What is a managed service in the cloud?
A managed service is a cloud offering where the provider runs and operates the underlying infrastructure — provisioning, patching, scaling, backups, availability — so you get to use the capability itself without managing the servers behind it. It is the difference between renting a finished capability and being handed the raw materials to build it yourself.
Managed vs self-managed, in plain terms
Imagine you need a database. You could rent a plain virtual machine and install, configure, patch and back up a database engine on it yourself — that is self-managing on raw infrastructure. Or you could use a managed database service, where the provider runs the database engine for you: it provisions the underlying servers, applies patches, handles backups, and scales resources, and you simply connect and use it. Both end up giving you “a database,” but one hands you a finished, operated capability and the other hands you a blank machine and a set of responsibilities.
What the provider actually takes on
With a managed service, the operational work that keeps a system healthy — installing security patches, monitoring for failures, replacing failed hardware, taking backups, scaling capacity up or down — moves from your team to the provider. This is sometimes described as shared operational responsibility: you are still responsible for how you configure and use the service, but the provider is responsible for keeping the underlying infrastructure running. That is a genuinely different division of labour from renting raw compute, where almost all of that responsibility sits with you.
Common examples
Managed databases are one of the clearest examples: a managed relational or NoSQL database service handles the database engine’s operations so you interact with it as a capability rather than as software you installed. Managed Kubernetes is another — running your own Kubernetes cluster means operating the control plane yourself, while a managed Kubernetes service runs that control plane for you and leaves you to manage the workloads on top of it. The pattern repeats across most cloud categories: managed caching, managed message queues, managed file storage, and more.
The trade-off: less toil, less control
Managed services are not free lunches — the trade is real in both directions. You give up some control and flexibility: you generally cannot access or tune the underlying server the way you could on a machine you fully control, and you are constrained to whatever configuration options the service exposes. In exchange, you remove a large amount of operational toil — the routine, repetitive work of patching, backing up and monitoring infrastructure — which frees a team to spend its time on the application itself rather than on keeping servers alive.
Where this sits among cloud service models
“Managed service” cuts across the more formal IaaS/PaaS/SaaS categories rather than being identical to any one of them — our IaaS vs PaaS vs SaaS explainer covers that spectrum, and in general, the further along it you go, the more “managed” the experience becomes. It is also closely related to serverless computing, where the provider manages not just a fixed service but the entire execution environment, scaling it invisibly with demand; our serverless vs containers explainer looks at that specific comparison. Broadly, choosing managed services over raw infrastructure is one of the main ways teams take advantage of what cloud computing actually offers beyond “someone else’s servers” — see our what is cloud computing explainer for that bigger picture.
Original practice questions, timed mock exams and revision notes. No card, nothing to pay.
Questions, answered
Sources
Exam details in this post come from the vendor's published exam guide, which is the authority on what is tested and how.
- AWS Certified Cloud Practitioner (CLF-C02) exam guide — Amazon Web Services
- Microsoft AZ-900 study guide — Microsoft Learn
- Google Cloud Associate Cloud Engineer exam guide — Google Cloud