Skip to content
GitLab
Projects
Groups
Snippets
/
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Sign in
Toggle navigation
Menu
Open sidebar
ExaHyPE
ExaHyPE-Documentation
Commits
525314f3
Commit
525314f3
authored
May 15, 2019
by
Dominic Etienne Charrier
Browse files
Add information. How to refine the mesh further.
parent
9691c549
Changes
1
Hide whitespace changes
Inline
Side-by-side
21_distributed-memory.tex
View file @
525314f3
...
...
@@ -392,19 +392,19 @@ metric if you switch this feature on (\texttt{hotspot}, e.g.).
\subsection
{
Meshes for weak and strong scaling
}
ExaHyPE
distributes work by decomposing the tripartitioned spacetree into
\exahype
distributes work by decomposing the tripartitioned spacetree into
subtrees that are deployed to worker processes.
Only subtrees that overlap with the computational domain are deployed.
This constraint can be used to steer work distribution.
ExaHyPE
can scale the bounding box such that
\texttt
{
outside
\_
cells
\_
left
}
and/or
\texttt
{
outside
\_
cells
\_
right
}
cells are placed outside of the
\exahype
can scale the bounding box such that
\texttt
{
outside
\_
cells
\_
left
}
and/or
\texttt
{
outside
\_
cells
\_
right
}
cells are placed outside of the
domain while the latter is still resolved as accurately as specified in the spec file.
Furthermore, there is an option to place exactly
\texttt
{
ranks
\_
per
\_
dimension" on the coarse grid
}
.
(Note: This overrules
\texttt
{
outside
\_
cells
\_
right
}
but not
\texttt
{
outside
\_
cells
\_
left
}
.
This feature is particular interesting for weak scaling experiments as it can scale the number of cells per dimension
of a mesh by arbitrary integers.
An example is given
below
:
An example is given
in Fig.
\ref
{
fig:weak-scaling
}
:
\begin{figure}
[h]
\begin{center}
...
...
@@ -416,23 +416,25 @@ An example is given below:
\includegraphics
[width=0.4\textwidth]
{
sketches/weak-scaling-8.pdf
}
\end{center}
\caption
{
Weak scaling experiment growing the domain by factors
$
1
$
,
$
2
^
d
$
,
$
4
^
d
$
, and
$
8
^
d
$
.
\texttt
{
ranks
\_
per
\_
dimension
}
is chosen 1,2,4,8.
}
}
\label
{
fig:weak-scaling
}
\end{figure}
The bounding box scaling controls can be used to create interesting meshes as
the
$
36
^
d
$
cell mesh
given below
. This mesh was created with 9 outside cells
the
$
36
^
d
$
cell mesh
shown in Fig.
\ref
{
fig:strong-scaling
}
. This mesh was created with 9 outside cells
on the left side and 12 ranks per dimension. The mesh
is interesting since it can be perfectly distributed among
$
2
^
d
$
,
$
4
^
d
$
, and
$
12
^
d
$
processes.
The same mesh refined by a factor
$
3
^
i
$
,
$
i
\leq
1
$
can be constructed if
$
3
^
i
\times
9
$
bounding box outside cells
are placed to the left of the domain.
\begin{figure}
[h]
\begin{center}
\includegraphics
[width=0.5\textwidth]
{
sketches/strong-scaling-grid-223.pdf
}
\end{center}
\caption
{
%
A grid with
$
36
^
d
$
cells on the coarse grid which allows to use
$
1
$
,
$
2
^
d
$
,
$
4
^
d
$
, and
$
12
^
d
$
ranks
to uniformly distribute the load.
}
\caption
{
A grid with
$
36
^
d
$
cells on the coarse grid which allows to use
$
1
$
,
$
2
^
d
$
,
$
4
^
d
$
, and
$
12
^
d
$
ranks
to uniformly distribute the load.
}
\label
{
fig:strong-scaling
}
\end{figure}
\section
{
MPI troubleshooting and inefficiency patterns
}
...
...
Write
Preview
Supports
Markdown
0%
Try again
or
attach a new file
.
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment