Commit Graph

27 Commits

Author SHA1 Message Date
Nik Clayton
044479f5ad Add information about the new options to newfs and tunefs which set the
expected average file size and number of files per directory.  Could do
with some fleshing out.
2001-04-10 10:36:44 +00:00
Kirk McKusick
a61ab64ac4 Directory layout preference improvements from Grigoriy Orlov <gluk@ptci.ru>.
His description of the problem and solution follow. My own tests show
speedups on typical filesystem intensive workloads of 5% to 12% which
is very impressive considering the small amount of code change involved.

------

  One day I noticed that some file operations run much faster on
small file systems then on big ones. I've looked at the ffs
algorithms, thought about them, and redesigned the dirpref algorithm.

  First I want to describe the results of my tests. These results are old
and I have improved the algorithm after these tests were done. Nevertheless
they show how big the perfomance speedup may be. I have done two file/directory
intensive tests on a two OpenBSD systems with old and new dirpref algorithm.
The first test is "tar -xzf ports.tar.gz", the second is "rm -rf ports".
The ports.tar.gz file is the ports collection from the OpenBSD 2.8 release.
It contains 6596 directories and 13868 files. The test systems are:

1. Celeron-450, 128Mb, two IDE drives, the system at wd0, file system for
   test is at wd1. Size of test file system is 8 Gb, number of cg=991,
   size of cg is 8m, block size = 8k, fragment size = 1k OpenBSD-current
   from Dec 2000 with BUFCACHEPERCENT=35

2. PIII-600, 128Mb, two IBM DTLA-307045 IDE drives at i815e, the system
   at wd0, file system for test is at wd1. Size of test file system is 40 Gb,
   number of cg=5324, size of cg is 8m, block size = 8k, fragment size = 1k
   OpenBSD-current from Dec 2000 with BUFCACHEPERCENT=50

You can get more info about the test systems and methods at:
http://www.ptci.ru/gluk/dirpref/old/dirpref.html

                              Test Results

             tar -xzf ports.tar.gz               rm -rf ports
  mode  old dirpref new dirpref speedup old dirprefnew dirpref speedup
                             First system
 normal     667         472      1.41       477        331       1.44
 async      285         144      1.98       130         14       9.29
 sync       768         616      1.25       477        334       1.43
 softdep    413         252      1.64       241         38       6.34
                             Second system
 normal     329         81       4.06       263.5       93.5     2.81
 async      302         25.7    11.75       112          2.26   49.56
 sync       281         57.0     4.93       263         90.5     2.9
 softdep    341         40.6     8.4        284          4.76   59.66

"old dirpref" and "new dirpref" columns give a test time in seconds.
speedup - speed increasement in times, ie. old dirpref / new dirpref.

------

Algorithm description

The old dirpref algorithm is described in comments:

/*
 * Find a cylinder to place a directory.
 *
 * The policy implemented by this algorithm is to select from
 * among those cylinder groups with above the average number of
 * free inodes, the one with the smallest number of directories.
 */

A new directory is allocated in a different cylinder groups than its
parent directory resulting in a directory tree that is spreaded across
all the cylinder groups. This spreading out results in a non-optimal
access to the directories and files. When we have a small filesystem
it is not a problem but when the filesystem is big then perfomance
degradation becomes very apparent.

What I mean by a big file system ?

  1. A big filesystem is a filesystem which occupy 20-30 or more percent
     of total drive space, i.e. first and last cylinder are physically
     located relatively far from each other.
  2. It has a relatively large number of cylinder groups, for example
     more cylinder groups than 50% of the buffers in the buffer cache.

The first results in long access times, while the second results in
many buffers being used by metadata operations. Such operations use
cylinder group blocks and on-disk inode blocks. The cylinder group
block (fs->fs_cblkno) contains struct cg, inode and block bit maps.
It is 2k in size for the default filesystem parameters. If new and
parent directories are located in different cylinder groups then the
system performs more input/output operations and uses more buffers.
On filesystems with many cylinder groups, lots of cache buffers are
used for metadata operations.

