Date: Tue, 29 Jun 2021 08:06:11 +0200 From: Gerrit Kuehn <gerrit.kuehn@aei.mpg.de> To: freebsd-net@freebsd.org Subject: Re: RFC: NFS trunking (multiple TCP connections for a mount Message-ID: <20210629080611.513b62d4@comet2.terra.ger> In-Reply-To: <YQXPR0101MB0968DC173855A82AAF45F08FDD039@YQXPR0101MB0968.CANPRD01.PROD.OUTLOOK.COM> References: <YQXPR0101MB0968DC173855A82AAF45F08FDD039@YQXPR0101MB0968.CANPRD01.PROD.OUTLOOK.COM>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, 29 Jun 2021 00:23:21 +0000 Rick Macklem <rmacklem@uoguelph.ca> wrote: > I don't understand how multiple TCP connections to the same > server IP address will distribute the load across multiple network > interfaces? > I thought that lagg would have handled this? I don't think this targets multiple interface situations. In my experience it is next to impossible to max. out a 10G connection with a single NFS connection running over it. The NICs have several independent queues like so (systat -ifstat): ixl0:aq ixl0:rxq0 ixl0:rxq1 ixl0:rxq2 ixl0:rxq3 ixl0:rxq4 ixl0:rxq5 ixl0:rxq6 ixl0:rxq7 A single connection will only use one of these, limiting the overall throughput. Spreading load across multiple connections sounds quite useful to me in this situation. cu Gerrit
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20210629080611.513b62d4>