Cloud Engineer · Step 2 of 8 · 9 min
Core services: compute, storage, networking, databases
Recognise the four building blocks every provider offers and the names each one uses.
Compute
Compute is where code runs. Virtual machines are the classic form: Amazon EC2, Azure Virtual Machines, Google Compute Engine. You choose the size, the operating system, and you patch it. Containers and managed container services sit one level up, and serverless, covered later, sits above that.
Storage
Object storage holds files of any size, cheaply and durably: Amazon S3, Azure Blob Storage, Google Cloud Storage. Block storage acts like a disk attached to a machine. File storage acts like a shared network drive.
Storage is cheap per gigabyte, but costs add up through copies, old snapshots nobody deleted, and data transferred out to the internet.
Networking
Each provider gives you a private network: a VPC on AWS and Google Cloud, a VNet on Azure. Inside it you create subnets, some public and some private, and rules for which traffic is allowed in and out. Load balancers spread traffic across machines, and DNS turns names into addresses.
A simple safe pattern: keep databases in private subnets, and expose only what must face the internet.
Databases
Managed databases run the engine for you, with backups and updates: Amazon RDS, Azure SQL Database and Azure Database for PostgreSQL, Google Cloud SQL. There are also NoSQL options such as Amazon DynamoDB, Azure Cosmos DB and Google Firestore, and analytics warehouses such as Amazon Redshift and Google BigQuery.
The names differ; the ideas match. Learn the concept once, then map it to each provider.
Try it
Price a small app before you build it
- 1.Open the free pricing calculator of any major provider: AWS Pricing Calculator, Azure Pricing Calculator or Google Cloud Pricing Calculator. No account is needed.
- 2.Add one small virtual machine, 20 GB of object storage and a small managed database.
- 3.Note the monthly total, then change the region and see if it moves.
- 4.Ask a chatbot the prompt below to compare your estimate across providers, and treat its prices as rough until you check them.
I want to run a small web app with one virtual machine, a managed PostgreSQL database and object storage for images. Map each part to the service names on AWS, Azure and Google Cloud, and list the cost items that people most often forget.