small blocks or more inodes?
To follow up on an email i sent not to long ago I have been digging
through ext2 documentation to try to find what is the best setup for
storing tons of small files(tens/hundreds of thousands).
The filesystem is composed of 2 9.1GB Ultra160 drives in raid1.
I have 2 identical systems setup with different parameters:
mail-wa:~# more /root/dumpe2fs-md0
Filesystem volume name: <none>
Last mounted on: <not available>
Filesystem UUID: 236531e2-03b8-4aac-ad6a-5ac880857bae
Filesystem magic number: 0xEF53
Filesystem revision #: 1 (dynamic)
Filesystem features: filetype sparse_super
Filesystem state: not clean
Errors behavior: Continue
Filesystem OS type: Linux
Inode count: 560128
Block count: 8956096
Reserved block count: 89560
Free blocks: 8883337
Free inodes: 560116
First block: 1
Block size: 1024
Fragment size: 1024
Blocks per group: 8192
Fragments per group: 8192
Inodes per group: 512
Inode blocks per group: 64
and
mail-ca:~# more /root/dumpe2fs
Filesystem volume name: <none>
Last mounted on: <not available>
Filesystem UUID: f79dcbca-4c5d-4b13-9a11-a46ee1bca1df
Filesystem magic number: 0xEF53
Filesystem revision #: 1 (dynamic)
Filesystem features: filetype sparse_super
Filesystem state: not clean
Errors behavior: Continue
Filesystem OS type: Linux
Inode count: 1121664
Block count: 2239024
Reserved block count: 111951
Free blocks: 2203810
Free inodes: 1121652
First block: 0
Block size: 4096
Fragment size: 4096
Blocks per group: 32768
Fragments per group: 32768
Inodes per group: 16256
Inode blocks per group: 508
Last mount time: Mon Dec 11 21:55:32 2000
I tried running bonnie++ and told it to create 8GB of data in 4 million
files but it didn't work, all i got was "aborted" after some time.
any opinions as to which is more effective at storing small files?(or
maybe another option ...)
thanks!
nate
:::
http://www.aphroland.org/
http://www.linuxpowered.net/
aphro@aphroland.org
4:15pm up 88 days, 1:33, 3 users, load average: 0.05, 0.03, 0.00
Reply to: