From gallen at cct.lsu.edu  Thu Jan  7 10:04:24 2010
From: gallen at cct.lsu.edu (Allen Gabrielle)
Date: Thu, 7 Jan 2010 10:04:24 -0600
Subject: [Users] Reminder: Cactus Survey
References: <8A19FA31-ADA5-4DD7-B834-C0CB4CDDDDB4@cct.lsu.edu>
Message-ID: <E03D2873-5120-44A7-9953-3247E957BA84@cct.lsu.edu>


If you didn't yet fill out our survey, please try and fill it out in the next week, we would very much appreciate your support with this. 

Thanks

Gabrielle

Begin forwarded message:

> From: Allen Gabrielle <gallen at cct.lsu.edu>
> Date: December 16, 2009 10:41:08 PM CST
> To: users at cactuscode.org, developers at lists.carpetcode.org
> Subject: Cactus Survey
> 
> 	
> Dear Cactus users, 
> 
> We've designed a short survey to gather feedback on how you are using Cactus and Carpet, what we should be targeting for future development, and how to better support the use and development of Cactus applications and associated software. Please take a few minutes to fill this out; we really appreciate your comments.
> 
> http://surveymonkey.com/s.aspx?sm=S0g62YndjGMx1MPvMxW9fA%3d%3d
> 
> Thanks!
> 
> The Cactus Team
> 
> --
> Gabrielle Allen
> Associate Professor
> Department of Computer Science
> Louisiana State University
> gallen at cct.lsu.edu
> 
> 
> 
> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.cactuscode.org/pipermail/users/attachments/20100107/4f92df66/attachment.html 

From korobkin at phys.lsu.edu  Fri Jan  8 12:00:27 2010
From: korobkin at phys.lsu.edu (Oleg Korobkin)
Date: Fri, 08 Jan 2010 20:00:27 +0200
Subject: [Users] CVS: TAT/Slab not found
Message-ID: <4B4772BB.700@phys.lsu.edu>

Hi,

I was checking out a new installation of Cactus and got the following
error when trying to download TAT/Slab:

Could not run command: cvs   -Q -z9  -d
:pserver:cvs_anon at cvs.cactuscode.org:/cactusdevcvs co    TAT/Slab 2>&1,
because of No such file or directory

Did we move the TAT arrangement to subversion, and if yes, then where?

Thanks for your help,
 - Oleg Korobkin

From schnetter at cct.lsu.edu  Fri Jan  8 12:03:55 2010
From: schnetter at cct.lsu.edu (Erik Schnetter)
Date: Fri, 8 Jan 2010 12:03:55 -0600
Subject: [Users] CVS: TAT/Slab not found
In-Reply-To: <4B4772BB.700@phys.lsu.edu>
References: <4B4772BB.700@phys.lsu.edu>
Message-ID: <D54C8E31-5E28-47D9-A858-879AA82DD0E6@cct.lsu.edu>

On Jan 8, 2010, at 12:00 , Oleg Korobkin wrote:

> Hi,
> 
> I was checking out a new installation of Cactus and got the following
> error when trying to download TAT/Slab:
> 
> Could not run command: cvs   -Q -z9  -d
> :pserver:cvs_anon at cvs.cactuscode.org:/cactusdevcvs co    TAT/Slab 2>&1,
> because of No such file or directory
> 
> Did we move the TAT arrangement to subversion, and if yes, then where?


No, the TAT arrangement is still in cvs.  However, it is in a different location:

	cvs.cactuscode.org:/arrangements

-erik

-- 
Erik Schnetter <schnetter at cct.lsu.edu>   http://www.cct.lsu.edu/~eschnett/





From romuald at th.if.uj.edu.pl  Sat Jan  9 04:51:45 2010
From: romuald at th.if.uj.edu.pl (Romuald Janik)
Date: Sat, 9 Jan 2010 11:51:45 +0100 (CET)
Subject: [Users] gxx_personality problem when compiling WaveDemo
Message-ID: <20100109114047.G12459@th.if.uj.edu.pl>

Dear All,

I have downladed Cactus (using GetCactus script) and the WaveDemo.
make WaveDemo-config runs ok [I give the output at the end of this mail]
but when I run
make WaveDemo
I get at the very end an error about undefined reference to 
`__gxx_personality_v0'   (output below)

My system is Ubuntu 8.04, gcc is version 4.2.4, as far as I know make is
a synonim of gmake on Ubuntu.
I would be grateful for some advice. Am I missing some library/option 
which is not set automatically by make WaveDemo-config?
Best regards,
Romuald
PS. Below are the outputs of the make commands..