My solution for this problem is very simple. I allocate many directories
in one cylinder group. I also do some things, so that the new allocation
method does not cause excessive fragmentation and all directory inodes
will not be located at a location far from its file's inodes and data.
The algorithm is:
/*
 * Find a cylinder group to place a directory.
 *
 * The policy implemented by this algorithm is to allocate a
 * directory inode in the same cylinder group as its parent
 * directory, but also to reserve space for its files inodes
 * and data. Restrict the number of directories which may be
 * allocated one after another in the same cylinder group
 * without intervening allocation of files.
 *
 * If we allocate a first level directory then force allocation
 * in another cylinder group.
 */

  My early versions of dirpref give me a good results for a wide range of
file operations and different filesystem capacities except one case:
those applications that create their entire directory structure first
and only later fill this structure with files.

  My solution for such and similar cases is to limit a number of
directories which may be created one after another in the same cylinder
group without intervening file creations. For this purpose, I allocate
an array of counters at mount time. This array is linked to the superblock
fs->fs_contigdirs[cg]. Each time a directory is created the counter
increases and each time a file is created the counter decreases. A 60Gb
filesystem with 8mb/cg requires 10kb of memory for the counters array.

  The maxcontigdirs is a maximum number of directories which may be created
without an intervening file creation. I found in my tests that the best
performance occurs when I restrict the number of directories in one cylinder
group such that all its files may be located in the same cylinder group.
There may be some deterioration in performance if all the file inodes
are in the same cylinder group as its containing directory, but their
data partially resides in a different cylinder group. The maxcontigdirs
value is calculated to try to prevent this condition. Since there is
no way to know how many files and directories will be allocated later
I added two optimization parameters in superblock/tunefs. They are:

        int32_t  fs_avgfilesize;   /* expected average file size */
        int32_t  fs_avgfpdir;      /* expected # of files per directory */

These parameters have reasonable defaults but may be tweeked for special
uses of a filesystem. They are only necessary in rare cases like better
tuning a filesystem being used to store a squid cache.

I have been using this algorithm for about 3 months. I have done
a lot of testing on filesystems with different capacities, average
filesize, average number of files per directory, and so on. I think
this algorithm has no negative impact on filesystem perfomance. It
works better than the default one in all cases. The new dirpref
will greatly improve untarring/removing/coping of big directories,
decrease load on cvs servers and much more. The new dirpref doesn't
speedup a compilation process, but also doesn't slow it down.

Obtained from:	Grigoriy Orlov <gluk@ptci.ru>
2001-04-10 08:38:59 +00:00
Ruslan Ermilov
0a5779d45b - Backout botched attempt to introduce MANSECT feature.
- MAN[1-9] -> MAN.
2001-03-26 14:33:27 +00:00
Ruslan Ermilov
fe655281c5 Set the default manual section for sbin/ to 8. 2001-03-20 18:13:31 +00:00
Ben Smithurst
e50fa3d247 Fix 'tunefs -p'
Reviewed by:	sheldonh
2001-01-29 11:00:16 +00:00
Philippe Charnier
77edab906e The tunefs code assumed that the last argument was the device specification.
We need to parse the arguments first, then open the device (if
specified) and then apply the changes. This change will disallow the
(undocumented) use of multiple instances of the same argument on the
same command line for the sack of a better error message.

Other changes are:
1) the softupdates (-n) now issue a warning about remaining unchanged
2) the usage and man page is changed to specify "space | time" instead of
"optimization preference".

PR:		bin/23335
Submitted by:Mark Peek <mark@whistle.com>
2000-12-10 20:59:30 +00:00
Philippe Charnier
2af14b60ed Remove .Op when arg is required (special | filesystem). Document that at
least one flag is required and check this in the code. Make use of getopt(3).
Generalyze printing `... remains unchanged ...'.
2000-11-28 18:17:15 +00:00
Ruslan Ermilov
7c7fb079b9 mdoc(7) police: use the new features of the Nm macro. 2000-11-20 16:52:27 +00:00
Sheldon Hearn
060ac658cc Open the device read-only initially and re-open read-write if necessary
later.  This allows tunefs -p on mounted filesystems.

Side-effects:
	Use K&R prototypes.
	Use definitions from fcntl.h for the flags argument to open(2).

