Q: about xfs pre-allocation
Could someone comment on my [mis-]understanding in regards to what the note below said that was posted by someone else to another list. The pre-allocation behavior for XFS that the note describes doesn't jive w/what I thought happened and I was wondering if my brain was out of date or something (at least in regards to this topic! ;-)). Names elided from Original Message, below for no great reason...
I thought file space pre-allocation ended when you closed the file??
But this note from the open-suse list indicates that, at least with ext2, a kernel thread removes this later.
I thought the FS-space allocator gave *preference* to having the next file start at least "filesize%('allocsize || 64K')" from the end of the previous, BUT, if needed it will allocate space from the end of the previous file (rounded to fs-blocksize) if space is really that tight.
-------- Original Message --------
FFFF,
Modern filesystems use preallocation.
Per <http://ext2.sourceforge.net/2005-ols/paper-html/node6.html> ext2 already had it by 2005.
That means when a file is created and written to they automatically allocate a unused tail at the end of each file.
Then some time later (hours / days) they have a background kernel thread that scavenges any tails that are still unused.
The positive is that files (like logs) growing slowly over time won't get fragmented so badly.
The bad is that for highspeed filesystem filling tasks like a massive rsync, the disk usage is anomalously high for a while (hours / days).
With XFS you can disable pre-allocation via the allocsize mount parameter. (That parameter has been around many years. so yes 11.4 has preallocation for XFS at a minimum and ext3/ext4 I think.)
allocsize=size
Sets the buffered I/O end-of-file preallocation size when doing delayed allocation writeout (default size is 64KiB). Valid values for this option are page size (typically 4KiB) through to 1GiB, inclusive, in power-of-2 increments.
size = 0 disables preallocation and is probably smart on your distination backup disk.
I assume other filesystems have a way to disable preallocation as well.
FYI: I don't know how to determine the total amount of preallocation space on a filesystem. I'm sure it can be done somehow.
gggg
----
To unsubscribe, e-mail: [email protected] To contact the owner, e-mail: [email protected]
_______________________________________________
xfs mailing list
[email protected]
http://oss.sgi.com/mailman/listinfo/xfs
Quoted 8 lines · LA Walsh
Could someone comment on my [mis-]understanding in regards to what the note below said that was posted by someone else to another list. The pre-allocation behavior for XFS that the note describes doesn't jive w/what I thought happened and I was wondering if my brain was out of date or something (at least in regards to this topic! ;-)). Names elided from Original Message, below for no great reason...
It's mostly wrong information.
Just a suggestion: take anything advice given about XFS on lists other than this one with a grain of salt. There's lots of people out there who *think* they know how XFS works, but there's very few who actually know how XFS works.
Quoted 1 line
I thought file space pre-allocation ended when you closed the file??
In some cases, yes. In others, no. it depends on the workload and what is optimal for minimising fragmentation for that workload. If it wasn't removed, then oncthe file has been clean for 5 minutes it gets removed by a background worker.
Quoted 2 lines
But this note from the open-suse list indicates that, at least with ext2, a kernel thread removes this later.
ext2 does not use tail preallocationm, background kernel threads or workqueues to do stuff in the background anymore. It uses reservation windows to keep concurrent allocations apart. i.e. it now uses the algorithm that the link talks about being designed for ext3. :P
Quoted 5 lines
I thought the FS-space allocator gave *preference* to having the next file start at least "filesize%('allocsize || 64K')" from the end of the previous, BUT, if needed it will allocate space from the end of the previous file (rounded to fs-blocksize) if space is really that tight.
It depends on the context. delayed allocation is where XFS does all this stuff, and it is quite complex to get it to behave in a sane manner.
Quoted 8 lines
-------- Original Message --------
FFFF,
Modern filesystems use preallocation.
Per <http://ext2.sourceforge.net/2005-ols/paper-html/node6.html> ext2 already had it by 2005.
Irix used tail preallocation like ext2 did with EFS back in the late 1980s. XFS improved on it in the 90s via delayed allocation, before ext2 was even conceived. So using ext2 as an example of a filesystem using tail preallaction when talking about what XFS does today is kinda funny.... :P
Quoted 2 lines
With XFS you can disable pre-allocation via the allocsize mount parameter.
Wrong. It only allows you to fix the pre-allocation size - it cannot be turned off.
Quoted 2 lines
(That parameter has been around many years. so yes 11.4 has preallocation for XFS at a minimum and ext3/ext4 I think.)
It's probably been around longer than ext2....
Quoted 4 lines
allocsize=size
Sets the buffered I/O end-of-file preallocation size when doing delayed allocation writeout (default size is 64KiB).
