[DRBD-user] Performance Issues on version 8.3.13

GG-Net Mailing mailing at gg-net.de
Thu Apr 25 18:07:52 CEST 2013

Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.


Hey there,

the speed is significatnly higher due to the Samsung 840 SSD's performance plus the performance of the raid controller.

And according to you, if I use protocol A or C is pretty much irrelevant to the performance of the DRBD, since it's always only as fast as the slowest node?

Thanks in advance,

Tim

-----Ursprüngliche Nachricht-----
Von: Sebastian Riemer [mailto:sebastian.riemer at profitbricks.com] 
Gesendet: Donnerstag, 25. April 2013 17:58
An: GG-Net Mailing
Cc: drbd-user at lists.linbit.com
Betreff: Re: AW: [DRBD-user] Performance Issues on version 8.3.13

Hi Tim,

from my experience there is no noticeable benefit of using Protocol A instead of C.

So you're doing random write IO with fio?

The 500kb/s are what a single HDD is only capable of. The higher number on the disconnected primary is caused by caching.

Cheers,
Sebastian


On 25.04.2013 17:18, GG-Net Mailing wrote:
> Well, that's why i use Protocol A so i have asynchronous sync. 
> 
> I tested using FIO, too. All my results there around 40mb/s with drbd disconnected.
> 
> Once I started DRBD I had round about 500kb/s in all of my tests.
> 
> 
> So the problem isn't the throughput of the network.
> 
> That's why I'm asking, if there is any known issue or workaround regarding this.






More information about the drbd-user mailing list