I've used those and haven't had issues, but also haven't tested them in a write-intensive workload like this.
PSA: avoid Crucial P1 SSDs. Write performance tanks after a few seconds of sustained random writes, and long-tail latencies are absolutely horrendous (>1 second). They drop to <5000 IOPS or worse, and sometimes stall. Seems QLC isn't anywhere near ready for prime time.
-
-
This Tweet is unavailable.
-
- Show replies
-
-
-
Did you use a reproducible test? I have a QLC Samsung 2TB 860 QVO be interesting to compare.
-
Just fio doing 4k writes. Ideally test it both with an empty drive and a full one (random data over entire drive). On the Crucial one though, you don't even need to do that, it happens on a fresh drive too.
End of conversation
New conversation -
-
-
is it thermal throttling?
-
Nope. Checked for that.
End of conversation
New conversation -
-
-
it's tired. you're hurting the poor thing. get it some lemonade
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
I put an Intel 660p in a Mid 2013 MacBook Air and I'm having similar issues. It works great right up until the OS needs a working swapfile and then suddently I get multiple second lockups. QLC is a bad joke.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
why are sata/pcie ssds superseeded by small form factor m.2 ssds in regular pcs? having a larger area available should mean that they can use MLC as opposed to TLC/QLC
-
It's not an area issue, it's a cost issue. TLC/QLC is cheaper per bit.
- Show replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.