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

end of output of make WaveDemo:

Creating cactus_WaveDemo in /home/romuald/cactus/Cactus/exe from 
CactusBase/Boundary CactusBase/CartGrid3D CactusBase/CoordBase 
CactusConnect/HTTPD CactusConnect/HTTPDExtra CactusWave/IDScalarWaveC 
CactusBase/IOASCII CactusBase/IOBasic CactusIO/IOJpeg CactusBase/IOUtil 
CactusPUGHIO/IsoSurfacer CactusBase/LocalReduce CactusPUGH/PUGH 
CactusPUGH/PUGHReduce CactusPUGH/PUGHSlab CactusConnect/Socket 
CactusBase/SymBase CactusBase/Time CactusWave/WaveBinarySource 
CactusWave/WaveToyC CactusExternal/jpeg6b
/home/romuald/cactus/Cactus/configs/WaveDemo/lib/libthorn_Cactus.a(flesh.cc.o):(.eh_frame+0x11): 
undefined reference to `__gxx_personality_v0'
collect2: ld returned 1 exit status
make[1]: *** [/home/romuald/cactus/Cactus/exe/cactus_WaveDemo] Error 1
make: *** [WaveDemo] Error 2

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%55

Output of make WaveDemo-config:

romuald at amorpha:~/cactus/Cactus$ make WaveDemo-config
________________________________________________________________________
Cactus - version: 4.0.b16
Reconfiguring WaveDemo.
creating cache ./config.cache
checking host system type... i686-pc-linux-gnu
Setting FPP to /lib/cpp
Setting FPPFLAGS to -traditional
checking whether make sets ${MAKE}... yes
checking for gcc... gcc
checking whether the C compiler (gcc  ) works... yes
checking whether the C compiler (gcc  ) is a cross-compiler... no
checking whether we are using GNU C... yes
checking for c++... no
checking for g++... no
checking for gcc... gcc
checking whether the C++ compiler (gcc  ) works... yes
checking whether the C++ compiler (gcc  ) is a cross-compiler... no
checking whether we are using GNU C++... yes
checking for ranlib... ranlib
checking for cpp... /lib/cpp
checking for perl... perl
checking for f90... no
checking for pgf90... no
checking for xlf90... no
checking for ifort... no
checking for gfortran... no
checking for f77... no
checking for pgf77... no
checking for g77... no
checking for f90... no
checking for pgf90... no
checking for xlf90... no
checking for ifort... no
checking for gfortran... no
checking for ar... ar
checking for mkdir... mkdir
checking for bash... /bin/sh
Setting CACHELINE_BYTES to 32
Setting CACHE_SIZE to 2048*1024 bytes
checking how to run the C preprocessor... /lib/cpp
checking for ANSI C header files... yes
checking whether byte ordering is bigendian... no
checking size of long long... 8
checking size of long int... 4
checking size of int... 4
checking size of short int... 2
checking size of long double... 12
checking size of double... 8
checking size of float... 4
checking size of char *... 4
checking for the null device... /dev/null
checking for gethostbyname... yes
checking whether mode_t is defined... yes
checking for availability of gettimeofday timing... yes
checking if gettimeofday needs timezone... yes
checking for availability of getrusage timing... yes
checking for availability of _ftime timing... no
checking for time.h... yes
checking for sys/time.h... yes
checking for sys/types.h... yes
checking for unistd.h... yes
checking for string.h... yes
checking for assert.h... yes
checking for tgmath.h... yes
checking for sys/stat.h... yes
checking for getopt.h... yes
checking for dirent.h... yes
checking for c_asm.h... no
checking for intrinsics.h... no
checking for mach/mach_time.h... no
checking for regex.h... yes
checking for sys/filio.h... no
checking for sys/ioctl.h... yes
checking for sys/socket.h... yes
checking for netinet/in.h... yes
checking for netdb.h... yes
checking for arpa/inet.h... yes
checking for winsock2.h... no
checking for crypt.h... yes
checking for signal.h... yes
checking for malloc.h... yes
checking whether time.h and sys/time.h may both be included... yes
checking whether socklen_t is defined... yes
checking whether SOCKET is defined... no
checking for hrtime_t... no
checking for gethrtime... no
checking for read_real_time... no
checking for time_base_to_time... no
checking for clock_gettime... no
checking for mach_absolute_time... no
checking for _rtc intrinsic... no
checking for mallinfo... yes
checking for mallopt... yes
checking whether M_MMAP_THRESHOLD is defined... yes
checking if __int64 is defined... no
checking for vector... yes
checking for vector.h... yes
checking for getopt_long_only... yes
checking for working const... yes
checking for C inline... inline
checking for C restrict... restrict
checking for C++ restrict... __restrict__
checking for C bool... no
checking for CXX bool... yes
checking for Fortran REAL*4... no
checking for Fortran REAL*8... no
checking for Fortran REAL*16... no
checking for Fortran COMPLEX*8... no
checking for Fortran COMPLEX*16... no
checking for Fortran COMPLEX*32... no
checking for crypt... no
checking for crypt in library crypt... yes
checking for finite... yes
checking for isnan... yes
checking for isinf... yes
checking for mkstemp... yes
checking for va_copy... yes
creating cctk_Archdefs.h
creating make.arch.defn
creating cctk_Extradefs.h
creating make.extra.defn
cctk_Archdefs.h is unchanged
make.arch.defn is unchanged
cctk_Extradefs.h is unchanged
make.extra.defn is unchanged
updating cache ./config.cache
creating ./config.status
creating make.config.defn
creating make.config.deps
creating make.config.rules
creating cctk_Config.h
cctk_Config.h is unchanged
Determining number of fortran underscores...
No Fortran compiler - creating null fortran name conversion routine.
No Fortran compiler - creating null fortran common name conversion 
routine.
________________________________________________________________________
Use make WaveDemo to build the configuration.
________________________________________________________________________


