An anonymous poster suggested the use of a program I completely missed called 'fileop' that exists in the Iozone distribution. From the source:
Unfortunately, my test blade is currently being used as part of the BlackBoard upgrade project so I won't get that back until June. But I plan on doing some tests to see how nss vs. reiser vs. reiser w/separate journal behaves.
Which happens to be a great way to test whonking huge directory trees. Balanced trees, of course, but still whonking huge. We see a problem with directories involving thousands of directory-entries, and this tool could be used to see if their behavior changes between when viewed over the network and when viewed locally.* Usage: fileop X
*
* X is a force factor. The total number of files will
* be X * X * X ( X ^ 3 )
* The structure of the file tree is:
* X number of Level 1 directorys, with X number of
* level 2 directories, with X number of files in each
* of the level 2 directories.
*
* Example: fileop 2
*
* dir_1 dir_2
* / \ / * sdir_1 sdir_2 sdir_1 sdir_2
* / \ / \ / \ / * file_1 file_2 file_1 file_2 file_1 file_2 file_1 file_2
*
* Each file will be created, and then 1 byte is written to the file.
*
Unfortunately, my test blade is currently being used as part of the BlackBoard upgrade project so I won't get that back until June. But I plan on doing some tests to see how nss vs. reiser vs. reiser w/separate journal behaves.