What's changed: Created Associate Cloud Engineer Chapter 3 (Domain 3 "Deploying and implementing": Compute Engine and GKE = instances/MIG-autoscaling/instance templates/OS Login/GKE (Autopilot/Standard/regional/private)/kubectl; serverless and data = Cloud Run/Cloud Functions/Eventarc/data products/AlloyDB/loading; networking and IaC = VPC/subnets/firewall/Cloud VPN/VPC Peering/Terraform/Config Connector).
3.2Deploying serverless and data
Understand deploying apps to Cloud Run and Cloud Functions, receiving events via Pub/Sub, Cloud Storage changes, and Eventarc, deciding where to deploy, and deploying data products (Cloud SQL, BigQuery, Firestore, Spanner, AlloyDB, Pub/Sub, Dataflow) and loading data.
Serverless lets you deploy code or containers without managing servers. We also cover deploying data products and loading data.
3.2.1Deploying Cloud Run and Cloud Functions
Cloud Run deploys containers serverless, triggered by HTTP requests or events. Cloud Functions runs small functions on events like Pub/Sub messages or Cloud Storage object changes. To handle diverse event sources uniformly, use Eventarc to route events to Cloud Run/Functions. Map "serverless containers = Cloud Run," "small task on an event = Cloud Functions," "event routing = Eventarc."
3.2.2Deploying data products and loading data
In the data layer, deploy Cloud SQL, Spanner, Firestore, BigQuery, Pub/Sub, Dataflow, and Cloud Storage by use case. For high-performance PostgreSQL compatibility, AlloyDB is an option. Load data by method based on scale and source: command-line upload (gcloud / bq CLI), loading from Cloud Storage, and Storage Transfer Service for large or continuous transfers. Map "bulk/continuous transfer of huge data = Storage Transfer Service" and "scheduled ingestion into BigQuery = BigQuery Data Transfer Service."
Common: requirement → service. E.g., "deploy containers serverless" = Cloud Run; "trigger on Pub/Sub or object changes" = Cloud Functions; "route diverse events" = Eventarc; "high-performance PostgreSQL-compatible" = AlloyDB; "continuous large data transfer" = Storage Transfer Service; "upload via command line" = gcloud/bq CLI.
Watch the mix-ups: (1) Cloud Run (containers) vs Cloud Functions (functions) differ in deploy unit. (2) Eventarc is the event-routing layer. (3) Choose load method by scale and source (CLI for small, Storage Transfer Service for huge).
3.2.3Section summary
- Cloud Run (serverless containers) / Cloud Functions (event-driven functions) / Eventarc (event routing)
- Data products = Cloud SQL/Spanner/Firestore/BigQuery/Pub/Sub/Dataflow/AlloyDB
- Load by scale = CLI / from Cloud Storage / Storage Transfer Service
Sign in to track progress — Log in.
Quick check
(just a quick review)Q1. To deploy a containerized app serverless, triggered by HTTP or events, which is best?
Q2. To run a small task on Pub/Sub messages or Cloud Storage object changes, which is best?
Q3. Which uniformly routes events from diverse sources (Pub/Sub, Cloud Storage) to Cloud Run/Functions?
Q4. To continuously transfer tens of terabytes from on-premises to Cloud Storage, which tool is best?
Q5. Which is an option when you need a high-performance, PostgreSQL-compatible managed database?
Q6. To upload a small amount of data to Cloud Storage or BigQuery from the command line, which is appropriate?