I always attributed speed issues to writing transaction history, but: Matthew Dillon discovered HAMMER was repeating itself when writing to disk. Fixing that issue doubled write speed. This fix is in 4.6 and the upcoming release.
A running description of activity related to DragonFly BSD.
I hope recent benchmarks can be re-run!
>>”This fix just about doubles HAMMER’s sequential write bandwidth.”
How often _sequential_ writes happen, especially given in the day and age of SSD (where everything is random)?
I love this. My experience is that a big chunk of optimizations is to stop doing something useless.