diff options
author | Mark Fasheh <mfasheh@suse.com> | 2010-04-06 05:17:15 +0400 |
---|---|---|
committer | Joel Becker <joel.becker@oracle.com> | 2010-05-06 05:18:07 +0400 |
commit | b07f8f24dfe54da0f074b78949044842e8df881f (patch) | |
tree | 8cc24b0a1e02a9b7f1241fbfecca50ac6881b938 | |
parent | 6b82021b9e91cd689fdffadbcdb9a42597bbe764 (diff) | |
download | linux-b07f8f24dfe54da0f074b78949044842e8df881f.tar.xz |
ocfs2: change default reservation window sizes
The default reservation size of 4 (32-bit windows) is a bit too ambitious.
Scale it back to 16 bits (resv_level=2). I have been testing various sizes
on a 4-node cluster which runs a mixed workload that is heavily threaded.
With a 256MB local alloc, I get *roughly* the following levels of average file
fragmentation:
resv_level=0 70%
resv_level=1 21%
resv_level=2 23%
resv_level=3 24%
resv_level=4 60%
resv_level=5 did not test
resv_level=6 60%
resv_level=2 seemed like a good compromise between not letting windows be
too small, but not so big that heavier workloads will immediately suffer
without tuning.
This patch also change the behavior of directory reservations - they now
track file reservations. The previous compromise of giving directory
windows only 8 bits wound up fragmenting more at some window sizes because
file allocations had smaller unused windows to poach from.
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
-rw-r--r-- | Documentation/filesystems/ocfs2.txt | 2 | ||||
-rw-r--r-- | fs/ocfs2/reservations.c | 7 | ||||
-rw-r--r-- | fs/ocfs2/reservations.h | 2 |
3 files changed, 6 insertions, 5 deletions
diff --git a/Documentation/filesystems/ocfs2.txt b/Documentation/filesystems/ocfs2.txt index 412df9095937..32339e584a9a 100644 --- a/Documentation/filesystems/ocfs2.txt +++ b/Documentation/filesystems/ocfs2.txt @@ -80,6 +80,6 @@ user_xattr (*) Enables Extended User Attributes. nouser_xattr Disables Extended User Attributes. acl Enables POSIX Access Control Lists support. noacl (*) Disables POSIX Access Control Lists support. -resv_level=4 (*) Set how agressive allocation reservations will be. +resv_level=2 (*) Set how agressive allocation reservations will be. Valid values are between 0 (reservations off) to 8 (maximum space for reservations). diff --git a/fs/ocfs2/reservations.c b/fs/ocfs2/reservations.c index 7fc6cfee95f1..87fa35791f18 100644 --- a/fs/ocfs2/reservations.c +++ b/fs/ocfs2/reservations.c @@ -55,9 +55,10 @@ static unsigned int ocfs2_resv_window_bits(struct ocfs2_reservation_map *resmap, if (!(resv->r_flags & OCFS2_RESV_FLAG_DIR)) { /* 8, 16, 32, 64, 128, 256, 512, 1024 */ bits = 4 << osb->osb_resv_level; - } else - bits = OCFS2_RESV_DIR_WINDOW_BITS; - + } else { + /* For now, treat directories the same as files. */ + bits = 4 << osb->osb_resv_level; + } return bits; } diff --git a/fs/ocfs2/reservations.h b/fs/ocfs2/reservations.h index 34bb308375c5..022aff601e15 100644 --- a/fs/ocfs2/reservations.h +++ b/fs/ocfs2/reservations.h @@ -22,7 +22,7 @@ #include <linux/rbtree.h> -#define OCFS2_DEFAULT_RESV_LEVEL 4 +#define OCFS2_DEFAULT_RESV_LEVEL 2 #define OCFS2_MAX_RESV_LEVEL 9 #define OCFS2_MIN_RESV_LEVEL 0 |