From bentivegna at cct.lsu.edu  Sat Jan  9 08:02:43 2010
From: bentivegna at cct.lsu.edu (Eloisa Bentivegna)
Date: Sat, 09 Jan 2010 08:02:43 -0600
Subject: [Users] gxx_personality problem when compiling WaveDemo
In-Reply-To: <20100109114047.G12459@th.if.uj.edu.pl>
References: <20100109114047.G12459@th.if.uj.edu.pl>
Message-ID: <4B488C83.7030301@cct.lsu.edu>

Romuald Janik ha scritto:
> Dear All,
> 
> I have downladed Cactus (using GetCactus script) and the WaveDemo.
> make WaveDemo-config runs ok [I give the output at the end of this mail]
> but when I run
> make WaveDemo
> I get at the very end an error about undefined reference to 
> `__gxx_personality_v0'   (output below)
> 
> My system is Ubuntu 8.04, gcc is version 4.2.4, as far as I know make is
> a synonim of gmake on Ubuntu.
> I would be grateful for some advice. Am I missing some library/option 
> which is not set automatically by make WaveDemo-config?

Romuald,

my first guess would be that since no g++ is found on your system (see 
the configuration stage's stdout), C++ files are compiled with gcc, 
which can lead to the error you see at link time.

Try and make sure that all the compilers (and other basic tools) are 
found by either putting them under your path or using a Cactus 
configuration file (see for instance the tutorials at 
http://www.cactuscode.org/documentation/tutorials/).

Does this solve the problem?

Eloisa

From romuald at th.if.uj.edu.pl  Sat Jan  9 08:57:00 2010
From: romuald at th.if.uj.edu.pl (Romuald Janik)
Date: Sat, 9 Jan 2010 15:57:00 +0100 (CET)
Subject: [Users] gxx_personality problem when compiling WaveDemo
In-Reply-To: <4B488C83.7030301@cct.lsu.edu>
References: <20100109114047.G12459@th.if.uj.edu.pl>
	<4B488C83.7030301@cct.lsu.edu>
Message-ID: <20100109155158.F12459@th.if.uj.edu.pl>

Thanks!

g++ was present but its name had a version string attached in the ubuntu 
package. A symbolic link g++ fixed this issue!
Many thanks again..
Romuald

On Sat, 9 Jan 2010, Eloisa Bentivegna wrote:

> Romuald Janik ha scritto:
>> Dear All,
>>
>> I have downladed Cactus (using GetCactus script) and the WaveDemo.
>> make WaveDemo-config runs ok [I give the output at the end of this mail]
>> but when I run
>> make WaveDemo
>> I get at the very end an error about undefined reference to
>> `__gxx_personality_v0'   (output below)
>>
>> My system is Ubuntu 8.04, gcc is version 4.2.4, as far as I know make is
>> a synonim of gmake on Ubuntu.
>> I would be grateful for some advice. Am I missing some library/option
>> which is not set automatically by make WaveDemo-config?
>
> Romuald,
>
> my first guess would be that since no g++ is found on your system (see
> the configuration stage's stdout), C++ files are compiled with gcc,
> which can lead to the error you see at link time.
>
> Try and make sure that all the compilers (and other basic tools) are
> found by either putting them under your path or using a Cactus
> configuration file (see for instance the tutorials at
> http://www.cactuscode.org/documentation/tutorials/).
>
> Does this solve the problem?
>
> Eloisa
> _______________________________________________
> Users mailing list
> Users at cactuscode.org
> http://www.cactuscode.org/mailman/listinfo/users
>

From schnetter at cct.lsu.edu  Sat Jan  9 11:32:35 2010
From: schnetter at cct.lsu.edu (Erik Schnetter)
Date: Sat, 9 Jan 2010 11:32:35 -0600
Subject: [Users] gxx_personality problem when compiling WaveDemo
In-Reply-To: <20100109114047.G12459@th.if.uj.edu.pl>
References: <20100109114047.G12459@th.if.uj.edu.pl>
Message-ID: <E7A34CA0-80FA-4D6C-9E9A-9701417C4A6A@cct.lsu.edu>

On Jan 9, 2010, at 4:51 , Romuald Janik wrote:

> Dear All,
> 
> I have downladed Cactus (using GetCactus script) and the WaveDemo.
> make WaveDemo-config runs ok [I give the output at the end of this mail]
> but when I run
> make WaveDemo
> I get at the very end an error about undefined reference to 
> `__gxx_personality_v0'   (output below)
> 
> My system is Ubuntu 8.04, gcc is version 4.2.4, as far as I know make is
> a synonim of gmake on Ubuntu.
> I would be grateful for some advice. Am I missing some library/option 
> which is not set automatically by make WaveDemo-config?

Romuald

I don't know what could be wrong in your case.  Can you send the contents of the auto-generated file Cactus/configs/WaveDemo/config-data/make.config.defn?  It contains the configuration settings that Cactus is using.

