> Any benchmark is going to be a benchmark of the OS as much > as it is going to be a benchmark of the filesystem. It's pretty > hard to separate the two. > > ZFS is best tested on Open Solaris. UFS is best tested on FreeBSD, > EXT3 is best tested on Linux, and HAMMER of course is best tested on > DragonFly. Plus the OS itself will depend greatly on whether the > machine environment is UP or SMP.
All true, of course. However, if you ran all of them on the same chunk of hardware, at least the hardware differences would be factored out and the results would be interesting to see, even if not necessarily indicative of the effects of the different file system structures and their implementations, imho. I'd like to see the talk, but there ain't no way I get to go there:-), rick
