lkml.org 
[lkml]   [2002]   [May]   [25]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
SubjectSlabinfo memory usage after big box benchmarks
From
I don't know if this is related to the negative 
dentry thread or not. After running various
benchmarks on quad Xeon with 3.75 GB RAM,
there are some big differences in slab size.

inode_cache and dentry_cache have the biggest
difference between kernels.

total slab size inode_cache
2.4.19-pre7-rmap13 45.9 MB 9.9 MB
2.4.19-pre8 278.0 MB 202.8 MB
2.4.19-pre8-aa2 306.1 MB 233.9 MB
2.4.19-pre8-ac4 44.3 MB 4.5 MB
2.4.19-pre8-jam2 263.9 MB 182.1 MB

The benchmarks create a million or more temp
files during the run. Updatedb runs several
times too.

bonnie++ is one of the benches that creates a lot
of files. On runs with vastly different "number
of files to create", there was little difference
in slabinfo memory usage on 2.4.19-pre8-jam2.

More slabinfo detail at:
http://home.earthlink.net/~rwhron/kernel/bigbox/slabinfo.txt

Script to create slabinfo.txt in url above:
http://home.earthlink.net/~rwhron/kernel/bigbox/si_sum

Big box benchmarks:
http://home.earthlink.net/~rwhron/kernel/bigbox.html

--
Randy Hron

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 13:26    [W:0.038 / U:0.032 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site