Btrfs/ZFS/bcachefs under workloads classic benchmarks skip
Thread
Unofficial Hacker News client; not affiliated with Y Combinator.
Btrfs/ZFS/bcachefs under workloads classic benchmarks skip
Unofficial Hacker News client; not affiliated with Y Combinator.
Farmadupe · · focus · HN ↗
I think if you're not using baremetal for such tests, it's likely that the results are simply not comparable at all? What if another tenant is also using the disk?
walrus01 · · focus · HN ↗
If it's something as simple as a KVM hypervisor that only runs 1 test VM at a time (with no other load from anything else other than the basic systemd daemons, ssh daemon etc running on the hypervisor), the results could be very close to bare metal.
I can see it being very time consuming and annoying to do repeated manual bare metal OS installs and new partitioning/filesystem creation for such a large variety of tests.
The author does also say that performance isn't really the main thing but rather, data integrity:
<a href="https://github.com/fenio/modern-fs-benchmark" rel="nofollow">https://github.com/fenio/modern-fs-benchmark
fenio · · focus · HN ↗
Real hardware is used in: <a href="https://bartosz.fenski.pl/modern-fs-benchmark/real-hw/" rel="nofollow">https://bartosz.fenski.pl/modern-fs-benchmark/real-hw/ <a href="https://bartosz.fenski.pl/modern-fs-benchmark/sas-hdd/" rel="nofollow">https://bartosz.fenski.pl/modern-fs-benchmark/sas-hdd/
But unfortunatelly it's much more limited number of actual runs. sas-hdd is still in progress so numbers for it should increase over time.