Flexible I/O Tester development
 help / color / mirror / Atom feed
From: Nicholas Krone <krone.nick@gmail.com>
To: fio@vger.kernel.org
Subject: Different percentages of job activity on a single drive, along with different i/o sizes for each drive
Date: Fri, 08 Aug 2014 00:49:38 -0700	[thread overview]
Message-ID: <53E48112.9070800@gmail.com> (raw)

Hi,

I'm currently trying to use FIO for some drive endurance testing and I 
have a few questions that I was hoping you guys could answer.

I need to stress a drive with a split blocksize range and ensure write 
verifications. Here's a small job I drafted up:

For example, a 100GB drive to test

[global]
ioengine=noop
ioengine=libaio
iodepth=4
rw=randwrite
direct=1
bssplit=512b/4:1024b/1:1536b/1:2048b/1:2560b/1:3072b/1:3584b/1:4096b/67:8192b/10:16384b/7:32768b/3:65536b/3 

do_verify=1
verify=meta
verify_backlog=512
error_dump=1
verify_dump=1
verify_fatal=1

[50-pct-ratio]
offset=0
size=5%

[30-pct-ratio]
offset=6g
size=15%

[20-pct-ratio]
offset=21g
size=80%


0-5% - 50 PCT write access (5 GB I/O size)
6%-20% - 30 PCT write access (15 GB I/O size)
21%-100% - 20 PCT write acces (80 GB I/O size)

Will using 3 sets of size params in each job work out correctly? Also, 
If during one of the jobs, a write verification occurs, will all the 
jobs fully stop because it wouldn't make sense for any of the test to 
continue if any write errors occur.

I read about the flow counter and a way control activity between jobs 
but I'm really unclear as to how implement it and the example didn't 
really help me out.

[50-pct-ratio]
flow=-2

[30-pct-ratio]
flow=3

[20-pct-ratio]
flow=5

Am I on the right track on anything? Any help would be great.


                 reply	other threads:[~2014-08-08  7:49 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=53E48112.9070800@gmail.com \
    --to=krone.nick@gmail.com \
    --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