fs_mark - Benchmark synchronous/async file creation
fs_mark [OPTIONS] -d DIRECTORY [-d DIRECTORY ...]
fs_mark is meant to give some low level bashing to file systems. The write pattern that it concentrates on is heavily synchronous IO across multiple directories, drives, etc.
- Print usage and exit.
- Keep files after each iteration.
- Run until filesystem is full.
- Method to be used for synchronizing changes to disk:
- 0: No Sync
- 1: fsyncBeforeClose
- 2: sync/1_fsync
- 3: PostReverseFsync
- 4: syncPostReverseFsync
- 5: PostFsync
- 6: syncPostFsync
- Use number subdirectories.
- Use number files in each subdirectory in round robin mode.
- Use directory. Can be specified multiple times.
- Log to logfile.
- Repeat the benchmark number of iterations.
- Use number of files per iteration.
- Use number of total bytes file names.
- Use number of random bytes in file names.
- Use number of bytes as size of each file.
- Use number of threads.
- Use number (of bytes per write() syscall.
fs_mark was written by Ric Wheeler <firstname.lastname@example.org>.
This manual page was written by Martin Steigerwald <email@example.com>, for the Debian project (but may be used by others).
|March 2012||fs_mark: Benchmark synchronous/async file creation|