All of lore.kernel.org
 help / color / mirror / Atom feed
From: Christian Mayrhuber <may@sbox.tugraz.at>
To: reiserfs-list@namesys.com
Subject: Mongo benchmark for latest reiserfs data logging
Date: Fri, 9 May 2003 23:53:21 +0200	[thread overview]
Message-ID: <200305092353.21291.may@sbox.tugraz.at> (raw)

Hi,
 
I've run a couple of tests with the different data logging modes
writeback, ordered and journal.
All data logging modes were perfectly stable, a kernel compile and a
bonnie benchmark on /dev/md0 mounted on / did not have any negative impact
on the filesystem. reiserfsck said that it was clean.

Testparameters:
===============
KERNEL=2.4.21-rc1 + Chris Masons data logging + quota patches (nr's: 02 to 10)
TEST PARTITITION SIZE=10GByte on LVM
PROCESSOR=AMD Athlon XP 2000+
RAM=512MB DDR 266
HARDDISK=Seagate Baracuda IV, 60GB
TEST WRITE_BUFFER=4096
TEST FSTYPE=reiserfs
TEST REP_COUNTER=3
TEST GAMMA=0.0
TEST NPROC=3
TEST SYNC=off
SER FILE_SIZE=4096
SER BYTES=250000000

                    reiser writeback  reiser ordered  reiser journal
====================================================================
PHASE Create
STAT REAL_TIME                 84.53           81.87          114.36
STAT CPU_TIME                  33.17           32.39           36.55
STAT DF                      1515040         1515032         1515136
PHASE Copy
STAT REAL_TIME                278.98          289.65          338.67
STAT CPU_TIME                  50.36           51.53            53.3
STAT DF                      3030296         3030276         3030308
PHASE Append
STAT REAL_TIME               1256.53         1251.68            1404
STAT CPU_TIME                  38.52           37.18           47.19
STAT DF                      4014560         4014392         4014392
PHASE Modify
STAT REAL_TIME               1793.68         1665.74         1767.42
STAT CPU_TIME                  29.05           26.61           33.65
STAT DF                      4014560         4014392         4014392 
PHASE Overwrite
STAT REAL_TIME               2294.99         2047.15         2120.25
STAT CPU_TIME                  32.27           30.68           40.06
STAT DF                      4014560         4014392         4014392
PHASE Read
STAT REAL_TIME               1460.46         1411.12         1454.93
STAT CPU_TIME                  30.83           30.94           32.82
STAT DF                      4014560         4014392         4014392
PHASE Stats
STAT REAL_TIME                  8.83            8.93            8.89
STAT CPU_TIME                   1.79            1.83             1.7
STAT DF                      4014560         4014392         4014392
PHASE Delete
STAT REAL_TIME                101.59          111.34           108.8
STAT CPU_TIME                  55.51           58.12           55.32
STAT DF                            0               0               0

-- 
lg, Chris


             reply	other threads:[~2003-05-09 21:53 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2003-05-09 21:53 Christian Mayrhuber [this message]
2003-05-10 11:24 ` Mongo benchmark for latest reiserfs data logging Hans Reiser

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=200305092353.21291.may@sbox.tugraz.at \
    --to=may@sbox.tugraz.at \
    --cc=reiserfs-list@namesys.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.