AWS vs Azure vs GCP
Compare core cloud services across AWS, Azure and Google Cloud.
What this comparison covers
Compare core cloud services across AWS, Azure and Google Cloud.
Set the priorities that matter to you and the tool scores each option against them, then explains the trade-off behind the recommendation.
Choose a service category and this page returns the AWS, Azure and Google Cloud product names that correspond to it: compute, serverless, containers, storage or database.
The answer is a fixed table that ships with the tool, printed as a JSON object with one key per provider plus a short description of the category. There is no pricing, no region list and no feature matrix.
It is a naming cross-reference for someone reading a document written for another cloud, not a comparison of what the three products actually do.
The five categories and their names
Compute maps EC2 to Virtual Machines to Compute Engine. Serverless maps Lambda to Functions to Cloud Functions. Containers maps ECS and EKS to Container Instances and AKS to Cloud Run and GKE.
Storage maps S3 to Blob Storage to Cloud Storage. Database maps RDS and DynamoDB to SQL Database and Cosmos DB to Cloud SQL and Firestore, which groups relational and non-relational products under one heading.
Choosing `all` returns the whole table rather than a category, which is the form to read when you want the five rows side by side.
- compute, serverless, containers, storage, database
- `all` returns every category
- Each row carries a short description of the category
What a name mapping leaves out
Two services with the same role can differ in consistency model, cold-start behaviour, maximum object size and billing unit. This table names the role and stops there.
The table is edited in the tool's own source, so a product renamed by a provider stays under its old name until that file is changed. Treat a name here as a starting point for the provider's own documentation.
How to use the category field
Enter a service category, compute, serverless, containers, storage or database, in the form field and press the Execute Tool Engine button. The labels on the form come from the tool's own parameter schema.
Read the JSON answer with one key per provider plus the category description. It is a naming cross-reference for documents written for another cloud: no pricing, no regions, no feature matrix.
What to Expect
Best for
- Translating an architecture document written for one cloud into another's vocabulary
- Teaching the shape of a three-provider service map
- Getting the canonical product name before searching a provider's docs
Not ideal for
- Choosing a provider on price, latency or regional availability
- Anything about a managed service the five categories do not cover
- A current product list, since the table is a fixed set of names
What this tool keeps
- One key per provider in every row, so the three columns always line up
- The category description alongside the three names
- A JSON envelope you can paste into a document
What may need cleanup
- Grouped entries such as `RDS / DynamoDB`, which you may need to split by hand
- Product names that a provider has since renamed
- The absence of pricing, which a real comparison will need from elsewhere
Common errors
- Reading the database row as if RDS and DynamoDB were interchangeable
- Expecting a region or a price column that the tool does not carry
- Sending a category that is not in the list and getting an error instead of a row
Example use cases
- Porting a runbook from AWS to Azure
- Explaining a GCP architecture to a team that knows AWS
- Naming the right product in a design document
Sample input
Choose serverless from the service dropdown and run the tool.
Sample output
A JSON result naming Lambda, Functions and Cloud Functions, under the label Event-driven functions.
Who this is for
- Engineers reading documentation written for another cloud
- Architects writing a multi-cloud design note
- Anyone who needs the product name rather than the feature list
Frequently asked questions
What does this comparison map?
A service category — compute, serverless, containers, storage or database — to the AWS, Azure and Google Cloud product names that correspond to it.
Can it tell me which provider is cheapest or fastest?
No. The table is a fixed name mapping; price, latency and regional availability are not modelled.
Why does a product name look outdated?
The table is a fixed set of names, so a provider's rename is not picked up automatically and the output may need a hand edit.
Build custom tools for your store
Join the SDK program to create useful ecommerce utilities for your store.