Parallel Programming Guide for HP-UX Systems

MPI

Running

Chapter 2 31

turning them off. This is accomplished by setting the time period of the s

option in the MPI_FLAGS environment variable (for example: s600). Time

is in seconds.

You can use the s[a][p]# option with the thread-compliant library as well

as the standard non thread-compliant library. Setting s[a][p]# for the

thread-compliant library has the same effect as setting MPI_MT_FLAGS=ct

when you use a value greater than 0 for #. The default value for the

thread-compliant library is sp0. MPI_MT_FLAGS=ct takes priority over the

default MPI_FLAGS=sp0.

Refer to “MPI_MT_FLAGS” on page 34 and “Thread-compliant library” on

page 21 for additional information.

Set MPI_FLAGS=sa1 to guarantee that MPI_Cancel works for canceling

sends.

]

Enables spin-yield logic.

is the spin value and is an integer between zero

and 10,000. The spin value speciﬁes the number of milliseconds a process

should block waiting for a message before yielding the CPU to another

process.

How you apply spin-yield logic depends on how well synchronized your

processes are. For example, if you have a process that wastes CPU time

blocked, waiting for messages, you can use spin-yield to ensure that the

process relinquishes the CPU to other processes. Do this in your appﬁle, by

setting y[#] to y0 for the process in question. This speciﬁes zero

milliseconds of spin (that is, immediate yield).

Specifying y without a spin value is equivalent to MPI_FLAGS=y10000.

If the time a process is blocked waiting for messages is short, you can

possibly improve performance by setting a spin value (between 0 and

10,000,) that ensures the process does not relinquish the CPU until after the

message is received, thereby reducing latency.

The system treats a nonzero spin value as a recommendation only. It does

not guarantee that the value you specify is used.

Refer to “Appﬁles” on page 42 for details about how to create an appﬁle and

assign ranks.

o Writes an optimization report to stdout. MPI_Cart_create and

MPI_Graph_create optimize the mapping of processes onto the virtual

topology if rank reordering is enabled.