Soooooooo, is it possible that a software RAID 1 can cut SSD performance in half?

Follow

@matrix Thats not as uncommon as you think. Thats the whole reason multiple runs are done on benchmarks... Keep in mind there is disk level caches, system level caches, as well as other applications you may have running as well (or even system level processes).

@freemo I know, but when I benchmarked the same drives alone, the difference between runs was like +-5MB. I didn't expect RAID to hit so much.

@matrix I suspect is the software raid caching data in memory, something linux loves to do when there is a lot of space memory to abuse.

@freemo Well it's a fresh server with 32GBs of RAM, so it can abuse that :lul:

@freemo @matrix
Also don't run a test for a couple of minutes, do a couple of hours to be sure you have exhausted enough of these caches and buffers to know with what you hare dealing with. 🙂

Sign in to participate in the conversation
Qoto Mastodon

QOTO: Question Others to Teach Ourselves
An inclusive, Academic Freedom, instance
All cultures welcome.
Hate speech and harassment strictly forbidden.