I moved the scripts to github where people can send me pull requests if they want something changed. Also, there is nothing about these scripts specific to Gluster. Finally, there is a new script, netperf-rpc-pairs.sh, which in my opinion may be the more important script of the two for predicting performance of distributed storage.
I think there is not enough information available to people on how to profile existing applications - not everyone can run synthetic workloads on their systems, particularly production systems where synthetic workloads could adversely impact the users.
