About
I work on storage and distributed infrastructure. Most of what I know came from breaking things in staging and then being surprised by a different failure mode in production.
This site is a set of working notes. I publish them mostly so I stop re-deriving the same conclusions every eighteen months, and because writing something down is the cheapest way to find out that I did not actually understand it.
What you will find here
- Write-ups of bugs that took more than a day to understand.
- Benchmarks, with the caveat that they are not yours.
- Reading notes on papers I actually finished.
Setup
A single small VM in Tokyo, a static site, and a TLS certificate that renews itself. Nothing clever. The latency to most of the services I use from here is around forty milliseconds, which is good enough that I stopped thinking about it.
Contact
I read mail but answer slowly. If something here is wrong, I would rather hear it than not.
This page is intentionally static. There is no analytics, no font CDN, and no third-party script on this site.