[Gate-users] Problem with 10^10 primaries send

[Sarah Blind]

I think it's working ! 
I just have some process that stop, not all of them go to the end. Maybe it's because of the number of cores the virtual machine has, I think a dual boot would be better :) 

Many thanks, you have been of great help! 
Regards, 
Sarah 

----- Le 8 Mar 22, à 14:26, Dimitrios Thanasas <dthanas at uoa.gr> a écrit : 

> Dear Sarah,

> Yes you are right, using "auto" a new seed is generated every time you run
> a different simulation. No you do not have to do something else.
> You can confirm that you each run use a different cpu with the command
> "top" when you run your simulations.
> You do not have to install job splitter (I tried some months ago and I had
> problems during the installation). I am not sure, but I think that using
> job splitter you get one output file (based on the job splits) and without
> it you get more output files and you have to find how to use them.

> Regards,
> Dimitris

> --
> --------------------------
> Dimitrios Thanasas, PhD
> Medical Physicist
> Medical Physics Lab
> Medical School
> National and Kapodistrian University of Athens
> --------------------------

> > Dear Dimitris,

> > Thank you so much for your response. I will try this right now!

> > What do you meen exactly by changing the engine seed ?
> > I have set

> >> /gate/random/setEngineName MersenneTwister
> >> /gate/random/setEngineSeed auto

> > I think a new seed is automaticallu generated that way?

> > I want to run my simulation for a specific number of primaries but not
> > especially for a specific time.

> > Do I have to set something in order for Gate to use a different CPU for
> > each simulation?

> > I use a DoseActor to produce a .hdr file, not a root file output.

> > Don't I need to install job splitter?

> > Thank you for your help,
> > Sarah

> > ----- Le 8 Mar 22, à 13:15, Dimitrios Thanasas <dthanas at uoa.gr> a écrit :

> >> Dear Sarah,

> >> The general idea is to create different folders having the same mac
> >> files
> >> but changing the "Engine seed" and the time start and time stop also (if
> >> you run the simulation for specific time and not for specific number of
> >> primaries). Then, you open a different terminal in each folder and run
> >> the
> >> corresponding mac file. The number of the folders depends on the numbers
> >> of CPUs that your machine has (each simulation run on a single CPU).
> >> In order not to mix up the output files you can set diffrent name of
> >> each
> >> output file (eg outputname_1, outputname_2,....).
> >> Then you can combine the results. Do you use a DoseActor to record the
> >> dose (or energy deposited) in a output file or do you produce a root
> >> file
> >> as output?
> >> In case you use a dose actor it is easy to sum up the dose of each run
> >> but
> >> it is not so easy to calculate the overall uncertainty. If your output
> >> is
> >> root files, you can merge them or load altogether in a root browser (I
> >> do
> >> not use root often so I cannot send you how to do this).

> >> I hope this will help you,
> >> Dimitris

> >> --
> >> --------------------------
> >> Dimitrios Thanasas, PhD
> >> Medical Physicist
> >> Medical Physics Lab
> >> Medical School
> >> National and Kapodistrian University of Athens
> >> --------------------------

> >> > Dear Dimitris Thanasas,

> >> > I come back to you about launching GATE with more than 10^9 primaries.
> >> > I have tried to set an activity to my source and to run the simulation
> >> for
> >> > a certain time as you suggested.
> >> > It allowed me to launch my simulation with 10^10 photons, but not
> >> more.

> >> > Now I want to split my simulation, but I don't now how exactly and I
> >> can't
> >> > find any information about it, would you give me some advices ?

> >> > Thank you very much for your help,
> >> > Regards,
> >> > Sarah Blind

> >> > ----- Le 25 Fév 22, à 7:59, Dimitrios Thanasas <dthanas at uoa.gr> a
> >> écrit :

> >> >> Dear Sarah,

> >> >> You could try to give an activity to your source, (e.g. 1 GBq -
> >> without
> >> >> half life) and then to run the simulation for a certain time (e.g.
> >> 10s).
> >> >> I
> >> >> think that this wiil work (you may need a statactor to know exactly
> >> the
> >> >> number of the emitted particles).
> >> >> If you try to split the simulation, you may need to change the engine
> >> >> seed
> >> >> in order not to have the same results in each simulation.

> >> >> Regards,
> >> >> Dimitris

> >> >> > Dear Gate users,

> >> >> > I have the following model : 1 cell inside of which there are 12
> >> >> clusters
> >> >> > of 8000 nanoparticles each.
> >> >> > My source is composed of a gamma beam of 5 keV.
> >> >> > My simulation works fine up to 10^9 photons send. But with 10^10
> >> >> primaries
> >> >> > nothing happens anymore (there is no error message).
> >> >> > Does someone have an idea what the problem might be ?

