Compare with Ext3/4, the performance of file creation and deletion on btrfs is very poor. the reason is that btrfs must do a lot of b+ tree insertions, such as inode item, directory name item, directory name index and so on. If we can do some delayed b+ tree insertion or deletion, we can improve the performance, so we made this patch which implemented delayed directory name index insertion and deletion. Beside that, we found we must map the page every time we want to set a member variable of the inode item, it is inefficient. We just do it at first to reduce the times of mmap(). By this way, we can also improve the performance of file creation and deletion. I did a quick test by the benchmark tool[1] and found we can improve the performance of file creation by ~11%, and file deletion by ~14%. Before applying this patch: Create files: Total files: 50000 Total time: 1.188547 Average time: 0.000024 Delete files: Total files: 50000 Total time: 1.662012 Average time: 0.000033 After applying this patch: Create files: Total files: 50000 Total time: 1.057432 Average time: 0.000021 Delete files: Total files: 50000 Total time: 1.422851 Average time: 0.000028 You can also try out the patchset by pulling: git://repo.or.cz/linux-btrfs-devel.git perf-improve [1] http://marc.info/?l=linux-btrfs&m=128212635122920&q=p3 --- fs/btrfs/Makefile | 2 +- fs/btrfs/btrfs_inode.h | 2 + fs/btrfs/ctree.c | 13 +- fs/btrfs/ctree.h | 21 +- fs/btrfs/delayed-dir-index.c | 790 ++++++++++++++++++++++++++++++++++++++++++ fs/btrfs/delayed-dir-index.h | 92 +++++ fs/btrfs/dir-item.c | 61 +++- fs/btrfs/extent-tree.c | 21 ++ fs/btrfs/inode.c | 189 +++++++--- fs/btrfs/transaction.c | 9 + fs/btrfs/transaction.h | 2 + 11 files changed, 1117 insertions(+), 85 deletions(-) -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
