From: "Aurélien Aptel" <aaptel@suse.com>
To: Xiaoli Feng <xifeng@redhat.com>
Cc: linux-cifs@vger.kernel.org
Subject: Re: How to verify multichannel
Date: Tue, 19 May 2020 18:54:13 +0200 [thread overview]
Message-ID: <878shnn9vu.fsf@suse.com> (raw)
In-Reply-To: <2024477496.29432550.1589904923298.JavaMail.zimbra@redhat.com>
Xiaoli Feng <xifeng@redhat.com> writes:
> Now I can see the multiple channel info in network package when mount with
> option "max_channel=2". If doesn't specify it. Client will only open one
> channel.
That's correct.
> And my smb.conf setup below can work for multichannel. Seems server
> and client don't require multiple network interfaces. And also don't need
> network team.
The client will try to reuse the same server interface if it has the RSS
capability flag set.
https://docs.microsoft.com/en-us/openspecs/windows_protocols/ms-smb2/fcd862d1-1b85-42df-92b1-e103199f531f
RSS flag means that this server interface can handle things in parallel
efficiently.
> But When test seedup, it isn't change. I use two vm in the same host. And
> each have 4 cpu and 1G memory. Maybe it's the problem.
Maybe the problem is not the number of CPU but the virtual network bus
bandwidth.
Let me explain: forget SMB and measure network speed (with iperf)
between 2 VMs. Let's say you get 1GB/s.
Now you add a virtual interface to the server VM and measure again on the 2
interfaces *at the same time*. You will see at most 500MB/s on each.
If you want to measure speedup, you need to limit the bandwidth of the
server interfaces with tc so that when you max out the server interface
bandwidth, you don't also max out the total virtual bus bandwidth.
If you limit 1MB/s on each, 1+1 = 2MB/s which is still << 1GB/s but you
will get speedup of 2.
> Thanks so much.
No problem :)
Cheers,
--
Aurélien Aptel / SUSE Labs Samba Team
GPG: 1839 CB5F 9F5B FB9B AA97 8C99 03C8 A49B 521B D5D3
SUSE Software Solutions Germany GmbH, Maxfeldstr. 5, 90409 Nürnberg, DE
GF: Felix Imendörffer, Mary Higgins, Sri Rasiah HRB 247165 (AG München)
prev parent reply other threads:[~2020-05-19 16:54 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <1371710722.29134084.1589816497911.JavaMail.zimbra@redhat.com>
2020-05-18 16:01 ` How to verify multichannel Xiaoli Feng
2020-05-18 17:17 ` Aurélien Aptel
2020-05-19 16:15 ` Xiaoli Feng
2020-05-19 16:37 ` Steve French
2020-05-19 16:54 ` Aurélien Aptel [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=878shnn9vu.fsf@suse.com \
--to=aaptel@suse.com \
--cc=linux-cifs@vger.kernel.org \
--cc=xifeng@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).