14 Oct
2022
14 Oct
'22
9:21 p.m.
Hi,
<snip all the very relevant and good points>
Anyway, my main concern wouldn't be the published
error rate but
just the fact that at 100MB/sec sustained write, it would take 2½
days to heal the array during which time there is no redundancy. And
100MB/s is really optimistic for a HDD in a system under other load.
I was pleasantly surprised to learn that the Toshibas I mentioned up thread
were specced at nearly 300MB/s. In my USB 3.0 dock I was getting a
sustained 160 or 170MB/s on to them, sequential write.
So, somehow, these specs are still improving! My previous disks could only
do about 140MB/s. With the new disks I'm clearly limited by my USB3.0 HDD
dock.
OTOH, on an active server, you're not going to get sequential reads and
writes for long before another process comes in and that'll destroy your
performance. You may still not be IOPs limited, but your rebuild won't get
anywhere near the single-threaded read performance.
So yeah, I'd agree that 100MB/s is very optimistic, but I've also been
shocked just how much HDD performance is still improving.
Best wishes,
@ndy
--
andyjpb(a)ashurst.eu.org
http://www.ashurst.eu.org/
0x7EBA75FF