-erik

-- 
Erik Schnetter <schnetter at cct.lsu.edu>   http://www.cct.lsu.edu/~eschnett/





From schnetter at cct.lsu.edu  Sat Jan  9 11:33:24 2010
From: schnetter at cct.lsu.edu (Erik Schnetter)
Date: Sat, 9 Jan 2010 11:33:24 -0600
Subject: [Users] gxx_personality problem when compiling WaveDemo
In-Reply-To: <E7A34CA0-80FA-4D6C-9E9A-9701417C4A6A@cct.lsu.edu>
References: <20100109114047.G12459@th.if.uj.edu.pl>
	<E7A34CA0-80FA-4D6C-9E9A-9701417C4A6A@cct.lsu.edu>
Message-ID: <2B5502BD-3A1B-464B-8C7A-A858C22E5EEA@cct.lsu.edu>

On Jan 9, 2010, at 11:32 , Erik Schnetter wrote:

> On Jan 9, 2010, at 4:51 , Romuald Janik wrote:
> 
>> Dear All,
>> 
>> I have downladed Cactus (using GetCactus script) and the WaveDemo.
>> make WaveDemo-config runs ok [I give the output at the end of this mail]
>> but when I run
>> make WaveDemo
>> I get at the very end an error about undefined reference to 
>> `__gxx_personality_v0'   (output below)
>> 
>> My system is Ubuntu 8.04, gcc is version 4.2.4, as far as I know make is
>> a synonim of gmake on Ubuntu.
>> I would be grateful for some advice. Am I missing some library/option 
>> which is not set automatically by make WaveDemo-config?
> 
> Romuald
> 
> I don't know what could be wrong in your case.  Can you send the contents of the auto-generated file Cactus/configs/WaveDemo/config-data/make.config.defn?  It contains the configuration settings that Cactus is using.


Forget this; Eloisa beat me to it...

-erik

-- 
Erik Schnetter <schnetter at cct.lsu.edu>   http://www.cct.lsu.edu/~eschnett/





From info at asfyon.nl  Tue Jan 26 04:52:06 2010
From: info at asfyon.nl (dr.r.j.slagter)
Date: Tue, 26 Jan 2010 11:52:06 +0100
Subject: [Users] parabolic PDE's
Message-ID: <45C65C749A3D46CB9BCC922EB98A8907@intel>

Dear all,

I should like to solve 12 coupled parabolic PDE's in 2D { i.e., (t,x,y) }.Is there some user who had experience with this problem?

Regards Reinoud ( info at asfyon.nl)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.cactuscode.org/pipermail/users/attachments/20100126/078a0676/attachment.html 

From ian.hinder at aei.mpg.de  Wed Jan 27 04:01:18 2010
From: ian.hinder at aei.mpg.de (Ian Hinder)
Date: Wed, 27 Jan 2010 11:01:18 +0100
Subject: [Users] Runtime statistics
Message-ID: <859B2268-C63C-4032-A5F4-11AA4087D5F7@aei.mpg.de>

Hi,

I find it very convenient to monitor at runtime the memory usage of a  
Cactus process.  For this, I have a thorn which sets a group of grid  
scalars from the information available in /proc/<pid>/statm.  This  
includes the process Resident Set Size which gives a good indication  
of the memory usage.  I then output this as an "info" variable on  
standard output, to give me a quick idea of the memory usage of a  
run.  This tells me if, for example, I am close to the physical memory  
limit of the machine and am likely to be swapping, and needing to use  
more processors.  This is very useful when setting up a new parameter  
file or changing the resolution of an existing one.  This information  
can also be output and stored with the simulation data.

I think that this behaviour should be included with Cactus so that all  
users can use it.

There are a couple of problems with my current approach:

1. It uses the /proc filesystem, which is not portable.  However, I  
expect that the vast majority of users are using Linux systems where  
this is available.  I propose that the functionality be disabled at  
compile time if /proc is unavailable.

2. It only tells me the memory usage on the root processor.  It would  
be nice to be able to use minimum and maximum reduction operators on  
grid scalars in the Info output (specified per-variable) but the  
reduction operators only seem to apply to grid functions.  One  
solution would be to do this by hand in the thorn.

It might be nice to have a thorn which monitors the current system  
status (e.g. like the memory usage discussed above), but also other  
useful things like the network utilisation, CPU utilisation, swap  
usage, whether there are other processes than the Cactus processes  
taking significant CPU.  All these things are currently possible on  
some clusters by sshing into nodes, but many clusters prohibit this  
and anyway this has to be caught when it happens and cannot be  
determined after the run has finished.

Where is the best place for this functionality?  The driver (e.g.  
Carpet)?  The flesh?  A thorn?

Comments?

