diff options
Diffstat (limited to 'doc')
-rw-r--r-- | doc/documentation.tex | 62 |
1 files changed, 62 insertions, 0 deletions
diff --git a/doc/documentation.tex b/doc/documentation.tex index ac5251f..73982d0 100644 --- a/doc/documentation.tex +++ b/doc/documentation.tex @@ -65,6 +65,8 @@ PUGH can be compiled with or without MPI. Compiling without MPI results in an executable which can only be used on a single processor, compiling with MPI leads to an executable which can be used with either single or multiple processors. +(Section~\ref{pugh_understanding} describes how you can tell if your +executable has been compiled with or without MPI). For configuring with MPI, see the Cactus User's Guide. @@ -177,6 +179,66 @@ pugh::partition_2d_y=''15:15'' Note that an empty string for a direction will apply the automatic distribution. +\section{Understanding PUGH Output} + +\label{pugh_understanding} + +PUGH reports information about the processor decomposition to standard output +at the start of a job. This section described how to interpret that output. + +\vskip .3cm + +\noindent +{\bf Single Processor (no MPI)} + +\begin{itemize} + +\item{\bf Type of evolution} + +If an executable has been compiled for only single processor use +(without MPI), the first thing which PUGH reports is this fact: + +{\tt INFO (PUGH): Single processor evolution} + +\end{itemize} + + +\vskip .3cm + +\noindent +{\bf Multiple Processor (with MPI)} + + + +\vskip .3cm + +\begin{itemize} + +\item{\bf Type of evolution} + +If an executable has been compiled using MPI, the first thing which +PUGH reports is this fact, together with the number of processors +being used: + +{\tt INFO (PUGH): MPI Evolution on 3 processors} + + +\item{\bf Maximum load skew} + +The maximum load skew describes the variance in the number of gridpoints on +each processor, and is defined by + +$$\mbox{Max Skew} = 100 \;\;\frac{\mbox{Max Points}- \mbox{Min Points}}{\mbox{Average Points}} $$ + +For most purposes, the maximum skew should ideally be close to zero, +however if your simulation has a different load at different grid +points, or if you are running across processors with different +properties, the optimal skew could be quite different. + +By default, PUGH tries to minize the skew in gridpoints, however this may +be overriden by performing the load balancing manually. + +\end{itemize} \section{Useful Parameters} |