From: Patrick Goetz <pgoetz@cs.utexas.edu>
To: fio@vger.kernel.org
Subject: NFS performance surpasses network bandwidth?!
Date: Wed, 24 Apr 2024 14:28:58 -0500 [thread overview]
Message-ID: <8af8b9ad-92dc-41b8-84ae-c4ec91a73cc3@cs.utexas.edu> (raw)
Hi -
I must either be doing something wrong or am interpreting the results
wrong. I'm running this command on an NFS mounted filesystem:
$ fio --name=random-write --ioengine=posixaio --rw=randwrite --bs=64k
--size=256m --numjobs=16 --iodepth=16 --runtime=60 --time_based
--end_fsync=1
With this result:
Run status group 0 (all jobs):
WRITE: bw=5258MiB/s (5513MB/s), 152MiB/s-766MiB/s (160MB/s-803MB/s),
io=498GiB (534GB), run=72987-96944msec
The client system is connected to the NFS server over a single 10Gb/s
connection. 5 GB/s far exceeds the max bandwidth of the wire!
What am I missing? I ran the test 3 times, each time checking my
location to make sure I was in an NFS-mounted directory. One test
resulted in 3.5GB/s write performance, which is still greater than the
theoretical max bandwidth for the network connection.
reply other threads:[~2024-04-24 19:28 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
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=8af8b9ad-92dc-41b8-84ae-c4ec91a73cc3@cs.utexas.edu \
--to=pgoetz@cs.utexas.edu \
--cc=fio@vger.kernel.org \
/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 read-only IMAP folder(s) and NNTP newsgroup(s).