-- 
Ian Hinder
ian.hinder at aei.mpg.de


From info at asfyon.nl  Wed Jan 27 05:42:03 2010
From: info at asfyon.nl (Dr. R. J. Slagter)
Date: Wed, 27 Jan 2010 12:42:03 +0100
Subject: [Users] wavedemo
Message-ID: <1264592523.12976.3.camel@localhost>

Hello,
I tried to install wavedemo, but I encounter some problems. I
execute ./GetCactus WaveDemo.th  and then the gmake commands as
described in the documentation. All went ok, but finally I get the
ouitpuit: see attachment

Hope someone knows the solution
Regards reinoud 


From info at asfyon.nl  Wed Jan 27 05:44:12 2010
From: info at asfyon.nl (Dr. R. J. Slagter)
Date: Wed, 27 Jan 2010 12:44:12 +0100
Subject: [Users] wavedemo
Message-ID: <1264592652.12976.5.camel@localhost>

Hello,
I tried to install wavedemo, but I encounter some problems. I
execute ./GetCactus WaveDemo.th  and then the gmake commands as
described in the documentation. All went ok, but finally I get the
ouitpuit: see attachment

Hope someone knows the solution
Regards reinoud 

-------------- next part --------------
A non-text attachment was scrubbed...
Name: wavedemo configure.odt
Type: application/vnd.oasis.opendocument.text
Size: 17887 bytes
Desc: not available
Url : http://www.cactuscode.org/pipermail/users/attachments/20100127/df8923d5/attachment-0001.bin 

From knarf at cct.lsu.edu  Wed Jan 27 08:03:05 2010
From: knarf at cct.lsu.edu (Frank Loeffler)
Date: Wed, 27 Jan 2010 08:03:05 -0600
Subject: [Users] wavedemo
In-Reply-To: <1264592652.12976.5.camel@localhost>
References: <1264592652.12976.5.camel@localhost>
Message-ID: <20100127140305.GX18541@numrel07.cct.lsu.edu>

Hi Reinoud,

On Wed, Jan 27, 2010 at 12:44:12PM +0100, Dr. R. J. Slagter wrote:
> I tried to install wavedemo, but I encounter some problems. I
> execute ./GetCactus WaveDemo.th  and then the gmake commands as
> described in the documentation. All went ok, but finally I get the
> ouitpuit: see attachment

Can you please send also the output from the configure stage? Something
went wrong there and without that output it is hard to tell what
exactly. That is the output generated by 'make NAME-config'.

If you know how, please send that as text file.

If you used a configuration file it would also be interesting to look at
that, so attach it as well please.

Frank

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: not available
Url : http://www.cactuscode.org/pipermail/users/attachments/20100127/029ab78a/attachment.bin 

From knarf at cct.lsu.edu  Wed Jan 27 08:09:29 2010
From: knarf at cct.lsu.edu (Frank Loeffler)
Date: Wed, 27 Jan 2010 08:09:29 -0600
Subject: [Users] Runtime statistics
In-Reply-To: <859B2268-C63C-4032-A5F4-11AA4087D5F7@aei.mpg.de>
References: <859B2268-C63C-4032-A5F4-11AA4087D5F7@aei.mpg.de>
Message-ID: <20100127140929.GY18541@numrel07.cct.lsu.edu>

Hi,

On Wed, Jan 27, 2010 at 11:01:18AM +0100, Ian Hinder wrote:
> I think that this behaviour should be included with Cactus so that all  
> users can use it.

I think this is a good idea.

> I propose that the functionality be disabled at  
> compile time if /proc is unavailable.

It should not be hard to disable this at runtime (testing if the
relevant files exist and are readable). This would prevent problems
where this information is not available on the headnode, but is on the
compute nodes or the other way around.

> 2. It only tells me the memory usage on the root processor.  It would  
> be nice to be able to use minimum and maximum reduction operators on  
> grid scalars in the Info output (specified per-variable) but the  
> reduction operators only seem to apply to grid functions.  One  
> solution would be to do this by hand in the thorn.

I agree.

> It might be nice to have a thorn which monitors the current system  
> status (e.g. like the memory usage discussed above), but also other  
> useful things like the network utilisation, CPU utilisation, swap  
> usage, whether there are other processes than the Cactus processes  
> taking significant CPU.

> Where is the best place for this functionality?  The driver (e.g.  
> Carpet)?  The flesh?  A thorn?

I believe that a thorn would be the best place for this. This is not
specific to a driver, and it does not need internal information from
the flesh. It might however, collect certain information from both.

Frank

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: not available
Url : http://www.cactuscode.org/pipermail/users/attachments/20100127/6cd8136a/attachment.bin 

From schnetter at cct.lsu.edu  Wed Jan 27 13:48:46 2010
From: schnetter at cct.lsu.edu (Erik Schnetter)
Date: Wed, 27 Jan 2010 14:48:46 -0500
Subject: [Users] Runtime statistics
In-Reply-To: <859B2268-C63C-4032-A5F4-11AA4087D5F7@aei.mpg.de>
References: <859B2268-C63C-4032-A5F4-11AA4087D5F7@aei.mpg.de>
Message-ID: <13ABBAE3-7A6F-4A59-AACF-5A781A109C5D@cct.lsu.edu>

On Jan 27, 2010, at 5:01 , Ian Hinder wrote:

> Hi,
> 
> I find it very convenient to monitor at runtime the memory usage of a  
> Cactus process.  For this, I have a thorn which sets a group of grid  
> scalars from the information available in /proc/<pid>/statm.  This  
> includes the process Resident Set Size which gives a good indication  
> of the memory usage.  I then output this as an "info" variable on  
> standard output, to give me a quick idea of the memory usage of a  
> run.  This tells me if, for example, I am close to the physical memory  
> limit of the machine and am likely to be swapping, and needing to use  
> more processors.  This is very useful when setting up a new parameter  
> file or changing the resolution of an existing one.  This information  
> can also be output and stored with the simulation data.
> 
> I think that this behaviour should be included with Cactus so that all  
> users can use it.
> 
> There are a couple of problems with my current approach:
> 
> 1. It uses the /proc filesystem, which is not portable.  However, I  
> expect that the vast majority of users are using Linux systems where  
> this is available.  I propose that the functionality be disabled at  
> compile time if /proc is unavailable.
> 
> 2. It only tells me the memory usage on the root processor.  It would  
> be nice to be able to use minimum and maximum reduction operators on  
> grid scalars in the Info output (specified per-variable) but the  
> reduction operators only seem to apply to grid functions.  One  
> solution would be to do this by hand in the thorn.

