[Camino-users] Possibility of parallelization for the datasynth command.

[Hall, Matt]

Hi,

The simplest thing to do is to run the single voxel command on several different nodes with different random seeds. You can submit each job separately to your cluster’s scheduling system and process the results when they’re finished. Otherwise you’ll just end up sequentially running 24 voxels on a single node.

Cheers,
Matt.

From: camino-users-bounces at www.nitrc.org [mailto:camino-users-bounces at www.nitrc.org] On Behalf Of Rajalekshmi Deepu
Sent: 07 June 2018 14:00
To: Cook, Philip <cookpa at pennmedicine.upenn.edu>
Cc: camino-users at www.nitrc.org
Subject: Re: [Camino-users] Possibility of parallelization for the datasynth command.


Hi,

We have a cluster with each compute node having 24 physical and 24 logical cores. If I wanted to run the below command, say for 24 voxels instead of 1, can I distribute it to the 24 cores? Probably with -separateruns option. Will this generate different data as we use MC?

 datasynth -walkers 100000 -tmax 10000 -p 0.0 -voxels 1 -initial spike -voxelsizefrac 1 -diffusivity 0.6E-9 -schemefile myScheme.scheme1 -substrate cylinder -packing square -cylinderrad 3E-6 -cylindersep 6.1E-6 -outputfile regCyl_r=3E-6_spike.Bfloat

Also I am not clear with the option -seed. Here<http://camino.cs.ucl.ac.uk/index.php?n=Tutorials.MCSimulator> it says, "-seed specifies the seed for the random number generators to use. "


On the other hand, here<http://web4.cs.ucl.ac.uk/research/medic/camino/pmwiki/pmwiki.php?n=Man.Datasynth> it says "Specifies the random seed to use for noise generation in simulation trials."

Bit confused now.

Any help will be appreciated.

Thanks and regards,

Lekshmi

On 29.05.18 16:31, Rajalekshmi Deepu wrote:
Hi,

Please find below the command:

datasynth -walkers 100000 -tmax 1000 -p 0.0 -voxels 1 -schemefile 59.scheme  -initial uniform

Best,

Lekshmi


On 29.05.18 16:19, Cook, Philip wrote:

Hi,

Can you give an example of the command you want to parallelize?



On May 29, 2018, at 9:20 AM, Rajalekshmi Deepu <r.deepu at fz-juelich.de><mailto:r.deepu at fz-juelich.de> wrote:

Dear all,

  I would like to know if there is a possibility of parallelization for the datasynth command?

Couldn't find any information about it.

Any help would be appreciated.

Thanks and regards,

Lekshmi


_______________________________________________
Camino-users mailing list
Camino-users at www.nitrc.org<mailto:Camino-users at www.nitrc.org>
https://www.nitrc.org/mailman/listinfo/camino-users


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.nitrc.org/pipermail/camino-users/attachments/20180607/c8ce7127/attachment-0001.html>