Storage, Data Platform, and Data Movement Options
Azure Storage services
Azure Storage accounts provide Blob, File, Queue, Table, and Disk services. Blob storage is for unstructured objects; choose access tiers (hot, cool, cold, archive) based on access frequency. Lifecycle management rules move or delete blobs automatically. Redundancy options include LRS, ZRS, GRS, GZRS, and read-access variants. GRS replicates to a secondary region but does not allow reads there unless RA-GRS or RA-GZRS is used. Archive tier has low storage cost but high rehydration latency.
Data platform choices
Azure SQL Database is a managed relational engine; SQL Managed Instance offers near-full SQL Server compatibility. Cosmos DB is globally distributed NoSQL with tunable consistency levels. Azure Database for MySQL and PostgreSQL are managed OSS databases. Azure Data Lake Storage Gen2 adds hierarchical namespace to Blob storage for analytics.
Movement and integration
Azure Data Factory orchestrates pipelines and data movement. Azure Synapse combines analytics, SQL pools, and Spark. Azure Data Explorer handles log and telemetry analytics. Azure Data Box moves large datasets offline. Storage Explorer and AzCopy are common tools.
Traps
Questions often confuse GRS with read access, assume archive data is instantly available, or mix up Table storage with Azure SQL. Cosmos DB consistency levels range from strong to eventual; stronger consistency costs latency. Also distinguish storage account replication from application-level backup.
Worked example
A compliance archive is accessed once or twice a year. Hot blob storage is wasteful. Use archive tier with lifecycle rules, accepting hours of rehydration time. If immediate reads are required after a regional outage, choose RA-GZRS rather than GRS.
Rule of thumb
Choose redundancy by RPO/RTO, tier by access frequency, and database by data model and consistency needs.