Edit: Unfortunately, the images originally included in this post are gone, because of hosting problems in late 2009. My apologies.
I’ve been toying with a minimal Ubuntu installation identical to the one I built for the Hardy speed guide, but this time I used
data=journal as a flag for the default ext3 filesystem. Boot time results were a second slower than a clean installation, as shown here by bootchart. Default is on the left, tweaked on the right.
If you want a hardware profile, take a look here.
I’d say that system responsiveness was a bit slower, but I can’t tangibly prove that. I also tried transferring files into and out of the system over a network, and it seemed slower than the Crux systems that used to the same things. Again, I don’t have any real numbers to show that though, so it’s not really worth much to say it.
I’ll have to double-check the situations where
data=journal would be useful. I don’t think I’d recommend it for desktop systems just because it doesn’t boot as quickly, and that, for me, is usually a vague indicator of system performance.
Now that I’ve finished toying with this, I have another dastardly plan in the making. More on that later, though.