I would create a Cactus clock for this.  Clocks (used by timers) measure resource consumption such as time, cycles, cache misses, etc.

There is a TAU clock that measures memory usage, provided by Performance/TAUClocks.  However, it is currently difficult to build with TAU, and this works only on a few systems.  A more generic clock would be useful.

> It might be nice to have a thorn which monitors the current system  
> status (e.g. like the memory usage discussed above), but also other  
> useful things like the network utilisation, CPU utilisation, swap  
> usage, whether there are other processes than the Cactus processes  
> taking significant CPU.  All these things are currently possible on  
> some clusters by sshing into nodes, but many clusters prohibit this  
> and anyway this has to be caught when it happens and cannot be  
> determined after the run has finished.

This information would be nice to have on stdout, in a separate file in a format that is easier to read, and on the web pages.

Timers/clocks are convenient because they already have the functionality to reduce over processors, output to file, etc.

I would try to separate the collection of information from the processing (reduction over processors, averaging over time, calculating statistics, warning users, aborting the job) and presentation (stdout, file, etc.)

-erik

-- 
Erik Schnetter <schnetter at cct.lsu.edu>   http://www.cct.lsu.edu/~eschnett/





From dprideout at gmail.com  Wed Jan 27 16:25:35 2010
From: dprideout at gmail.com (David Rideout)
Date: Wed, 27 Jan 2010 16:25:35 -0600
Subject: [Users] funny cygpath problem with compiling Cactus on cygwin
Message-ID: <1ce81abb1001271425j60b02dc0m216547d5ee9b87a7@mail.gmail.com>

I just downloaded a fresh copy of cygwin and Cactus.  When compiling,
it repeatedly gives the following message.  It seems to call cygpath
with invalid arguments upon every change of directory or something.
Has anyone seen this before?  Any tips on how to fix?

Thanks,
David


Usage: cygpath (-d|-m|-u|-w|-t TYPE) [-f FILE] [OPTION]... NAME...
       cygpath [-c HANDLE]
       cygpath [-ADHOPSW]
       cygpath [-F ID]
Convert Unix and Windows format paths, or output system path information

Output type options:
  -d, --dos             print DOS (short) form of NAMEs (C:\PROGRA~1\)
  -m, --mixed           like --windows, but with regular slashes (C:/WINNT)
  -M, --mode            report on mode of file (binmode or textmode)
  -u, --unix            (default) print Unix form of NAMEs (/cygdrive/c/winnt)
  -w, --windows         print Windows form of NAMEs (C:\WINNT)
  -t, --type TYPE       print TYPE form: 'dos', 'mixed', 'unix', or 'windows'
Path conversion options:
  -a, --absolute        output absolute path
  -l, --long-name       print Windows long form of NAMEs (with -w, -m only)
  -p, --path            NAME is a PATH list (i.e., '/bin:/usr/bin')
  -s, --short-name      print DOS (short) form of NAMEs (with -w, -m only)
  -C, --codepage CP     print DOS, Windows, or mixed pathname in Windows
                        codepage CP.  CP can be a numeric codepage identifier,
                        or one of the reserved words ANSI, OEM, or UTF8.
                        If this option is missing, cygpath defaults to the
                        character set defined by the current locale.
System information:
  -A, --allusers        use `All Users' instead of current user for -D, -O, -P
  -D, --desktop         output `Desktop' directory and exit
  -H, --homeroot        output `Profiles' directory (home root) and exit
  -O, --mydocs          output `My Documents' directory and exit
  -P, --smprograms      output Start Menu `Programs' directory and exit
  -S, --sysdir          output system directory and exit
  -W, --windir          output `Windows' directory and exit
  -F, --folder ID       output special folder with numeric ID and exit
Try `cygpath --help' for more information.

From knarf at cct.lsu.edu  Wed Jan 27 17:06:19 2010
From: knarf at cct.lsu.edu (Frank Loeffler)
Date: Wed, 27 Jan 2010 17:06:19 -0600
Subject: [Users] funny cygpath problem with compiling Cactus on cygwin
In-Reply-To: <1ce81abb1001271425j60b02dc0m216547d5ee9b87a7@mail.gmail.com>
References: <1ce81abb1001271425j60b02dc0m216547d5ee9b87a7@mail.gmail.com>
Message-ID: <20100127230619.GA18541@numrel07.cct.lsu.edu>

On Wed, Jan 27, 2010 at 04:25:35PM -0600, David Rideout wrote:
> I just downloaded a fresh copy of cygwin and Cactus.  When compiling,
> it repeatedly gives the following message.  It seems to call cygpath
> with invalid arguments upon every change of directory or something.
> Has anyone seen this before?  Any tips on how to fix?

I don't and I don't have an easy way to try (no read-to-use windows).
What would be nice to know is the actual command Cactus tries to
execute. Do you know of a way to find that out and post it?

Frank

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: not available
Url : http://www.cactuscode.org/pipermail/users/attachments/20100127/461b77ce/attachment.bin 

From info at asfyon.nl  Wed Jan 27 14:51:57 2010
From: info at asfyon.nl (Dr. R. J. Slagter)
Date: Wed, 27 Jan 2010 21:51:57 +0100
Subject: [Users] wavedemo
Message-ID: <1264625517.4485.0.camel@localhost>

Hello Frank,
Thanks for your reply. I attached the config.log in the dir
configs/wavedemo
Regards Reinoud
-------------- next part --------------
A non-text attachment was scrubbed...
Name: config.log
Type: text/x-log
Size: 25809 bytes
Desc: not available
Url : http://www.cactuscode.org/pipermail/users/attachments/20100127/f324663a/attachment-0002.bin 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: config.status
Type: application/x-shellscript
Size: 26700 bytes
Desc: not available
Url : http://www.cactuscode.org/pipermail/users/attachments/20100127/f324663a/attachment-0003.bin 

From info at asfyon.nl  Thu Jan 28 10:53:31 2010
From: info at asfyon.nl (Dr. R. J. Slagter)
Date: Thu, 28 Jan 2010 17:53:31 +0100
Subject: [Users] wavedemo
Message-ID: <1264697611.7513.3.camel@localhost>

Hello Frank,
I send you config.log ones again after reinstalling the demo.
I guess I had to change Makefile and add gfortran?
Regards Reinoud
-------------- next part --------------
A non-text attachment was scrubbed...
Name: config.log
Type: text/x-log
Size: 25809 bytes
Desc: not available
Url : http://www.cactuscode.org/pipermail/users/attachments/20100128/3c52478c/attachment.bin 

From dprideout at gmail.com  Thu Jan 28 12:40:21 2010
From: dprideout at gmail.com (David Rideout)
Date: Thu, 28 Jan 2010 12:40:21 -0600
Subject: [Users] funny cygpath problem with compiling Cactus on cygwin
In-Reply-To: <20100127230619.GA18541@numrel07.cct.lsu.edu>
References: <1ce81abb1001271425j60b02dc0m216547d5ee9b87a7@mail.gmail.com> 
	<20100127230619.GA18541@numrel07.cct.lsu.edu>
Message-ID: <1ce81abb1001281040i4a049a17rda37c56b79de5181@mail.gmail.com>

The command seems to arise from

export TRANSFORM_DIRS = $(shell cygpath -wa $(1))

in make.config.defn, in which the argument is empty.  i.e.

cygpath -wa

On Wed, Jan 27, 2010 at 5:06 PM, Frank Loeffler <knarf at cct.lsu.edu> wrote:
> On Wed, Jan 27, 2010 at 04:25:35PM -0600, David Rideout wrote:
>> I just downloaded a fresh copy of cygwin and Cactus. ?When compiling,
>> it repeatedly gives the following message. ?It seems to call cygpath
>> with invalid arguments upon every change of directory or something.
>> Has anyone seen this before? ?Any tips on how to fix?
>
> I don't and I don't have an easy way to try (no read-to-use windows).
> What would be nice to know is the actual command Cactus tries to
> execute. Do you know of a way to find that out and post it?
>
> Frank
>
>
> _______________________________________________
> Users mailing list
> Users at cactuscode.org
> http://www.cactuscode.org/mailman/listinfo/users
>
>


From info at asfyon.nl  Thu Jan 28 16:12:33 2010
From: info at asfyon.nl (Dr. R. J. Slagter)
Date: Thu, 28 Jan 2010 23:12:33 +0100
Subject: [Users] WaveDemo runs now
Message-ID: <1264716753.18942.6.camel@localhost>

Hello Frank,

That was the trick: export F77=gfortran and then the gmake
I have now the executable cactus_testwave in the dir exe/ and it runs
with the WaveDemo.dat
Now I will inspect the output files  and try to visualize it in GNUPLOT.
Thanks for the help! I come back for the visualization and later (if you
have time) with my set of equations.
Reinoud



From schnetter at cct.lsu.edu  Thu Jan 28 19:52:47 2010
From: schnetter at cct.lsu.edu (Erik Schnetter)
Date: Thu, 28 Jan 2010 20:52:47 -0500
Subject: [Users] PDE's Fortran
In-Reply-To: <CAABE5AE9F5F4B7B99D0C975E1CA9A3A@intel>
References: <CAABE5AE9F5F4B7B99D0C975E1CA9A3A@intel>
Message-ID: <75A22D11-B689-4775-9380-253D4176D9D9@cct.lsu.edu>

On Jan 6, 2009, at 11:12 , dr.r.j.slagter wrote:

> Hello Erik, Thanks for your reply.I work on GR for some time now ( see http://www.asfyon.nl/webdoc1.htm )
> I used for my numerical calculations several routines (Cadsol,PDECOL RNPL) but I want to try for my latest
> rsearch ( see attachtment) a different routine in CACTUS. The equations are ( 11) like equation 20.
> The independent variables are t (time) r and y.
> Hope you can help me. Regards Reinoud

Reinoud

You mentioned in your earlier letter that you want to solve parabolic PDEs.  While parabolic and elliptic equations usually require an external solver, hyperbolic (wave-type) equations can be solved with much cheaper methods.  Cactus contains several time integrators for hyperbolic equations.  Equation (20) looks at first sight like a hyperbolic equation to me, since it contains both second spatial and temporal derivatives.

