of the customer. They are also obligated to communicate with their customers, including informing them when services are changed or removed. The cloud users, in turn, are responsible for keeping up with information sent by their cloud providers, looking out for changes that may affect an application that has a dependency on a cloud service. Fortunately, most of these changes are backward compatible and may not force changes to applications that leverage them. Removing services is mostly rare these days. If cloud providers did it so much that their customers and guys like me noticed, they would find cloud users voting with their feet. This is pretty consistent with how technology has been supported in the past, either in the data center or in the cloud. The patterns are much the same. Therefore, planning for changes or removal of technology services, cloud or not, is the same. This includes understanding what and when changes need to be made. How they are to be tested and accepted? Also, what improvements can you take advantage of? Don’t be concerned when services are retired. We’re more than 10 years into this, and many public cloud services have gray whiskers and should be updated, replaced, or improved. This benefits both sides. You get updated and upgraded services. The cloud providers can move resources to more productive purposes. This is just a normal part of cloud operations. Related content analysis Strategies to navigate the pitfalls of cloud costs Cloud providers waste a lot of their customers’ cloud dollars, but enterprises can take action. By David Linthicum Nov 15, 2024 6 mins Cloud Architecture Cloud Management Cloud Computing analysis Understanding Hyperlight, Microsoft’s minimal VM manager Microsoft is making its Rust-based, functions-focused VM tool available on Azure at last, ready to help event-driven applications at scale. By Simon Bisson Nov 14, 2024 8 mins Microsoft Azure Rust Serverless Computing how-to Docker tutorial: Get started with Docker volumes Learn the ins, outs, and limits of Docker's native technology for integrating containers with local file systems. By Serdar Yegulalp Nov 13, 2024 8 mins Devops Cloud Computing Software Development news Red Hat OpenShift AI unveils model registry, data drift detection Cloud-based AI and machine learning platform also adds support for Nvidia NIM, AMD GPUs, the vLLM runtime for KServe, KServe Modelcars, and LoRA fine-tuning. By Paul Krill Nov 12, 2024 3 mins Generative AI PaaS Artificial Intelligence Resources Videos