Here is an interesting idea: recording traces of the filesystem operations applications do, and allowing these traces to be replayed later.
> ioreplay is mainly intended for replaying of recorded (using strace) IO > traces, which is useful for standalone > benchmarking. It provides many features to ensure validity of such > measurements. http://code.google.com/p/ioapps/ Sounds like something we should consider doing for HDFS performance testing... regards, Colin