Edited by tkaiser at Wed Nov 12, 2014 05:23 |
dupont-y replied at Wed Nov 12, 2014 03:20
Yes, + 100 Mbits. Take it with a grain of salt, this is absolutely not tuned, and not optimal beca ...
Ok, understood. Based on incomplete information regarding the 'environment' of the tests it's a bit too early to expect speed improvements due to a newer kernel (or at least it's not clear what's responsible for the increased throughput). But it still looks promising.
For me it was really funny to realize which parameters on this sort of lowend hardware are reponsible for performance gains compared to the server hardware we normally deal with (x86/Sparc and 10 GbE)
BTW: It would be really great if you could post a short summary regarding your SATA port multiplier setup in a new thread -- here or maybe in the "advanced users"/"server/networking" subforums. At least containing the necessary kernel config, the chipset your port multiplier uses and maybe a product link as well.