[prev in list] [next in list] [prev in thread] [next in thread] 

List:       fuse-devel
Subject:    Re: [fuse-devel] slow reads
From:       Sergey Novgorodsky <svnovgorodsky () hotmail ! com>
Date:       2009-08-31 16:06:59
Message-ID: 25225997.post () talk ! nabble ! com
[Download RAW message or body]


Let me rephrase my question.
I ran another set of tests, comparing Fuse read performance vs ext3 -

Fuse: 2.8.0
Linux kernel: 2.6.27.21-0.1-default  (64 bit)
CPU: Xeon 3.0GHz
Memory: 8GB

I compiled fusexmp_fh.c :    gcc -Wall `pkg-config fuse --cflags --libs`
-lulockmgr fusexmp_fh.c -o fusexmp_fh
and started fuse with -omodules=subdir,subdir="/mydir/fuseread"
-oallow_other

I ran my script, which reads 261000 files of different sizes , total 11G,
using this command:
dd if=$filename bs=$bsize &> /dev/null

I tested with bs parameter 4k,8k,16k, and 32k

According to my tests, Fuse mounted on ext3  performed worse than plain ext3
by 34% to 62%:
4k - 62% 
8k - 58%
16k - 34%
32k - 46% 

Do those numbers look reasonable (under conditions I described)?
Can Fuse read performance be improved to be at most 10-15% worse than ext3?
If yes, how?

Thank you,
Sergey




The purpose of my testing is to determine if 

-- 
View this message in context: http://www.nabble.com/slow-reads-tp25151431p25225997.html
Sent from the Fuse - Dev mailing list archive at Nabble.com.


------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with 
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
fuse-devel mailing list
fuse-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/fuse-devel
[prev in list] [next in list] [prev in thread] [next in thread]