Cloud services are capabilities hosted by a cloud service provider such as Google Cloud, AWS, or Azure, and accessed by consumers over the internet. The opposite arrangement is on-premises, where the IT resources sit inside the facilities of the organization using them. Before working through any specific database product, it helps to be clear on how these services are categorized and what a provider takes off your plate. The Professional Cloud Database Engineer exam assumes this framing as background, and questions are easier to read once the service model behind a scenario is obvious.
Cloud offerings are commonly grouped into three models that differ in how much of the stack the provider manages for you: Infrastructure as a Service, Platform as a Service, and Software as a Service. The distinction comes down to the level of abstraction and the responsibilities each one removes.
Infrastructure as a Service, or IaaS, gives you virtualized computing resources online. You rent the infrastructure you need, such as servers and storage, without maintaining physical hardware. Compute Engine is the Google Cloud example, alongside Amazon EC2 and Azure VMs. This model suits organizations that want full control over their infrastructure and need to set up custom environments to their own specifications.
Platform as a Service, or PaaS, provides a platform for building, running, and managing applications without handling the underlying infrastructure. The provider takes care of the operating system, storage, and runtime, so developers can focus on code rather than servers. App Engine is the Google Cloud example here, with AWS Elastic Beanstalk and Azure App Services as counterparts.
Software as a Service, or SaaS, delivers ready-to-use applications over the internet. Users access the software without installing or maintaining it. Google Workspace, Salesforce, and Microsoft Office 365 fall into this category. The emphasis is simplicity for the end user, which makes SaaS a natural fit for everyday business needs.
Read left to right from IaaS to PaaS to SaaS, the level of abstraction and simplicity increases at each step. IaaS provides the infrastructure, PaaS provides the platform, and SaaS provides complete applications. The further along you go, the less management complexity is left to the consumer.
Google Cloud offers a large catalog of services spanning compute, storage, databases, data analytics, networking, development, security, and AI and machine learning. Most of them fall under either IaaS or PaaS. A short and incomplete list gives a sense of the range: App Engine, Compute Engine, Kubernetes Engine, Cloud Run, Cloud Storage, BigQuery, Spanner, Bigtable, Cloud SQL, Pub/Sub, Dataflow, Data Fusion, Agent Platform, Cloud IAM, Cloud Logging, and Cloud Build, among many others.
The point of grouping them this way is to see what the cloud removes. In a traditional on-premises setting, an organization that wanted these capabilities would have to stand up its own clusters and software and keep them running. Using cloud services shifts that work to the provider, and how much it shifts depends on the model. The exam does not test every one of these services, and only a subset appears on it. As a database engineer you will spend most of your attention on the database and data services in that list, but the broader catalog is the backdrop those products sit in.
One practical wrinkle is that Google Cloud service names change over time, and the official exam does not always reflect those updates the moment they happen. A question can still reference an older name well after the branding has moved on.
Dataplex is a useful example. The service was released in 2022 to help manage and govern data across various storage systems. By 2025 it had become the Dataplex Universal Catalog, and into 2026 the branding shifted again to the Knowledge Catalog. The name changed at each step, but the underlying purpose, data discovery and governance, stayed the same. A professional-level exam taken in 2026 could still present the 2022 name rather than the current one, and we have seen this happen in practice.
The takeaway is to recognize a service regardless of which naming convention a question uses. We aim to stay aligned with whichever name is most likely to appear on the exam, whether that is a legacy version or the most recent update, and we mention alternative names for the same service so you recognize it on sight. When you encounter a name in a Professional Cloud Database Engineer question, the goal is to match it to the right service immediately and not lose time wondering whether it is something new.
Our Professional Cloud Database Engineer course covers cloud service models and Google Cloud's database services alongside service terminology and how to read exam questions, with practice questions that drill these distinctions.