If you have not used Cactus before, then I suggest starting with a tutorial, which solves a single wave equation.  This would later be a good starting point to expand the system to the 11 equations you want to solve.  A recent tutorial is available at <http://www.cactuscode.org/documentation/tutorials/cactustutorialtg2009/>.

The standard Cactus WaveToy example uses a particular formulation of the wave equation which works nicely in flat space, but which is not suitable for more complex geometries or coordinate system.  You will probably need to rewrite your system into a form that is first order in time.  You can then use standard methods in Cactus to integrate your system in time.

If the 11 equations are complicated (they probably are), you should consider using an automated code generation system to evaluate them in C or Fortran.

-erik

-- 
Erik Schnetter <schnetter at cct.lsu.edu>   http://www.cct.lsu.edu/~eschnett/





From info at asfyon.nl  Fri Jan 29 02:42:46 2010
From: info at asfyon.nl (dr.r.j.slagter)
Date: Fri, 29 Jan 2010 09:42:46 +0100
Subject: [Users] PDE's Fortran
References: <CAABE5AE9F5F4B7B99D0C975E1CA9A3A@intel>
	<75A22D11-B689-4775-9380-253D4176D9D9@cct.lsu.edu>
Message-ID: <0C64BF7314F848939453BCED4B93D927@intel>

Hello Erik,
Thanks for your reply. With the help of Frank I managed to run WaveDemo.th
I get the output and will try to visualize the files with gnuplot.
My (indeed hyperbolic) 12 equations I attached( equations 9-14). I can write 
them in first order form together with the boundary values. I tried to solve 
them in some other routines, but not succesful.
Now I want to solve these equations in Cactus. Hopefully you and Frank can 
help me.
I visited the site you mentioned. What is the automated code generation 
system ?
Regards Reinoud


----- Original Message ----- 
From: "Erik Schnetter" <schnetter at cct.lsu.edu>
To: "Cactus users mailing list" <users at cactuscode.org>
Sent: Friday, January 29, 2010 2:52 AM
Subject: Re: [Users] PDE's Fortran


> On Jan 6, 2009, at 11:12 , dr.r.j.slagter wrote:
>
>> Hello Erik, Thanks for your reply.I work on GR for some time now ( see 
>> http://www.asfyon.nl/webdoc1.htm )
>> I used for my numerical calculations several routines (Cadsol,PDECOL 
>> RNPL) but I want to try for my latest
>> rsearch ( see attachtment) a different routine in CACTUS. The equations 
>> are ( 11) like equation 20.
>> The independent variables are t (time) r and y.
>> Hope you can help me. Regards Reinoud
>
> Reinoud
>
> You mentioned in your earlier letter that you want to solve parabolic 
> PDEs.  While parabolic and elliptic equations usually require an external 
> solver, hyperbolic (wave-type) equations can be solved with much cheaper 
> methods.  Cactus contains several time integrators for hyperbolic 
> equations.  Equation (20) looks at first sight like a hyperbolic equation 
> to me, since it contains both second spatial and temporal derivatives.
>
> If you have not used Cactus before, then I suggest starting with a 
> tutorial, which solves a single wave equation.  This would later be a good 
> starting point to expand the system to the 11 equations you want to solve. 
> A recent tutorial is available at 
> <http://www.cactuscode.org/documentation/tutorials/cactustutorialtg2009/>.
>
> The standard Cactus WaveToy example uses a particular formulation of the 
> wave equation which works nicely in flat space, but which is not suitable 
> for more complex geometries or coordinate system.  You will probably need 
> to rewrite your system into a form that is first order in time.  You can 
> then use standard methods in Cactus to integrate your system in time.
>
> If the 11 equations are complicated (they probably are), you should 
> consider using an automated code generation system to evaluate them in C 
> or Fortran.
>
> -erik
>
> -- 
> Erik Schnetter <schnetter at cct.lsu.edu>   http://www.cct.lsu.edu/~eschnett/
>
>
>
>
> _______________________________________________
> Users mailing list
> Users at cactuscode.org
> http://www.cactuscode.org/mailman/listinfo/users
>


--------------------------------------------------------------------------------



No virus found in this incoming message.
Checked by AVG - www.avg.com
Version: 8.5.432 / Virus Database: 271.1.1/2653 - Release Date: 01/28/10 
14:55:00
-------------- next part --------------
A non-text attachment was scrubbed...
Name: warp cosm string1.tex
Type: application/octet-stream
Size: 8502 bytes
Desc: not available
Url : http://www.cactuscode.org/pipermail/users/attachments/20100129/77d4cf20/attachment.obj