Storage estimates

Storage estimates combine object size, daily write volume, retention period, replication factor, and index overhead. Example: 1M new photos/day × 2 MB × 365 days × 3 replicas ≈ 2 PB/year before indexes and thumbnails.

Separate hot data (recent, frequently accessed) from cold archives. Media, logs, and analytics often dominate cost more than relational metadata. Plan compaction, tiering, and legal hold retention separately.

On interviews: walk through one line-item calculation, mention growth rate, and explain when you would move blobs to object storage versus a database.

Common pitfalls: ignoring thumbnail and derivative multiples; no retention policy so storage grows forever; underestimating index size for search features.

Checklist:

• Estimate record size and daily new objects.
• Multiply by retention and replication.
• Split hot versus cold storage tiers.
• Include index and derivative overhead.