> >> >> > Thank you very much for your help,
> >> >> > Regards,
> >> >> > Sarah Blind

> >> >> > world.mac




> >> >> > /gate/world/geometry/setXLength 30 um
> >> >> > /gate/world/geometry/setYLength 30 um
> >> >> > /gate/world/geometry/setZLength 30 um
> >> >> > /gate/world/setMaterial Air

> >> >> > /gate/world/daughters/name cell
> >> >> > /gate/world/daughters/insert sphere
> >> >> > /gate/cell/geometry/setRmin 0.0 mm
> >> >> > /gate/cell/geometry/setRmax 10 um
> >> >> > /gate/cell/placement/setTranslation 0 0 0 um
> >> >> > /gate/cell/setMaterial G4_WATER
> >> >> > /gate/cell/vis/setVisible 1
> >> >> > /gate/cell/vis/setColor blue
> >> >> > /gate/cell/vis/forceSolid false

> >> >> > /gate/cell/daughters/name cluster
> >> >> > /gate/cell/daughters/insert sphere
> >> >> > /gate/cluster/geometry/setRmin 0.0 mm
> >> >> > /gate/cluster/geometry/setRmax 500 nm
> >> >> > /gate/cluster/placement/setTranslation 0 0 0 um
> >> >> > /gate/cluster/setMaterial G4_WATER
> >> >> > /gate/cluster/vis/setVisible 1
> >> >> > /gate/cluster/vis/setColor yellow
> >> >> > /gate/cluster/vis/forceSolid false

> >> >> > /gate/cluster/daughters/name NP
> >> >> > /gate/cluster/daughters/insert sphere
> >> >> > /gate/NP/geometry/setRmin 0 mm
> >> >> > /gate/NP/geometry/setRmax 2 nm
> >> >> > /gate/NP/placement/setTranslation 0 0 0 mm
> >> >> > /gate/NP/setMaterial Bismuth
> >> >> > /gate/NP/vis/setVisible 1
> >> >> > /gate/NP/vis/setColor red
> >> >> > /gate/NP/vis/forceSolid true


> >> >> > /gate/NP/repeaters/insert cubicArray
> >> >> > /gate/NP/cubicArray/setRepeatNumberX 20
> >> >> > /gate/NP/cubicArray/setRepeatNumberY 20
> >> >> > /gate/NP/cubicArray/setRepeatNumberZ 20
> >> >> > /gate/NP/cubicArray/setRepeatVector 5 5 5 nm

> >> >> > /gate/cluster/repeaters/insert cubicArray
> >> >> > /gate/cluster/cubicArray/setRepeatNumberX 3
> >> >> > /gate/cluster/cubicArray/setRepeatNumberY 2
> >> >> > /gate/cluster/cubicArray/setRepeatNumberZ 2
> >> >> > /gate/cluster/cubicArray/setRepeatVector 5 5 5 um




> >> >> > physics.mac

> >> >> > BQ_BEGIN


> >> >> > /gate/physics/addPhysicsList QGSP_BERT_EMV

> >> >> > /gate/physics/Gamma/SetCutInRegion world 50 nm
> >> >> > /gate/physics/Electron/SetCutInRegion world 50 nm

> >> >> > /gate/physics/Gamma/SetCutInRegion cell 20 nm
> >> >> > /gate/physics/Electron/SetCutInRegion cell 20 nm

> >> >> > /gate/physics/Gamma/SetCutInRegion cluster 5 nm
> >> >> > /gate/physics/Electron/SetCutInRegion cluster 5 nm

> >> >> > /gate/physics/Gamma/SetCutInRegion NP 0.5 nm
> >> >> > /gate/physics/Electron/SetCutInRegion NP 0.5 nm

> >> >> > /gate/random/setEngineName MersenneTwister

> >> >> > /gate/physics/addAtomDeexcitation

> >> >> > /gate/physics/addProcess PhotoElectric gamma
> >> >> > /gate/physics/processes/PhotoElectric/setModel LivermoreModel
> >> >> > /gate/physics/processes/PhotoElectric/setAugerElectron true
> >> >> > /gate/physics/processes/PhotoElectric/setDeltaRayCut 10 eV
> >> >> > /gate/physics/processes/PhotoElectric/setXRayCut 10 eV
> >> >> > /gate/physics/processes/PhotoElectric/setStepFunction e- 0.2 0.1 nm

