HCL - New pages [en]

Virtual memory overcommit

2013-09-09T13:29:46Z

Davepc:

== Paging and the OOM-Killer ==

Due to the nature of experiments our group runs, we often induce heavy paging and complete exhaustion of available memory on certain nodes. Linux has a pair of strategies to deal with heavy memory use. First, is overcommitting. This is where a process is allowed allocate or fork even when there is no more memory available. You can seem some interesting numbers here:[http://opsmonkey.blogspot.com/2007/01/linux-memory-overcommit.html]. The assumption is that processes may not use all memory that they allocate and failing on allocation is worse than failing at a later date when the memory use is actually required. More processes may be supported by allowing them to allocate memory (provided they do not use it all). The second part of the strategy is the Out-of-Memory killer (OOM Killer). When memory has been exhausted and a process tries to use some 'overcommitted' part of memory, the OOM killer is invoked. It's job is to rank all processes in terms of their memory use, priority, privilege and some other parameters, and then select a process to kill based on the ranks.

The argument for using overcommital+OOM Killer is that rather than failing to allocate memory for some random unlucky process, which as a result would probably terminate, the kernel can instead allow the unlucky process to continue executing and then make a some-what-informed decision on which process to kill. Unfortunately, the behaviour of the OOM-killer sometimes causes problems which grind the machine to a complete halt, particularly when it decides to kill system processes. There is a good discussion on the OOM-killer here: [http://lwn.net/Articles/104179/]

For this reason overcommit has been disabled on the HCL cluster.

cat /proc/sys/vm/overcommit_memory
2
cat /proc/sys/vm/overcommit_ratio
100

To restore to default overcommit

# echo 0 > /proc/sys/vm/overcommit_memory
# echo 50 > /proc/sys/vm/overcommit_ratio

<br>

Run the following two programmes from here: [http://www.win.tue.nl/~aeb/linux/lk/lk-9.html www.win.tue.nl/~aeb/linux/lk/lk-9.html] to test results.

Demo program 1: allocate memory without using it.

<source lang="">#include <stdio.h>
#include <stdlib.h>

int main (void) {
int n = 0;

while (1) {
if (malloc(1<<20) == NULL) {
printf("malloc failure after %d MiB\n", n);
return 0;
}
printf ("got %d MiB\n", ++n);
}
}</source><br>

Demo program 2: allocate memory and actually touch it all.

<source lang="">#include <stdio.h>
#include <string.h>
#include <stdlib.h>

int main (void) {
int n = 0;
char *p;

while (1) {
if ((p = malloc(1<<20)) == NULL) {
printf("malloc failure after %d MiB\n", n);
return 0;
}
memset (p, 0, (1<<20));
printf ("got %d MiB\n", ++n);
}
}</source>

== Manually Limit the Memory on the OS level ==

as root edit /etc/default/grub

GRUB_CMDLINE_LINUX_DEFAULT="quiet mem=128M"

then run the command

update-grub
reboot

BlueGene/P

2013-03-19T15:00:15Z

Xalid:

Some members of HCL have access to Shaheen BlueGene/P at King Abdullah University of Science and Technology ([http://www2.hpc.kaust.edu.sa/documentation/shaheen/]) . In addition, some members has access to BleGene/P at West University of Timisoara, Romania ([http://hpc.uvt.ro/infrastructure/bluegenep/]).

===== Fupermod on Shaheen BlueGene/P =====

In order to compile fupermod on the BG/P the following commands should be run to load some libraries:

#module load bluegene
#module load essl
#module load gsl

Then, configure command can be executed as follows:

/fupermod/configure --with-gsl-dir=/opt/share/math_libraries/gsl/ppc64/IBM --with-blas=essl CFLAGS="-O3 -qarch=450 -qtune=450" --with-essl-dir=/opt/share/ibmmath/essl/4.4/<br>

Desktop Backup

2012-12-06T18:30:32Z

Davepc:

Members of the HCL group may backup their desktops to heterogeneous server in the following directory:

heterogeneous:/home/desktops/<user>

This can easly be done with rsync as follows:

rsync -axv /home/<your_desktop_username>/ <user>@heterogeneous:/home/desktops/<user>/ --exclude-from=.bkup_excludes

and make the file .backup_excludes with files and directorys you would like to exclude, for example your download folder, internet cache, etc. An example of Daves excludes file:

