Google Cloud Essentials — selected guide
How to store data on Google Cloud
Eight concise orientations to compute, storage, containers, projects, delivery, logs, and errors
Eight 2020–2022 explainers introduce compute, storage, containers, Firebase projects, build tools, logging, and error reporting through the workloads each service supports.
Google Cloud Tech 3 of 8
In this collection Browse 8 summaries 3 of 8
Files, application records, and large analytical datasets have different storage needs. Ryan compares Google Cloud services by how they organize data and how an application reads or analyzes it. 00:00–06:18(opens in a new tab)
Key Points Covered
- Files become objects in buckets. Cloud Storage can hold images, videos, or documents. In the example, a bucket groups profile photos, and each photo is an individual object. 00:00–01:02(opens in a new tab)
- Relational records use tables with defined columns. Cloud SQL supports MySQL, PostgreSQL, and SQL Server, while Google handles operations such as patching, backups, and replicas. The application keeps a familiar database model without managing all of its infrastructure. 01:02–03:07(opens in a new tab)
- Documents allow records with different fields. Firestore groups documents into collections: a
userscollection might contain one document per person, with different fields for different people. Its real-time updates synchronize changes across clients, supporting examples such as leaderboards and collaborative whiteboards. 02:04–04:11(opens in a new tab) - Analysis and high-volume application access are distinct needs. BigQuery is a data warehouse for large-scale SQL analysis; Ryan queries a public dataset for popular baby names. The overview also introduces Spanner for strongly consistent relational data at scale and Bigtable for workloads needing many low-latency reads and writes. 04:11–06:18(opens in a new tab)
Full video: https://www.youtube.com/watch?v=w2mL4sUOjVo(opens in a new tab)