What's changed: Created Associate Data Practitioner Chapter 1 (Domain 1 "Preparation and ingestion": methodologies and quality = ETL/ELT/ETLT/data quality/cleaning (Cloud Data Fusion/BigQuery/Dataflow)/transfer (Storage Transfer Service/Transfer Appliance); formats and extract/load = CSV/JSON/Parquet/Avro/BigQuery Data Transfer Service/Database Migration Service/CLI; choosing storage and location = destinations/structured classification/regional-dual-multi-zonal).
1.3Choosing storage and data location
Understand choosing among storage destinations (Cloud Storage, BigQuery, Cloud SQL, Firestore, Bigtable, Spanner, AlloyDB), classifying data requirements as structured, semi-structured, or unstructured, and selecting data location types: regional, dual-regional, multi-regional, and zonal.
Where to store data depends on its kind (structured/semi-structured/unstructured) and use case. You also choose a location type that affects availability, latency, cost, and compliance.
1.3.1Choosing a destination
| Data/use | Destination |
|---|---|
| Unstructured (images, video, files) | Cloud Storage |
| Large-scale analytics | BigQuery |
| Business relational | Cloud SQL / AlloyDB |
| Global core | Spanner |
| Mobile/web documents | Firestore |
| Huge-scale/time series | Bigtable |
Use Cloud Storage for unstructured data, BigQuery for large-scale analytics, Cloud SQL or high-performance AlloyDB for business relational, Spanner for global core systems, Firestore for mobile/web documents, and Bigtable for huge-scale/time series. Classify the data requirement as structured (tables)/semi-structured (JSON)/unstructured and pick the fitting destination.
1.3.2Location types
Location types include zonal (within one zone), regional (within one region, multi-zone redundant), dual-regional (spanning two regions), and multi-regional (spread across regions). Regional/multi-region increases availability and redundancy, and limiting to specific areas meets data residency requirements. Choose by balancing redundancy, cost, and regulation.
Common: requirement → destination/location. E.g., "images/files" = Cloud Storage; "analytics" = BigQuery; "business RDB" = Cloud SQL/AlloyDB; "global core" = Spanner; "huge time series" = Bigtable; "multi-zone redundancy" = regional; "multi-region high availability" = multi-regional; "limit to in-country" = region selection (data residency).
Watch the mix-ups: (1) choose the destination by data kind (structured/semi-structured/unstructured) and use case. (2) Availability rises zonal (one zone) < regional (one region redundant) < dual/multi-regional. (3) Limiting to specific areas meets data residency.
1.3.3Section summary
- Destination = Cloud Storage (unstructured)/BigQuery (analytics)/Cloud SQL, AlloyDB (business RDB)/Spanner (global)/Firestore, Bigtable (NoSQL)
- Classify the requirement as structured/semi-structured/unstructured to choose
- Location types = zonal/regional/dual/multi-regional (choose by redundancy, cost, regulation)
Sign in to track progress — Log in.
Quick check
(just a quick review)Q1. Which destination best stores unstructured data like images, video, and files?
Q2. To store redundantly across zones within one region and tolerate zone failures, which location type?
Q3. Which destination best suits analyzing large-scale data with SQL?
Q4. To spread data across regions for the highest availability and redundancy, which location type?
Q5. Which data requirement should you classify first when choosing a destination?
Q6. Which can be used as a high-performance, PostgreSQL-compatible managed database?