IaaS vs PaaS vs SaaS for the PCA Exam: Cloud Service Models Explained

GCP Study Hub
Ben Makansi
April 19, 2026

The three cloud service models

Cloud providers package their offerings into three service models: Infrastructure as a Service, Platform as a Service, and Software as a Service. They differ in how much of the stack the provider runs for you and how much you run yourself. The Professional Cloud Architect exam expects you to identify which model a service belongs to and pick the right one for a given workload.

Infrastructure as a Service (IaaS)

IaaS gives you virtualized computing resources online. You rent servers, storage, and networking from the provider instead of buying physical hardware. Examples are Google Compute Engine, Amazon EC2, and Azure VMs. You get full control over the infrastructure, which lets you set up custom environments to fit whatever the workload needs.

Platform as a Service (PaaS)

PaaS gives you a platform to build, run, and manage applications without dealing with the underlying infrastructure. The provider handles the operating system, storage, and runtime. You write the code and deploy it. Examples are Google App Engine, AWS Elastic Beanstalk, and Azure App Services.

Software as a Service (SaaS)

SaaS gives users access to ready-to-use applications over the internet. There is no installation and no maintenance on the user side. Examples are Google Workspace, Salesforce, and Microsoft Office 365.

Increasing abstraction from IaaS to SaaS

Moving from IaaS to PaaS to SaaS, the level of abstraction increases and the user takes on less management. IaaS provides infrastructure, PaaS provides the platform, and SaaS provides the complete application.

Where GCP services fit

GCP has a large catalog of services across compute, storage, databases, data analytics, networking, development, security, and AI/ML. Most are either IaaS or PaaS. A subset of these will appear on the Professional Cloud Architect exam, including App Engine, Compute Engine, BigQuery, Kubernetes Engine, Cloud Storage, Composer, Cloud Run, Cloud Functions, Dataproc, Vertex AI, Cloud IAM, Data Fusion, Dataflow, Pub/Sub, Spanner, Bigtable, Cloud SQL, and Cloud Build. Each is hosted by the cloud service provider and accessed by consumers over the internet. On-prem means the IT resources sit inside the facilities of the organization using them, which is what these services replace.

My Professional Cloud Architect course covers IaaS, PaaS, and SaaS alongside the rest of the foundational architecture material.

arrow