Note on The Slotted Counter Pattern via planetscale.com

The idea here is that instead of incrementing a single row for a counter, we are now picking a slot and incrementing the count in that slot. This means instead of hammering a single row, we are spreading the updates across 100 rows and reducing the potential for contention.

FROM:

planetscale.com

The Slotted Counter Pattern

Source

The slotted counter trades write contention for read complexity, which begs the question Marc Brooker raised in Optimism vs Pessimism in Distributed Systems: what optimistic assumption about your workload are you embedding when you choose this pattern (that reads are rarer or cheaper than writes) and at what read-to-write ratio does that assumption flip and make the pattern self-defeating?