.Skype/
.Trash-1000/
.adobe/
.cache/
.config/chromium
.dropbox/
.mozilla/
.ssh/
.svn/
.thumbnails/
.thunderbird/
Downloads/
Dropbox/
backups/

CUDA SDK

2012-07-12T11:04:57Z

Zhongziming: Created page with "http://developer.nvidia.com/gpu-computing-sdk"

http://developer.nvidia.com/gpu-computing-sdk

Bash Scripts

2012-05-28T09:54:13Z

Davepc:

A collection of useful bash scripts here.

Open in vi all .c files containing a string
vim -p `grep STRING *.[c]| cut -f1 -d ":"| uniq`

Pgfplot

2012-05-09T11:43:11Z

Quintin:

PgfPlot is a package which let you do the same things as gnuplot but directly inside your latex file.
And this has a lot of advantages:
- no additional file
- no conversion from eps or ps or png to the pdf
- the quality is really improved
- the picture text has the same font as the rest of the document

=Preambule=
<source lang="latex">
\usepackage{tikz}
\usepackage{pgfplots}
</source>

=example=

<source lang="latex">
\begin{tikzpicture}[transform shape,scale=0.9]
\begin{axis}[
ylabel style={yshift=-15pt}, % place of the ylabel
width=\textwidth,height=7.4cm, %size of the picture
xlabel=Number of processors,
ylabel=Execution time ($10^3$ s),
legend style={at={(.9,.9)}}, % place of the legend
ymin=0, ymax=8.8, % range for the y axe
mark size=1.5 % size of the points
]
\addplot+[] table[header=false,x index=0,y expr=\thisrowno{2}/1000]{data};
\addplot+[] table[header=false,x index=0,y expr=\thisrowno{2}/1000]{data};
\addplot+[] table[header=false,x index=0,y expr=\thisrowno{2}/1000]{data};

\legend{WSCOM$_{\text{pf}}$,WSCOM,list\_min};

\end{axis}\end{tikzpicture}

</source>

a part of the file "data" is shown just after:

<source lang="text">
1 list_min 8241.04532500001 0.592721444672621 400
2 list_min 4276.264 0.541485563775196 400
3 list_min 2840.83145 0.240012784271948 400
4 list_min 2171.53825 0.204766932333766 400
5 list_min 1771.9397 0.187049380812677 400
</source>

Matplotlib

2012-01-30T17:28:02Z

Jun.zhu: Created page with "http://matplotlib.sourceforge.net/"

http://matplotlib.sourceforge.net/

PGF/Tikz

2012-01-30T17:26:54Z

Davepc:

* [http://en.wikipedia.org/wiki/PGF/TikZ Tikz on Wikipedia]
* [http://www.ctan.org/tex-archive/graphics/pgf/base/doc/generic/pgf/pgfmanual.pdf PGF manual]
* [http://www.texample.net/tikz/examples/ Examples]

= Write a figure =

The preamble of the latex file must contain: <source lang="latex">\usepackage{tikz}</source> Some optional libraries could be add like this: <source lang="latex">\usetikzlibrary{calc}</source>

To start a figure, the code must be inside the tikzpicture environment like this: <source lang="latex">\begin{tikzpicture} ... TikZ code here... \end{tikzpicture}</source>

= Exemple =

<source lang="latex">
% Author: Quintin Jean-Noël
% <http://moais.imag.fr/membres/jean-noel.quintin/>
\documentclass{article}
\usepackage{tikz}
\usetikzlibrary[topaths]
% A counter, since TikZ is not clever enough (yet) to handle
% arbitrary angle systems.
\newcount\mycount
\begin{document}
\begin{tikzpicture}[transform shape]
%the multiplication with floats is not possible. Thus I split the loop
%in two.
\foreach \number in {1,...,8}{
% Computer angle:
\mycount=\number
\advance\mycount by -1
\multiply\mycount by 45
\advance\mycount by 0
\node[draw,circle,inner sep=0.125cm] (N-\number) at (\the\mycount:5.4cm) {};
}
\foreach \number in {9,...,16}{
% Computer angle:
\mycount=\number
\advance\mycount by -1
\multiply\mycount by 45
\advance\mycount by 22.5
\node[draw,circle,inner sep=0.125cm] (N-\number) at (\the\mycount:5.4cm) {};
}
\foreach \number in {1,...,15}{
\mycount=\number
\advance\mycount by 1
\foreach \numbera in {\the\mycount,...,16}{
\path (N-\number) edge[->,bend right=3] (N-\numbera) edge[<-,bend
left=3] (N-\numbera);
}
}
\end{tikzpicture}
\end{document}
</source>

