What's changed: Created Associate Cloud Engineer Chapter 2 (Domain 2 "Planning and configuring": choosing compute = Compute Engine/GKE/Cloud Run/Cloud Functions/Spot VM/custom machine types; choosing storage and databases = Cloud SQL/Spanner/Firestore/Bigtable/BigQuery/Cloud Storage classes/Persistent Disk; planning networking = regions-zones/load balancers/Network Service Tiers).
2.3Planning networking
Understand the basics of planning network resources: designing availability with regions and zones, choosing a load balancer by use case, and selecting Network Service Tiers (Premium / Standard) for performance and cost.
Network planning decides "where to place resources for availability," "how to distribute traffic," and "performance vs cost."
2.3.1Regions and zones
Google Cloud is divided into regions (geographic areas), each with multiple zones (independent failure domains). For higher availability, spread resources across zones (tolerate a zone failure) and, if critical, across multiple regions. Choosing a region near users gives low latency. Remember "single zone = down on a zone failure" vs "multi-zone/region = high availability."
2.3.2Load balancers and Network Service Tiers
A load balancer distributes traffic across instances, improving availability and scalability. Google Cloud load balancers are chosen by target (HTTP(S) vs TCP/UDP), global vs regional, and external vs internal (e.g., serving web to worldwide users = global external HTTP(S) load balancer). Network Service Tiers let you choose Premium (high performance, global, Google backbone) or Standard (cost-focused, regional).
Common: requirement → design. E.g., "tolerate a zone failure" = spread across zones; "tolerate a regional disaster / global low latency" = multiple regions; "distribute traffic for availability/scale" = load balancer; "serve web to worldwide users" = global external HTTP(S) LB; "favor performance" = Premium Tier; "favor cost" = Standard Tier.
Watch the mix-ups: (1) distinguish zone (failure domain in a region) from region (geographic area). (2) Availability comes from redundant placement—single zone is weak. (3) Trade-off of Premium (performance, global) vs Standard (cost, regional).
2.3.3Section summary
- Regions (geography) and zones (failure domains); raise availability by spreading across zones/regions
- Load balancers distribute traffic (choose by target/scope/internal-external)
- Network Service Tiers = Premium (performance/global) vs Standard (cost/regional)
Sign in to track progress — Log in.
Quick check
(just a quick review)Q1. To keep service running through a single-zone failure, which design is most appropriate?
Q2. Which distributes traffic across instances to improve availability and scalability?
Q3. Which Network Service Tier favors global high performance over Google’s premium backbone?
Q4. Which correctly describes the relationship between regions and zones?
Q5. Which Network Service Tier do you choose to favor cost when regional is enough?
Q6. To serve a web app to worldwide users via one global endpoint, which load balancer fits best?