The fifth problem was released yesterday! It's about databases, paging, and the importance of good primary keys. sirupsen.com/napkin/problem…

6th problem is live! An exercise in quickly seeing if the simplest possible solution might be sufficient. buff.ly/2Tzezhe

The 7th Napkin Problem is live! This one's about ballpark figures for implementing revision history. sirupsen.com/napkin/problem…

Another napkin problem published! Included is my answer to the last problem on storing revision history in a relational database. This week's problem is about syncing! sirupsen.com/napkin/problem…

9th problem is live. Also includes a detailed answer to problem 8: how to do Merkle tree based syncing! sirupsen.com/napkin/problem…

Another edition out, this time looking at whether MySQL transactions per second = fsyncs per second. sirupsen.com/napkin/problem…

This month's napkin math is about circuit breakers. If you've put one in production, I can almost guarantee you that you've configured it wrong. Read this edition to learn how to configure your circuit breaker correctly. 🧯 sirupsen.com/napkin/problem…

This edition is about filtering with inverted indexes. E.g. answering queries fast across millions of products of the type "give me all products that are blue, tagged summer, and are sneakers." sirupsen.com/napkin/problem…

Issue 14 is out! If you've ever tried to sync databases at scale, you know dropping records on the floor is inevitable. Here we tackle how to verify whether 100M+ records is synced with checksums in less than a minute. sirupsen.com/napkin/problem…

Another edition just went out: Increase Performance by Fitting In the Initial TCP Slow Start Window sirupsen.com/napkin/problem…

Issue 16 on when to write simulators has been published! sirupsen.com/napkin/problem…

What metrics should be on your web application's dashboard? Another computer napkins out! sirupsen.com/metrics

Scaling Causal's Spreadsheet Engine from Thousands to Billions of Cells: From Maps to Arrays sirupsen.com/causal

New Napkin Math out! Composite Indexes are about ~10x faster than index merges in Postgres and MySQL The gap between MySQL and Postgres' can be significant because Postgres doesn't support index-only scans when doing an index merge sirupsen.com/index-merges