This makes very little sense. The author shows a fsync() benchmark comparing an actualy fsync (Pixel on ext4 without nobarrier) with a nobarrier (no-op fsync) alternative. The only thing this benchmark shows is that a no-op is faster than the real thing.
Yes, computation is done in memory and is reasonably fast, with some clever caching. Right now the backend server is overloaded - I am working on it and you will be able to try it with custom addresses very soon.
Thank you extrapolate and ajaimk!
Yes, the backend is suffering a bit - but it will be back in a few minutes.
This is just a very limited demo of what BitIodine can do! For more info click on "About" or wait till new features ("Get insights") make it to the web interface.