Skip to contentSkip to conversation

Performance results on Experimental branch

DISCUSSION Steven Pratt 2 participants 4 messages Feb 25, 2009 Btrfs file system

lore
Share & tools
lore
#0 lore raw

Have completed the first run of the experimental branch on my RAID system and the results are encouraging.

Compared to 2.6.29-rc2:
Large file creates increase 50% plus across all thread counts. Mail Server is up about 20%
Random reads are pretty much flat
Random Writes are up 50% or more
Sequential reads took a 10% hit on single thread, but 20% gain on 128 threads.

All in all good progress. Results and graphs can be found here:

http://btrfs.boxacle.net/repository/raid/history/History.html

Steve

#1 lore raw

On Mi=E9rcoles 25 Febrero 2009 22:07:33 Steven Pratt escribi=F3:

Quoted 3 lines

All in all good progress. Results and graphs can be found here:
=20
http://btrfs.boxacle.net/repository/raid/history/History.html

Some graphs seem to be broken...btrfs gets a "transparent" color.

=46or example here:
http://btrfs.boxacle.net/repository/raid/history/History_Mail_server_si=
mulation._num_threads=3D16.FFSB_Ops_per_Sec.png
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" = in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

#2 lore raw
Quoted 11 lines · Diego Calleja

On Mi=E9rcoles 25 Febrero 2009 22:07:33 Steven Pratt escribi=F3:

=20

Quoted 4 lines

All in all good progress. Results and graphs can be found here:

http://btrfs.boxacle.net/repository/raid/history/History.html
=20

Some graphs seem to be broken...btrfs gets a "transparent" color.
=20

Unless I am missing something, what you are referring to is a simple=20 wraping/alignment issue in the key on the long name on the experimental= =20
btrfs. It is the Brown bar. Let me know if this is not the issue.

Steve

Quoted 2 lines

For example here:
http://btrfs.boxacle.net/repository/raid/history/History_Mail_server_=

simulation._num_threads=3D16.FFSB_Ops_per_Sec.png

Quoted 2 lines

--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs=

" in

Quoted 3 lines

the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
=20

--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" = in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

#3 lore raw

On Mi=E9rcoles 25 Febrero 2009 23:55:08 Steven Pratt escribi=F3:

Quoted 1 line

Unless I am missing something, what you are referring to is a simple=

=20

Quoted 1 line

wraping/alignment issue in the key on the long name on the experiment=

al=20

Quoted 1 line

btrfs. It is the Brown bar. Let me know if this is not the issue.

doh, you're right. Sorry, it's time to sleep here ;(
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" = in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html