On (Thu) Apr 16 2009 [15:42:01], Chris Mason wrote: > Hello everyone, > > Here's an updated (v4) patch for ext3 data=guarded mode. The first two > patches in the series are unchanged, and it looks like Linus pulled them > in this morning. I had written a small program that calculates the time needed to allocate a file and zero it using various methods (posix_fallocate, mmap, 4k-chunk writes, 8k-chunk writes). I did this on a 20G partition with each method creating a file 4G in size. The system has 3G RAM. The program that does this is at http://fedorapeople.org/gitweb?p=amitshah/public_git/alloc-perf.git;a=blob;f=test-file-zero-alloc-speed.c;hb=HEAD with the script to run it for the multiple filesystems at http://fedorapeople.org/gitweb?p=amitshah/public_git/alloc-perf.git;a=blob;f=run_test.sh;hb=HEAD I have a few results from those runs, time in seconds: # 4GiB file, kernel b0cbc861a3c05e634520b049b5cc27ad6febb51f filesystem posix-fallocate mmap chunk-4096 chunk-8192 ext2 74 96 761 81 ext3-writeback 87 97 202 93 ext3-ordered 86 94 134 104 ext4 0 84 120 91 xfs 0 84 274 81 reiserfs 85 84 187 98 btrfs 0 86 121 85 # 4GiB file, kernel 9f76208c33984ab777eace5d07a4e36e88703e02 + ext3-guarded filesystem posix-fallocate mmap chunk-4096 chunk-8192 ext3-guarded 85 97 459 90 ext3-writeback 86 95 140 94 ext3-ordered 86 96 277 95 btrfs 0 81 499 93 xfs 0 79 184 84 These were with a desktop running with a few terminal sessions and one konqueror session (to gauge the times a user will actually see while working on her desktop). Running the test in single user mode, I get the following results: # 4GiB file, kernel 9f76208c33984ab777eace5d07a4e36e88703e02 + ext3-guarded filesystem posix-fallocate mmap chunk-4096 chunk-8192 ext3-guarded 84 86 163 91 ext3-writeback 84 88 217 91 ext3-ordered 84 86 226 91 btrfs 0 76 86 79 ext4 0 73 195 76 Amit -- To unsubscribe from this list: send the line "unsubscribe linux-ext4" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
