Reading how data actually gets accessed before picking storage
Analyzing data access patterns means understanding how frequently and by whom your data is read/written so you can choose the right storage class, location, and partitioning strategy. Google Cloud offers native tools to observe usage before optimizing cost and performance rather than guessing.
Must-know
- Cloud Storage's Autoclass and Object Lifecycle Management can automatically transition objects between Standard, Nearline, Coldline, and Archive based on age or access recency, but Autoclass and lifecycle rules cannot be combined on the same bucket.
- Storage Insights datasets (bucket-level metrics exported to BigQuery) let you analyze object age, size, and storage class distribution at scale to inform lifecycle and class decisions, this is more scalable than manually inspecting bucket listings.
- BigQuery's
INFORMATION_SCHEMAviews (e.g., JOBS,TABLE_STORAGE) and the Cloud Monitoring metrics expose query frequency, bytes scanned, and table access patterns, which help decide between clustering, partitioning, or materialized views. - BigQuery flat-rate vs. on-demand pricing decisions should be driven by observed query volume and concurrency patterns, not assumptions, heavy, predictable workloads favor reservations/capacity-based pricing.
- Access frequency directly affects retrieval cost and minimum storage duration in Cloud Storage: Nearline/Coldline/Archive classes impose early-deletion fees if objects are deleted or moved before their minimum storage duration (30/90/365 days respectively).
- Cloud Audit Logs (Data Access logs) provide the authoritative source for who accessed which object/table and when, which is essential for both security analysis and validating actual (not assumed) access patterns before re-architecting storage.
Coverage checked against the published exam guide on Aug 11, 2026.
These are independent practice questions, written against this certification's published exam guide. They are not the certification vendor's own questions, and not the real exam.