= Voir aussi =

*[[Pfgplot]]

Beamer

2012-01-30T17:23:53Z

Root: Created page with "[http://en.wikipedia.org/wiki/Beamer_(LaTeX) Beamer] - a package for presentation slides"

[http://en.wikipedia.org/wiki/Beamer_(LaTeX) Beamer] - a package for presentation slides

BibTeX

2012-01-30T17:23:33Z

Root: Created page with "[http://en.wikipedia.org/wiki/BibTeX BibTeX] - a reference management software"

[http://en.wikipedia.org/wiki/BibTeX BibTeX] - a reference management software

FORTRAN

2012-01-30T16:55:05Z

Root: Created page with "[http://www.yolinux.com/TUTORIALS/LinuxTutorialMixingFortranAndC.html Tutorial on mixing FORTRAN and C/C++ code]"

[http://www.yolinux.com/TUTORIALS/LinuxTutorialMixingFortranAndC.html Tutorial on mixing FORTRAN and C/C++ code]

GDB

2012-01-30T16:54:29Z

Davepc:

Debugging with GDB

compile programme with -g -o0 (or ./configure --enable-debug)

For serial programme (or MPI running with 1 process)
gdb ./programme_name

To debug MPI application in parallel
add this line to somewhere in the code:
if (!rank) getc(stdin); MPI_Barrier(MPI_COMM_WORLD);

Then run the code and it will hang on that line.

If you get the gdb error:
fupermod_....c: No such file or directory.

Run the following command to add to gdb's source directories to be searched:
directory ~/fupermod/

BitTorrent (B. Cohen's version)

2011-12-13T11:42:45Z

Kiril:

The modified bittorrent tarball is currently under the MPIBlib repository:

svn co https://hcl.ucd.ie/repos/CPM/trunk/MPIBlib/tests/bittorrent/bittorrent.tar.gz

* Extract this under your home directory
** You might need
mv $HOME/lib/python $HOME/lib/python2.6

* export following variables to tweak the way Python paths are searched:
export PYTHONPATH=/usr/lib/python2.6/:<local-installation-path-of-Python-libs>
* Modify logfile path in /home/kdichev/lib/python2.6/BitTorrent/StorageWrapper.py and create according directory structure
* Create a file of any size consisting of "s" characters only.
* Create torrent file:
btmakemetafile myfile.ext http://<frontend on G5K>:6969/announce
* Stick the torrent file into $HOME/public at the frontend.
** It is then available from within G5K under http://public.lille.grid5000.fr/~kdichev/{torrent file}
* Start the tracker on the frontend:
bttrack --port 6969 --dfile dstate
* Book nodes interactively:
qsub -I -lnodes=...
* Start one client that has the full copy
ssh <one-node> btdownloadheadless --url http://public.lille.grid5000.fr/~kdichev/{torrent file} &

* On all other nodes, launch the client at the same time:
script-per-process.sh
#!/bin/bash
cd /tmp ; btdownloadheadless --display_interval 100 --url http://public.lille.grid5000.fr/~kdichev/{torrent file}
run.sh
mpirun -n 14 --machinefile hostfile $PWD/script-per-process.sh

NLOPT

2011-09-29T18:54:39Z

Davepc: Created page with "[http://ab-initio.mit.edu/wiki/index.php/NLopt Download nlopt from here] Unpack and configure with: ./configure --with-cxx --enable-shared make make install"

[http://ab-initio.mit.edu/wiki/index.php/NLopt Download nlopt from here]
Unpack and configure with:
./configure --with-cxx --enable-shared
make
make install