Wrong. The default behaviour is dynamic preallocation.
Quoted 6 lines
Valid values for
this option are page size (typically 4KiB) through to 1GiB, inclusive, in power-of-2 increments.size = 0 disables preallocation and is probably smart on your distination backup disk.
No it doesn't - it's invalid.
# mount -o allocsize=0 /dev/vda /mnt/test
mount: wrong fs type, bad option, bad superblock on /dev/vda,
missing codepage or helper program, or other error
In some cases useful info is found in syslog - try
dmesg | tail or so
# dmesg |tail -3
[23843.767642] XFS (vda): Mounting Filesystem
[23843.835922] XFS (vda): Ending clean mount
[95648.513436] XFS (vda): invalid log iosize: 255 [not 12-30]
#Cheers,
Dave.
Signature · 7 lines
Dave Chinner [email protected] _______________________________________________ xfs mailing list [email protected] http://oss.sgi.com/mailman/listinfo/xfs
Quoted 10 lines · Dave Chinner
Quoted 2 lines · LA Walsh
Could someone comment on my [mis-]understanding in regards to what the note below said
It's mostly wrong information.
Just a suggestion: take anything advice given about XFS on lists other than this one with a grain of salt. There's lots of people out there who *think* they know how XFS works, but there's very few who actually know how XFS works.
---
Well, even thinking I had a clue, I wanted to check here, cuz, things change and I might give out totally wrong info... yuck...
I appreciate your expertise and tireless[?] efforts on a great FS.
=====
Brian,
Thanks for the different detail and emphasis... Sometimes I need to read something in multiple ways to fully get it (assuming I 'get' it...)
_______________________________________________
xfs mailing list
[email protected]
http://oss.sgi.com/mailman/listinfo/xfs
Quoted 11 lines · LA Walsh
Could someone comment on my [mis-]understanding in regards to what the note below said that was posted by someone else to another list. The pre-allocation behavior for XFS that the note describes doesn't jive w/what I thought happened and I was wondering if my brain was out of date or something (at least in regards to this topic! ;-)). Names elided from Original Message, below for no great reason...
I thought file space pre-allocation ended when you closed the file??
XFS normally frees post-eof preallocation on close, but there is the possibility for the preallocated space to hang around longer if a file is opened, written and closed repeatedly. You could observe this by repeatedly doing an open/write/close cycle on a file with xfs_io. After a few cycles, you'll see the blocks allocated to the file exceed the size of the file.
Quoted 8 lines
But this note from the open-suse list indicates that, at least with ext2, a kernel thread removes this later.
I thought the FS-space allocator gave *preference* to having the next file start at least "filesize%('allocsize || 64K')" from the end of the previous, BUT, if needed it will allocate space from the end of the previous file (rounded to fs-blocksize) if space is really that tight.
XFS has a background scanning thread for this purpose as of 3.8. It runs on a 5 minute interval by default and trims the speculative preallocation from inodes that haven't been written to recently.
Brian
Quoted 44 lines
-------- Original Message --------
FFFF,
Modern filesystems use preallocation.
Per <http://ext2.sourceforge.net/2005-ols/paper-html/node6.html> ext2 already had it by 2005.
That means when a file is created and written to they automatically allocate a unused tail at the end of each file.
Then some time later (hours / days) they have a background kernel thread that scavenges any tails that are still unused.
The positive is that files (like logs) growing slowly over time won't get fragmented so badly.
The bad is that for highspeed filesystem filling tasks like a massive rsync, the disk usage is anomalously high for a while (hours / days).
With XFS you can disable pre-allocation via the allocsize mount parameter. (That parameter has been around many years. so yes 11.4 has preallocation for XFS at a minimum and ext3/ext4 I think.)
allocsize=size
Sets the buffered I/O end-of-file preallocation size when doing delayed allocation writeout (default size is 64KiB). Valid values for this option are page size (typically 4KiB) through to 1GiB, inclusive, in power-of-2 increments.
size = 0 disables preallocation and is probably smart on your distination backup disk.
I assume other filesystems have a way to disable preallocation as well.
FYI: I don't know how to determine the total amount of preallocation space on a filesystem. I'm sure it can be done somehow.
gggg
----
Quoted 8 lines
To unsubscribe, e-mail: [email protected] To contact the owner, e-mail: [email protected]
_______________________________________________
xfs mailing list
[email protected]
http://oss.sgi.com/mailman/listinfo/xfs
_______________________________________________
xfs mailing list
[email protected]
http://oss.sgi.com/mailman/listinfo/xfs