Visit Our UNIX and Linux User Community


Gfs2 vs xfs vs ext4


 
Thread Tools Search this Thread
Special Forums Hardware Filesystems, Disks and Memory Gfs2 vs xfs vs ext4
# 1  
Old 11-12-2009
Gfs2 vs xfs vs ext4

Looking for suggestions as to which filesystem to go with. I currently use gfs2 on hosts with 3.4tb useable. I understand gfs2 is being left behind but xfs and ext4 are not quite certified completely on CentOS 5.2. I have email storage hosts that have a decent i/o requirement and 12TB usable after raid 1+0. We tried ext3 but it was just too slow, I am not sure if ext4 is any faster or not. I also would like to hear peoples experience with recovery on these. How long to fsck, how successful are they, etc?

Any feedback is appreciated
# 2  
Old 11-12-2009
FWIW, we use EXT3 and it works fine.
# 3  
Old 11-12-2009
Quote:
Originally Posted by king_hippo
We tried ext3 but it was just too slow
Large ext3 partitions can be slow to fsck, but aren't that bad in operation. It's also important to note that the default ext3 mount options are brain-dead for large, heavily-cached systems; for instance, a commit interval of 5 seconds is rather small, and the default 'ordered' writing mode is extremely safe but sometimes a bottleneck.

On the other hand ext3 is excellent at safety. I've seen ext3 recover from horrible abuse.

Quote:
I am not sure if ext4 is any faster or not.
ext4 is somewhat faster but the difference is not gigantic. Its fsck is much faster than ext3's for partitions larger than hundreds of gigs. I don't feel it's quite mature, though. Only time will tell if it's as reliable as ext3.

Another filesystem you might consider is xfs. It's fairly mature, and designed for huge, fast transfers...
# 4  
Old 11-12-2009
ext3 fail

We tested ext3 before, it can't handle our load
# 5  
Old 11-13-2009
Quote:
Originally Posted by king_hippo
We tested ext3 before, it can't handle our load
With what mount options? As corona688 said there are *VAST* differences between the default options and a workload tuned setup.

Secondly it is not at all clear from what you have said whether it is the file system or your storage configuration that is the problem, nor is it possible for anyone to give you a good recommendation based only on the size of the storage.

1. What type of storage are you using.
2. What type of disks
3. If using a controller based array, what you optimized for sequential or random operation.
4. Will you mirror on-host or using hardware RAID
4. What is your storage block size.
5. What is the average size of your writes
6. What is the breakdown of reads/writes
7. Is it really optimal to create a 12TB LUN and put all you eggs in one basket, could you achieve better results with more smaller luns.
8. Do you have file hotspots ( certain files heavily accessed )
9. How many files per directory do you have ( files includes sub-directories) in the largest directory.


For a quick overvirew of options search for a presentation called "Choosing and Tuning Linux File Systems" written by Val Henson from the Intel Linux group.
# 6  
Old 11-19-2009
1. What type of storage are you using.
Penguin x8dtn, 24 1tb disks
2. What type of disks
sata
3. If using a controller based array, what you optimized for sequential or random operation.
dont know
4. Will you mirror on-host or using hardware RAID
controller based raid 10. six disks per raid set
4. What is your storage block size.
64k
5. What is the average size of your writes
dunno, guess is a few meg. No more than 16m. These are email attachments
6. What is the breakdown of reads/writes
not sure
7. Is it really optimal to create a 12TB LUN and put all you eggs in one basket, could you achieve better results with more smaller luns.
yes, we are using 3tb
8. Do you have file hotspots ( certain files heavily accessed )
just newer ones a bit
9. How many files per directory do you have ( files includes sub-directories) in the largest directory.
millions
# 7  
Old 11-19-2009
Millions of files in one giant directory isn't optimal in any filesystem I've ever heard of. I bet your giant untuned ext3 filesystem didn't have directory indexes enabled, which is highly reccomended for dirs larger than a few thousand entries.

Previous Thread | Next Thread
Test Your Knowledge in Computers #495
Difficulty: Easy
Blocks of code cannot be nested within each other to form sub-blocks.
True or False?

8 More Discussions You Might Find Interesting

1. UNIX for Advanced & Expert Users

VMWare,XFS and iSCSI issues!

I have a RHEL6 VM that requires the use of remote storage using iSCSI and XFS for the mount point. Here's the issue: With XFS you can't use the _netdev option for your mount point (pause for network) so my mount point doesn't mount properly because the network isn't up yet. I've moved the... (3 Replies)
Discussion started by: ksfolsom
3 Replies

2. Red Hat

Convert ext4 to ext3

Is there any way to conver ext4 to ext3 filesystem without formatting the partition/disk .. Had ext3 filesystem and had converted it to ext4 by issuing following command # tune2fs -O extents,uninit_bg,dir_index /dev/sda1 # fsck -pf /dev/sda1 # blkid /dev/sda1 /dev/sda1:... (1 Reply)
Discussion started by: Shirishlnx
1 Replies

3. Red Hat

XFS - Custom Kernel or Module?

Hey everyone. I am going to be using XFS for a project coming up. We're running RHEL 5.5. Simply typing modprobe xfs works just fine. The kernel module loads without any issue. Is there any issue with doing this and inserting "modprobe xfs" into /etc/rc.modules? Is there a major reason to... (0 Replies)
Discussion started by: msarro
0 Replies

4. SuSE

iscsi network share + xfs

Hello, we got a MD3000i used as a network share between two servers (say A and B). The problem we are facing is that file/directories created by server A are not visible on server B (and viceversa). It's not a problem with permission (chmod 777 doesn't help). The MD3000i was at first used ony... (0 Replies)
Discussion started by: neutrino
0 Replies

5. Linux

GFS2 needed or not?

Hey everyone. I am in the process of setting up an iSCSI SAN to function as a log storage device for a number of servers. All of the initiators see the volume, and originally I formatted it with ext3 and went on my merry way. However after some research I'm having concerns that I should nuke the... (1 Reply)
Discussion started by: msarro
1 Replies

6. UNIX for Advanced & Expert Users

[XFS] How to use real-time subvolume

Hi! I created filesystem XFS on partition hda8 with subvolume real-time on partition hda5: mkfs.xfs -r rtdev=/dev/hda5 /dev/hda8 and i mounted it: mount -t xfs -o rtdev=/dev/hda5 /dev/hda8 /xfs But I don't know how can I use this partition hda5 with subvolume real-time. I don't know how to... (3 Replies)
Discussion started by: Cadi2108
3 Replies

7. Filesystems, Disks and Memory

ext4 - ready for production system?

Gidday, Are you using ext4 for production system? Or is it better to opt for a more conservative strategy, like ext3 for instance? What are your experiences? Thanks in advance, Loc. (3 Replies)
Discussion started by: Loic Domaigne
3 Replies

8. Solaris

Does Solaris support XFS filesytem?

Hi, Does solaris support the XFS file sytems? If so, how can I mount a shared directory from another machine (which is using XFS)? thanks (3 Replies)
Discussion started by: orahi001
3 Replies

Featured Tech Videos