Cloud Providers

What Is Google Cloud Platform Used For?

Google Cloud Platform runs applications, stores and analyzes data, builds AI models, and manages networks. Learn its main uses and how it differs from Workspace.

On this page
  1. What Google Cloud Platform is
  2. Running applications and websites
  3. Storing and managing data
  4. Analytics with BigQuery
  5. AI and machine learning
  6. Networking and security
  7. Mobile and web backends
  8. Google Cloud vs Google Workspace
  9. Who GCP tends to suit
  10. Free tier, credits, and discounts
  11. Common misconceptions
  12. Certifications and next steps

Google Cloud Platform (GCP) is Google’s cloud platform. Organizations use it to run applications and websites, store and manage data, analyze large datasets, build and deploy machine learning models, and connect and protect their networks. It is strongest in data analytics, AI, and Kubernetes, which are areas where Google’s own products have deep roots. GCP is separate from Google Workspace, the productivity suite that includes Gmail and Google Docs.

What Google Cloud Platform is

Google began offering its infrastructure to outside customers with App Engine in 2008. Compute Engine became generally available in December 2013. The platform is now branded Google Cloud, and GCP is the common short name for its cloud platform products.

GCP is the third-largest cloud provider by infrastructure share, at about 12–13% in Synergy Research’s 2025 estimates. It has about 42 regions and 127 zones, according to 2025 figures. Google reported 2024 revenue of $43.2 billion for Google Cloud, a figure that includes Workspace.

Running applications and websites

GCP offers several ways to run applications, depending on how much you want to manage:

  • Compute Engine provides virtual machines with per-second billing, live migration during maintenance, and sustained use discounts on eligible machine types.
  • Google Kubernetes Engine (GKE) runs containers managed by Kubernetes, which Google created and later donated to the Cloud Native Computing Foundation. The GKE, EKS, and AKS comparison covers how GKE compares with the other managed options.
  • Cloud Run runs any container that serves requests, scales to zero when idle, and is based on the open-source Knative project.
  • Cloud Run functions runs single functions in response to events or HTTP requests. Google renamed Cloud Functions to Cloud Run functions in August 2024. For a comparison with other providers’ function services, see the serverless comparison.
  • App Engine runs web applications without you managing the servers underneath.

Storing and managing data

Cloud Storage holds files, backups, and other unstructured data, with storage classes that trade access speed for cost. For relational databases, Cloud SQL runs managed MySQL and PostgreSQL. AlloyDB is a PostgreSQL-compatible database for demanding workloads. Spanner is a globally distributed relational database. Bigtable and Firestore cover wide-column and document data.

Analytics with BigQuery

BigQuery is Google Cloud’s serverless data warehouse, generally available since 2011. It separates storage from compute, so you can query large datasets without managing servers for the query engine. Dataflow handles stream and batch data processing, and Looker, which Google acquired in 2020 for $2.6 billion, provides business intelligence dashboards.

BigQuery is often the first GCP service data teams adopt, because it can answer questions about large datasets with standard SQL.

AI and machine learning

Vertex AI, launched in May 2021, is the platform for building, training, and deploying machine learning models. Google’s Gemini models and a catalog of other models, called Model Garden, are available through it. BigQuery ML lets teams train and run some models with SQL.

Google also designs its own accelerators, called Tensor Processing Units (TPUs). The Trillium generation became generally available in December 2024, and Google announced the Ironwood generation in April 2025. TPUs are available only on Google Cloud. Availability depends on region and quota.

Networking and security

Google Cloud’s network tiers offer a choice. Premium Tier carries traffic over Google’s private backbone for more of the path. Standard Tier uses the public internet and is cheaper, with different performance characteristics. Cloud CDN, load balancing, and Cloud DNS are available for traffic management.

For security, Google acquired Mandiant in 2022 for $5.4 billion, and it announced an agreement to acquire Wiz, a cloud security company, in March 2025 for $32 billion. Check the current status of that deal before relying on it. Google Cloud also offers identity and access management, encryption, and threat detection services.

Mobile and web backends

Firebase, which Google acquired in October 2014, adds services for mobile and web apps, including databases, authentication, and hosting. Many teams use Firebase alongside Cloud Run or other GCP services.

Google Cloud vs Google Workspace

Google Workspace is a software subscription for email, documents, calendars, and video meetings. Google Cloud Platform is a set of infrastructure and platform services for building and running systems. Both sit within Google’s cloud business and both are reported together, but they are different products. A company that uses Gmail does not necessarily run workloads on GCP, and a company that runs workloads on GCP does not have to use Workspace.

Who GCP tends to suit

GCP is often a strong choice when:

  • Your product depends on data analytics, machine learning, or AI models
  • Your team runs containerized microservices on Kubernetes
  • You want per-second billing and discounts that apply without long upfront commitments
  • You want a platform built around Google’s data and AI tooling

GCP may be less natural for organizations that rely on a large set of specialized managed services or on Microsoft-based identity and licensing. Compare the services you need on each platform before deciding. The AWS vs Azure vs Google Cloud comparison maps the major services side by side.

Free tier, credits, and discounts

New accounts receive a $300 credit valid for 90 days, along with Always Free resources. An example is one e2-micro virtual machine in select US regions. Sustained use discounts apply automatically to many machine types but not all, and the E2 family is one example that does not qualify. Committed use discounts require a commitment. The cloud pricing comparison explains these models alongside AWS and Azure.

Common misconceptions

  • Google Workspace is not GCP. Both are part of Google’s cloud business, but they are separate products.
  • GCP is not only for AI and data. It runs general virtual machines, web apps, databases, and networks.
  • Calling Google Maps APIs is not running infrastructure on GCP. Using a Maps API is a service call, not a deployment on the platform.
  • Sustained use discounts do not apply to every machine type. Check the eligibility for the family you plan to use.

Certifications and next steps

Google Cloud’s certifications include Cloud Digital Leader for business and non-technical roles, Associate Cloud Engineer for hands-on administration, and Professional Cloud Architect for design work. Start with the Digital Leader path if you are new to cloud concepts, and move to the associate certification once you have run workloads.

To try GCP, create an account to use the free credit and Always Free resources. Keep an eye on usage and set a budget alert, because free allowances are capped and usage above them is billed at normal rates.

The complete guide to cloud providers

AWS vs Azure vs Google Cloud: A Full Comparison
Read the guide

Keep exploring

Guides from other topics

All topics →