Parallel Programming Guide for HP-UX Systems

MPI
Tuning
Chapter 2 65
Tuning
This section provides information about tuning HP MPI applications to improve performance.
The topics covered are:
• MPI_FLAGS options
• Message latency and bandwidth
• Multiple network interfaces
• Processor subscription
• MPI routine selection
• Multilevel parallelism
• Coding considerations
The tuning information in this chapter improves application performance in most but not all
cases. Use this information together with the output from counter instrumentation to
determine which tuning changes are appropriate to improve your application’s performance.
When you develop HP MPI applications, several factors can affect performance, whether your
application runs on a single computer or in an environment consisting of multiple computers
in a network. These factors are outlined in this chapter.
MPI_FLAGS options
The function parameter error checking is turned off by default. It can be turned on by setting
MPI_FLAGS=Eon.
If you are running an application stand-alone on a dedicated system, setting MPI_FLAGS=y
allows MPI to busy spin, thereby improving latency.
Message latency and bandwidth
Latency is the time between the initiation of the data transfer in the sending process and the
arrival of the first byte in the receiving process.
Latency is often dependent upon the length of messages being sent. An application’s
messaging behavior can vary greatly based upon whether a large number of small messages
or a few large messages are sent.