For a single parallel run, try to use machines which have a similar level of performance.

Avoid using swap space. Try to use machines which can perform their assigned tasks using their own physical memory.

If you are using a network of workstations, use high speed Ethernet connections (100 Mbit is preferred to 10 Mbit).

High speed interconnects are only supported by the vendor specific run modes (HP MPI & SGI MPI). These are not supported by PVM.

Before starting a parallel job, check to see if your selected machines or processors can be used exclusively. If you have to share their use with others, this may increase the cpu static load.

Using two processors on a multi-processor machine is generally faster than using two processors on different machines. Dual CPU PCs are often the exception to this since the memory bus cannot provide enough memory access for both CPU's.

Try to avoid assigning more than one process per processor.

Use a sensible number of partitions. There should always be a balance between the individual partition size and the number of partitions.