Although not strictly part of MPICH itself, the Scalable Unix
Tools (SUT) [26] are a useful part of the MPICH programming
environment on workstation clusters. Basically, SUT implements
parallel versions of common Unix commands such as ls, ps,
cp, or rm. Perhaps the most useful is a cross between
find and ps that we call pfps (parallel find in the
process space). For example, one can find and send a KILL signal to
runaway jobs on a workstation network during a debugging session with
pfps -all -tn myprog -kill KILLor locate all of one's own jobs on the network that have been running for more than an hour with
pfps -all -o me -and -rtime 1:00 -printGraphical displays also show the load on each workstation and can help one choose the sub-collection of machines to run an MPICH job on. Details can be found in [26].