> >> >> > /gate/physics/addProcess Compton gamma
> >> >> > /gate/physics/processes/Compton/setModel LivermoreModel
> >> >> > /gate/physics/processes/Compton/setStepFunction e- 0.2 0.1 nm

> >> >> > /gate/physics/addProcess RayleighScattering
> >> >> > /gate/physics/processes/RayleighScattering/setModel LivermoreModel

> >> >> > /gate/physics/addProcess GammaConversion
> >> >> > /gate/physics/processes/GammaConversion/setModel LivermoreModel

> >> >> > /gate/physics/addProcess ElectronIonisation
> >> >> > /gate/physics/processes/ElectronIonisation/setModel LivermoreModel
> >> e-
> >> >> > /gate/physics/processes/ElectronIonisation/setModel PenelopeModel
> >> e+
> >> >> > /gate/physics/processes/ElectronIonisation/setStepFunction e- 0.2
> >> 0.1
> >> >> nm

> >> >> > /gate/physics/addProcess Bremsstrahlung e-
> >> >> > /gate/physics/addProcess Bremsstrahlung e+
> >> >> > /gate/physics/processes/Bremsstrahlung/setModel LivermoreModel e-
> >> >> > /gate/physics/processes/Bremsstrahlung/setModel PenelopeModel e+

> >> >> > /gate/physics/addProcess eMultipleScattering e-
> >> >> > /gate/physics/addProcess eMultipleScattering e+
> >> >> > /gate/physics/processes/eMultipleScattering/setModel UrbanModel e-
> >> >> > /gate/physics/processes/eMultipleScattering/setModel UrbanModel e+

> >> >> > /gate/physics/SetMinKineticEnergyInRegion world 1 eV
> >> >> > /gate/physics/SetMinKineticEnergyInRegion cell 1 eV
> >> >> > /gate/physics/ActivateSpecialCuts gamma

> >> >> > BQ_END


> >> >> > source.mac

> >> >> > BQ_BEGIN


> >> >> > /gate/source/addSource mybeam gps
> >> >> > /gate/source/mybeam/gps/particle gamma
> >> >> > /gate/source/mybeam/gps/pos/type Beam
> >> >> > /gate/source/mybeam/gps/ene/mono 5 keV
> >> >> > /gate/source/mybeam/gps/ene/type Gauss
> >> >> > /gate/source/mybeam/gps/ene/sigma 1.0 keV
> >> >> > /gate/source/mybeam/gps/direction 0 0 -1

> >> >> > /gate/source/mybeam/gps/position 0 0 15 um
> >> >> > /gate/source/mybeam/gps/pos/type Volume
> >> >> > /gate/source/mybeam/gps/pos/shape Cylinder
> >> >> > /gate/source/mybeam/gps/pos/inner_radius 0 mm
> >> >> > /gate/source/mybeam/gps/pos/radius 10 um
> >> >> > /gate/source/mybeam/gps/pos/halfz 0.01 um
> >> >> > /gate/source/mybeam/gps/ang/type beam2d





> >> >> > BQ_END



> >> >> > _______________________________________________
> >> >> > Gate-users mailing list
> >> >> > Gate-users at lists.opengatecollaboration.org
> >> >> > http://lists.opengatecollaboration.org/mailman/listinfo/gate-users

> >> > --
> >> > Sarah Blind
> >> > Ingénieure de recherche - CNRS
> >> > Centre de Recherche en Automatique de Nancy (CRAN)
> >> > Département Biologie, Signaux et Systèmes en Cancérologie et
> >> Neurosciences
> >> > (BioSiS)
> >> > Site de la faculté de médecine
> >> > 9, Avenue de la Forêt de Haye
> >> > 54505 Vandoeuvre les Nancy
> >> > Tél : 03.72.74.61.13


> > --
> > Sarah Blind
> > Ingénieure de recherche - CNRS
> > Centre de Recherche en Automatique de Nancy (CRAN)
> > Département Biologie, Signaux et Systèmes en Cancérologie et Neurosciences
> > (BioSiS)
> > Site de la faculté de médecine
> > 9, Avenue de la Forêt de Haye
> > 54505 Vandoeuvre les Nancy
> > Tél : 03.72.74.61.13


-- 
Sarah Blind 
Ingénieure de recherche - CNRS 
Centre de Recherche en Automatique de Nancy (CRAN) 
Département Biologie, Signaux et Systèmes en Cancérologie et Neurosciences (BioSiS) 
Site de la faculté de médecine 
9, Avenue de la Forêt de Haye 
54505 Vandoeuvre les Nancy 
Tél : 03.72.74.61.13 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opengatecollaboration.org/pipermail/gate-users/attachments/20220308/4a3a7ba5/attachment-0001.html>