AWS, Azure & GCP · Cloud across providers
How do IaaS, PaaS and SaaS differ?
Short answer
IaaS gives virtualised infrastructure you manage; PaaS gives a managed runtime where you deploy code; SaaS gives a finished application. Higher abstraction means less operational work and less control.
The models are a spectrum of managed responsibility. Moving up the stack removes operational work but also removes levers you may need for custom networking, kernels, agents or licensing.
What you operate
- IaaS: you manage OS patching, runtime, scaling agents and network configuration on top of provider hardware
- PaaS: the provider manages servers, scaling and patching; you manage build, configuration, data and access
- SaaS: the provider manages the whole application; you manage users, permissions and data usage
- FaaS: you manage individual functions and their permissions; the provider manages everything else
A typical web product might use IaaS for a legacy component that needs a custom kernel module, PaaS for the main API, and SaaS for observability or email. Mixing models is normal; the goal is to avoid operating what does not differentiate the product.
Follow-ups you should expect
Interviewers use these to check whether the first answer was understanding or memorisation.
- 1Which model would you pick for a legacy application with custom kernel modules?
- 2What operational work disappears when moving from IaaS to PaaS?
- 3How does vendor lock-in differ across the three models?
- 4Can a single system use all three models at once?
Say it in the interview
IaaS gives me virtual machines and networks to manage, PaaS gives me a managed runtime where I deploy code, and SaaS gives me a finished application. As I move up, the provider takes on more patching and scaling, but I give up control. In practice I mix them: IaaS for legacy or custom components, PaaS for the main service, and SaaS for commodity functions.
Use this as a model, not a script. Saying it in your own words is what makes it survive follow-up pressure.
Related questions
Reading is step one
Now say it out loud.
Understanding an answer and delivering it under pressure are different skills. Rehearse this topic in a live AWS, Azure & GCP mock interview and get scored feedback on depth, clarity and confidence.