filebench(1) - Linux man page
Name
filebench - interprets WML script and generates appropriate filesystem workloadSynopsis
filebench [-f <filename>] [-h]Description
Filebench includes many features to facilitate file system benchmarking:
- [bu]
- Multiple workload types support via loadable personalities
- [bu]
- Ships with more than 40 pre-defined personalities, including the one that describe mail, web, file, and database servers behaviour
- [bu]
- Easy to add new personalities using reach Workload Model Language (WML)
- [bu]
- Multi-process and multi-thread workload support
- [bu]
- Configurable directory hierarchies with depth, width, and file sizes set to given statistical distributions
- [bu]
- Support of asynchronous I/O and process synchronization primitives
- [bu]
- Integrated statistics for throughput, latency, and CPU cycle counts per system call
Options
- -f <filename>
- use specified file as an input instead of stdin
- -h
- Display verbose help
Examples
- Run filebench interactively:
- $ filebench
- Start filebench interactive shell, filebench prompt will appear after that.
- filebench> load fileserver
- Load fileserver personality using load command.
- filebench> set $dir=/mnt
- After personality is loaded, tunables of the workload personality can be set. Here we change the benchmark directory to /mnt where the file system we want to benchmark is presumably mounted.
- filebench> run 60
- Start workload for 60 seconds. After 60 seconds of the run the statistics is printed and Filebench exits.
- Run filebench non-interactively:
- filebench -f /usr/share/filebench/workloads/fileserver.f
- Run filebench non-interactively, however, you need to add 'run <time>' to the end of the workload personality file before calling filebench.
Exit Status
- 0 if OK,
- 1 if workload file is not found.