Re: bioctl raid5 disk performance questions

2025-03-17 Thread Stuart Henderson
On 2025-03-16, Jon Higgs wrote: > On 16/03/25 11:51, Rub??n Llorente wrote: >> As far as I understand, raid 5 performance is usually slower than you >> would expect because parity data needs to be computed and written to >> the drives too. softraid's implementation of RAID5 is not particularly qu

Re: bioctl raid5 disk performance questions

2025-03-16 Thread Jon Higgs
On 16/03/25 11:51, Rub??n Llorente wrote: > As far as I understand, raid 5 performance is usually slower than you > would expect because parity data needs to be computed and written to > the drives too. I just tested this on my machine with a pair of SSDs in RAID1. It was still under 50MB/s.

Re: bioctl raid5 disk performance questions

2025-03-16 Thread Rubén Llorente
Divan Santana wrote: Greetings :) I have a openbsd 76 system (dmesg [1]) with bioctl raid5 setup. It achieves performance writes of between 18-43 MB/s with a dd test[2] and 7-11 MB/s with rsync tests[3]. But the backend drives are reported to write at 220MB/s according to it's vendor spec [4].

bioctl raid5 disk performance questions

2025-03-15 Thread Divan Santana
Greetings :) I have a openbsd 76 system (dmesg [1]) with bioctl raid5 setup. It achieves performance writes of between 18-43 MB/s with a dd test[2] and 7-11 MB/s with rsync tests[3]. But the backend drives are reported to write at 220MB/s according to it's vendor spec [4]. I would expect with a