There are cosmetic differences between this and the submitted patch.

PR:		17143
Reported by:	Peter Edwards <peter.edwards@ireland.com>
Submitted by:	luoqi
2000-03-14 07:44:32 +00:00
Sheldon Hearn
ef8f7ac935 Remove single-space hard sentence breaks. These degrade the quality
of the typeset output, tend to make diffs harder to read and provide
bad examples for new-comers to mdoc.
2000-03-01 11:27:47 +00:00
Luoqi Chen
51003344a8 Remove unused #include and prototype declaration. 2000-01-30 05:24:55 +00:00
Luoqi Chen
b20ae6a0b3 Typo fix. While I am at it, remove the name translation from block to raw
device, they are equivalent now (or more accurately we no longer have block
devices).

Submitted by:	Gregory Sutter <gsutter@pobox.com>
2000-01-30 05:09:48 +00:00
Alexey Zelkin
eee003f78e Document a waring that tunefs(8) emits when enabling/disabling
soft updates on an unmounted filesystem.

PR:		docs/15657
Submitted by:	Mark Ovens <mark@ukug.uk.FreeBSD.org>
2000-01-07 13:05:04 +00:00
Peter Wemm
7f3dea244c $Id$ -> $FreeBSD$ 1999-08-28 00:22:10 +00:00
Luoqi Chen
9ea6e95e0d Check if an fs is mounted before checking if it is mounted read-only.
Pointed out by:		Mike Smith	<msmith@freebsd.org>
1999-07-19 21:04:25 +00:00
Luoqi Chen
7382c45aea Allow tuning of read-only mounted file system.
Reviewed by:	Bruce Evans	<bde@zeta.org.au>
1999-01-20 01:22:39 +00:00
Dag-Erling Smørgrav
90b2fe743a Sort options alphabetically. 1999-01-13 08:07:56 +00:00
Philippe Charnier
8679b1b4d8 Document -n (soft-update) flag.
Add rcsid, remove unused #includes. Sync usage() and SYNOPSIS.
1998-08-03 06:41:20 +00:00
Julian Elischer
b1897c197c Reviewed by: dyson@freebsd.org (john Dyson), dg@root.com (david greenman)
Submitted by:	Kirk McKusick (mcKusick@mckusick.com)
Obtained from:  WHistle development tree
1998-03-08 09:59:44 +00:00
Philippe Charnier
49d430cc1f Cosmetic in usage string. 1997-06-19 14:42:10 +00:00
Mike Pritchard
097d42f425 Correct some man page cross references and some file
locations.
1996-02-05 17:32:16 +00:00
Peter Wemm
f8e782eea4 Add (apparently) Larry McVoy's warning.... 1995-08-12 15:49:12 +00:00
Joerg Wunsch
16a7269ee5 When tuneing filesystems with tunefs, it is not obvious what the current
parameters are.  You can use dumpfs, but that's not obvious which settings
are tuneable, and is far from clear to the non-guru (it's like using a
hexdump of a tar archive to get a table-of-contents).

There is also an undocumented option in the man page that can be dangerous.
Suppose your disk driver decides to scramble all writes while you tell
tunefs to update all backup superblocks.

This suggested change adds a '-p' (print) switch to bring it in
line with some SVR4 systems.

(Slightly changed by me, mostly for optics. - joerg)

Submitted by:	peter@haywire.dialix.com
1995-06-25 17:46:13 +00:00
Rodney W. Grimes
5ebc7e6281 Remove trailing whitespace. 1995-05-30 06:12:45 +00:00
David Greenman
3669a02157 Changed manual page to conform to the reality in FreeBSD. 1995-03-15 07:08:33 +00:00
Garrett Wollman
404c1a5fc1 Convert to our man installation style. Also fixed long-standing bug
in `fastboot'/`fasthalt' in which the interpreter would hang around
after `reboot' or `halt' is run, causing an irritating ``Killed'' message.
1994-08-05 02:42:42 +00:00
Rodney W. Grimes
8fae3551ec BSD 4.4 Lite sbin Sources
Note:  XNSrouted and routed NOT imported here, they shall be imported with
usr.sbin.
1994-05-26 06:35:07 +00:00