ljishen/blogbench repository overview

⁠Blogbench - A filesystem benchmark for Unix (Version 1.1)

Blogbench is a portable filesystem benchmark that tries to reproduce the load of a real-world busy file server.

It stresses the filesystem with multiple threads performing random reads, writes and rewrites in order to get a realistic idea of the scalability and the concurrency a system can handle.

⁠Docker Image

⁠Sample Usage

docker run --rm \
    -v `pwd`:/root/results \
    ljishen/blogbench \
    /tmp /root/results/output.prof

⁠Available options

--commenters=<n> (-c <n>): number of comments posters
--help (-h): usage
--iterations=<n> (-i <n>): number of iterations
--readers=<n> (-r <n>): number of writers
--rewriters=<n> (-W <n>): number of rewriters
--sleep=<secs> (-s <secs>): delay after every iteration
--writers=<n> (-w <n>): number of writers

By default, there are 3 concurrent writers, 1 rewriters, 5 commenters and 100 readers, and it will run for 30 iterations.

See Homepage⁠ of Blogbench for more details.