Database Capacity Planning: Storage Growth, Index Size, IOPS & Query Cost
Size a production database the way DBAs do it: bytes per row and table size, B-tree index pages with fill factor, compound storage growth to the disk alert, IOPS and throughput from the query mix, cache hit ratio and buffer pool, replicas and backup retention, connection pools with Little's law and provisioned vs serverless cost, in 8 hands-on labs with illustrative numbers.
The first lesson is free — no account needed.
About this course
A database capacity planning course that sizes the production database of one fictional parcel-tracking SaaS, Parcelwise, in a spreadsheet, for DBAs, backend and platform engineers, SREs and computer science students. You estimate bytes per row from the column types and the size of a 48-million-row table on 8 KB pages, size four B-tree indexes with a 90% fill factor, forecast compound storage growth until the 80% disk alert fires, turn the query mix into read and write IOPS with write amplification, size the buffer pool from the cache hit ratio and the working set, add up replicas, backups and WAL for a 30-day recovery window, size connection pools with Little's law, and compare provisioned instances with serverless cost in a chart. Illustrative numbers, with no affiliation with any database or cloud vendor. The first lab is free.
Lessons
- Row Size from Column Types & Table Size: Bytes per Row, Rows per Page, GiBFree
- Index Size & Overhead: B-tree Leaf Pages, Fill Factor & Indexes vs Table Pro
- Database Storage Forecast: Compound Monthly Growth, Overhead & Months Until the Disk Alert (Chart) Pro
- IOPS & Throughput from the Query Mix: Reads, Amplified Writes & the Peak Pro
- Cache Hit Ratio & Buffer Pool Sizing: Working Set, Headroom & the RAM to Buy Pro
- Replicas, Backups & Retention: The Real Storage Footprint of One Database Pro
- Connection Pool Sizing with Little's Law: Busy Connections, Pool per Instance & Limits Pro
- Cost Comparison: Instance Sizes vs Serverless per Month (with a Chart) Pro