here it is with 128k blocksize instead of 4MB
https://gist.githubusercontent.com/Zorlin/55024e6cccc6096c9efd7cc26c2fc95e/raw/8671f6272edc24ce1fb37beb805641dbf3a661e9/fio-read-sequential-128k
https://gist.githubusercontent.com/Zorlin/55024e6cccc6096c9efd7cc26c2fc95e/raw/910f6ebd7617ebafe33d48cea751dbff04750682/fio-read-sequential-4MB
this'll give you some idea, this is not erasure coded however. I expect with erasure coding you'd see slightly higher latency and maybe 880MiB/s.
It works in the literal sense, but you should really try it as bare metal. You will notice a huge difference in latency and performance when doing any kind of metadata operations, and this effect is compounded as you get to millions of objects. My cluster has 1.1 billion objects.
There is *an* impact but honestly I don't notice the difference. I will happily serve datasets out of EC goals in production with nobody caring :)
Master cannot run as a VM. I know that's a funny statement, but... try it, it'll work but you'll have absolutely terrible performance. I didn't...
I wouldn't mind sharing but they don't publicly disclose it and I'd like to be respectful of that, sorry. I'll comment that it's very reasonable pricing to me for the feature set and reliability, even when comparing to other similar SDS like Ceph and SeaweedFS. I try to be fairly agnostic and...
For what it's worth -
I'm running MooseFS Pro at home (150TiB licence, 1.3PB cluster) and at work (1.5PB licence, 1.8PB cluster) and the erasure coding and high availability is solidly in "just ****ing works" territory. Failover is reliable. I sleep better with it than I do with my Ceph...
Can someone help walk me through how the OCVID values work when using ZenStates? I tried --oc-vid 80 which I thought was 1.05v CPU core and it consistently causes a lockup. --oc-vid 50 seems more stable but I haven't figured out how to get CPU core voltage readouts on Proxmox 7.2..
Neither the 32c nor 64c Epyc were vendor locked, thankfully. I could be wrong but I don't think the ES chips can be vendor locked?
Either way, both booted on the H11SSL-I.
I got the chip running on H11SSL-i v2.0 using the custom BIOS found in a certain thread. Thrilled!
Had to stick IPMI on and flash the custom BIOS before it would POST.
Nah, their lab bios was the one on the beta page publicly available. I think the 7502 is simply a later generation chip even as an ES so it works for them. I'll still have a play to learn, but yeah.
Ohhhh. They were using a 7502 ES, not this chip. Drat. Well, I was happy to get a different mobo anyways, if needed. I'll be throwing some more traditional Epycs in here or getting a later ES if I get brave enough.
That's a good point but I have read some posts here claiming to have it on AGESA 1.0.0.3 which you can get, and someone claims to be using a ROMED8-2T... I've already DM'ed those people.
I've had a look, I'm not after spoon feeding and I'm not a stranger to fairly complex mods. I just have come up empty when I've tried to figure out how. If it's not a vendor lock issue and I could use the ROMED8-2T I'd love to just do that rather than buying a specific board.
If it's easy...
Sorry, one is Dell one is HP.
Supposedly the locked ES chips can work on any bios that has been sufficiently edited but there's a fabled two boards -
H11SSL v2.0 + + ASUS KRPA-U16 flashed with bios 0302 (per someone else's thread about the same issue apparently)
which seem to work well and...
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.