RE: Trouble running the same simulation with gas particles

From: Phil Sutton <P.J.Sutton_at_lboro.ac.uk>
Date: Wed, 21 Jan 2015 09:12:41 +0000

Thanks for everyone's help, I have gone through again checking everything that was mentioned. I have made sure that the temperature of my particles are set to a non-zero value along with a minimum value for their temperature. I assume that the latter is done only through the parameterfile? In my IC I set the U array to have N-1 particles because of the central mass and give them a non-zero value. Also in the parameterfile I do the same, therefore it should at least be read somewhere. Before stopping it does write the first snapshot it appears to have read in these temperatures correctly.

With regards to the MASSARR I have instead set them in the MASS array for each particle. I have always gone about it this (probably not the best). However, if I now tried to set it in the MASSARR for particle types it will come back as incorrect blocks sizes. I assume to set this it's as simple as just setting: MASSARR=[1000D, 0, 0, 0, 0, 0], for example?

The example test IC's (Gas sphere) run fine with no errors and the my parameterfile matches that almost as the example cases.

Additionally I compiled Gadget2 to switch off gravity (NOGRAVITY) which resulted in the same error of zero time steps being assigned, hence it removes this as a potential problem.

When writing IC for gas particles do I need any other arrays except for the following, as this is what I am writing to the IC file at the end:


    writeu,1,npart,massarr,time,redshift,flag_sfr,flag_feedback,npartTotal,la
    writeu,1,pos
    writeu,1,vel
    writeu,1,id
    writeu,1,mass
    writeu,1,u


Thanks for the help.

Phil



-----Original Message-----
From: Volker Springel [mailto:volker_at_MPA-Garching.MPG.DE]
Sent: 20 January 2015 13:17
To: Gadget General Discussion
Subject: Re: [gadget-list] Trouble running the same simulation with gas particles



Hi Phil,

You would get an error of this kind if you set the temperature of your particles to zero and don't impose a temperature floor. Then the sound-speed ends up being zero, causing a division by zero when the Courant timestep criterion is evaluated.

Volker

On 20 Jan 2015, at 13:26, Arman Khalatyan <arm2arm_at_gmail.com> wrote:

>
> Hi,
> If you have all particles with the same mass why the output is:
>> Type 1 (halo): 32767 (tot= 0000032767) masstab=0
> Are there MASS block per particle type? If not you should correct your header.
>
>
> ***********************************************************
>
> Dr. Arman Khalatyan eScience -SuperComputing
> Leibniz-Institut für Astrophysik Potsdam (AIP) An der Sternwarte 16,
> 14482 Potsdam, Germany
>
> ***********************************************************
>
>
> On Tue, Jan 20, 2015 at 1:16 PM, Matthias Kühtreiber
> <a0749244_at_unet.univie.ac.at> wrote:
>>
>> Hi Phil,
>>
>> I assume that all physical units set in the parameter file fit the
>> units in your initial conditions. You could also try if you can get
>> the gas sphere ICs to run which are delivered with Gadget-2. There is
>> also a parameter file included which could be helpful for comparisons.
>>
>> Otherwise you could try to track down the error in timestep.c. The
>> value for dt seems to get "nan". There could be for instance a
>> division by 0 somewhere in the calculation of dt. The endrun(818) is
>> in function "get_timestep". I hope this helps a little.
>>
>> Best wishes,
>> Matthias
>>
>> On Di, 20.01.2015, 12:26, Phil Sutton wrote:
>>>
>>> Hi Matthias,
>>>
>>> Unfortunately I have this set to a rather large non-zero value.
>>> Would there be anything else in the parameter file that might throw
>>> up this error? Although I have tried to not have any zero values. I
>>> can only assume it something to do with hydrodynamical element of
>>> the time step performed by Gadget?
>>>
>>> Thanks
>>>
>>> Phil
>>>
>>> -----Original Message-----
>>> From: Matthias Kühtreiber [mailto:a0749244_at_unet.univie.ac.at]
>>> Sent: 20 January 2015 10:52
>>> To: Gadget General Discussion
>>> Subject: Re: [gadget-list] Trouble running the same simulation with
>>> gas particles
>>>
>>>
>>> Hi Phil,
>>>
>>> Have you double-checked that the value of "SofteningGas" in your
>>> parameter file is set to a non-zero value?
>>> A gravitational softening of 0 can lead to similar errors you
>>> described above.
>>>
>>> Best wishes,
>>>
>>> Matthias Kühtreiber
>>> Department of Astrophysics
>>> University of Vienna
>>>
>>>
>>>
>>> On Di, 20.01.2015, 10:58, Phil Sutton wrote:
>>>> Hi all,
>>>>
>>>> For a while now I have been running a few simulations successfully
>>>> using non-gas particles. However, I now wish to include some gas
>>>> particles and have run into some problems getting Gadget2 to
>>>> initial run. I now have two initial conditions that creates a ring
>>>> of particles around a larger central mass. The velocities all
>>>> assume circular Keplerian motions with the following:
>>>> [cid:image003.jpg_at_01D03497.9310C440]
>>>>
>>>> All particle number densities, velocities and masses are the same
>>>> (I added in a new array for internal energies of gas particles).
>>>> The only difference is the particle type. The non-gas IC runs fine
>>>> with no errors and continues to evolve in a Keplerian way. However,
>>>> when gas particles are used for the ring a zero time step is
>>>> assigned. I have tried altering the smoothing lengths, mass, number
>>>> of particles and scale of the system from ridiculously small and large with no effect.
>>>> The only thing that allows Gadget2 to run is if I reduce the
>>>> central mass down by a factor of 10, which obviously results in the
>>>> ring particles no longer having Keplerian motions about the centre.
>>>> I have probably made a simple error in setting up of the initial
>>>> conditions or an assumptions I’ve made when using gas particles but
>>>> am running out of ideas to try. Any help would be great.
>>>>
>>>> Here is the errors for the gas particle IC followed by the
>>>> successful non-gas IC if it helps:
>>>>
>>>>
>>>> This is Gadget, version `2.0'.
>>>>
>>>>
>>>>
>>>> Running on 12 processors.
>>>>
>>>>
>>>>
>>>> Allocated 30 MByte communication buffer per processor.
>>>>
>>>>
>>>>
>>>> Communication buffer has room for 604946 particles in gravity
>>>> computation
>>>>
>>>> Communication buffer has room for 245760 particles in density
>>>> computation
>>>>
>>>> Communication buffer has room for 196608 particles in hydro
>>>> computation
>>>>
>>>> Communication buffer has room for 196608 particles in domain
>>>> decomposition
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> Hubble (internal units) = 3.24078e-18
>>>>
>>>> G (internal units) = 6.672e-08
>>>>
>>>> UnitMass_in_g = 1
>>>>
>>>> UnitTime_in_s = 1
>>>>
>>>> UnitVelocity_in_cm_per_s = 1
>>>>
>>>> UnitDensity_in_cgs = 1
>>>>
>>>> UnitEnergy_in_cgs = 1
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> Allocated 0.567307 MByte for particle storage. 68
>>>>
>>>>
>>>>
>>>> Allocated 0.70079 MByte for storage of SPH data. 84
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> reading file `gas.dat' on task=0 (contains 35000 particles.)
>>>>
>>>> distributing this file to tasks 0-11
>>>>
>>>> Type 0 (gas): 34999 (tot= 0000034999) masstab=0
>>>>
>>>> Type 1 (halo): 0 (tot= 0000000000) masstab=0
>>>>
>>>> Type 2 (disk): 1 (tot= 0000000001) masstab=0
>>>>
>>>> Type 3 (bulge): 0 (tot= 0000000000) masstab=0
>>>>
>>>> Type 4 (stars): 0 (tot= 0000000000) masstab=0
>>>>
>>>> Type 5 (bndry): 0 (tot= 0000000000) masstab=0
>>>>
>>>>
>>>>
>>>> reading done.
>>>>
>>>> Total number of particles : 0000035000
>>>>
>>>>
>>>>
>>>> allocated 0.0762939 Mbyte for ngb search.
>>>>
>>>>
>>>>
>>>> Allocated 1.63065 MByte for BH-tree. 64
>>>>
>>>>
>>>>
>>>> domain decomposition...
>>>>
>>>> NTopleaves= 1436
>>>>
>>>> work-load balance=1.01143 memory-balance=1.01143
>>>>
>>>> exchange of 0000032087 particles
>>>>
>>>> domain decomposition done.
>>>>
>>>> begin Peano-Hilbert order...
>>>>
>>>> Peano-Hilbert done.
>>>>
>>>> Begin Ngb-tree construction.
>>>>
>>>> Ngb-Tree contruction finished
>>>>
>>>> ngb iteration 1: need to repeat for 0000034999 particles.
>>>>
>>>> ngb iteration 2: need to repeat for 0000034999 particles.
>>>>
>>>> ngb iteration 3: need to repeat for 0000034967 particles.
>>>>
>>>> ngb iteration 4: need to repeat for 0000034163 particles.
>>>>
>>>> ngb iteration 5: need to repeat for 0000027944 particles.
>>>>
>>>> ngb iteration 6: need to repeat for 0000014669 particles.
>>>>
>>>> ngb iteration 7: need to repeat for 0000004986 particles.
>>>>
>>>> ngb iteration 8: need to repeat for 0000001284 particles.
>>>>
>>>> ngb iteration 9: need to repeat for 0000000446 particles.
>>>>
>>>> ngb iteration 10: need to repeat for 0000000211 particles.
>>>>
>>>> ngb iteration 11: need to repeat for 0000000088 particles.
>>>>
>>>> ngb iteration 12: need to repeat for 0000000026 particles.
>>>>
>>>> ngb iteration 13: need to repeat for 0000000005 particles.
>>>>
>>>>
>>>>
>>>> Setting next time for snapshot file to Time_next= 0
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> writing snapshot file...
>>>>
>>>> done with snapshot.
>>>>
>>>>
>>>>
>>>> Setting next time for snapshot file to Time_next= 500
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> Begin Step 0, Time: 0, Systemstep: 0
>>>>
>>>> domain decomposition...
>>>>
>>>> NTopleaves= 1436
>>>>
>>>> work-load balance=1.01143 memory-balance=1.01143
>>>>
>>>> domain decomposition done.
>>>>
>>>> begin Peano-Hilbert order...
>>>>
>>>> Peano-Hilbert done.
>>>>
>>>> Start force computation...
>>>>
>>>> Tree construction.
>>>>
>>>> Tree construction done.
>>>>
>>>> Begin tree force.
>>>>
>>>> tree is done.
>>>>
>>>> Begin tree force.
>>>>
>>>> tree is done.
>>>>
>>>> Start density computation...
>>>>
>>>> Start hydro-force computation...
>>>>
>>>> force computation done.
>>>>
>>>>
>>>>
>>>> Error: A timestep of size zero was assigned on the integer timeline!
>>>>
>>>> We better stop.
>>>>
>>>> Task=0 Part-ID=0 dt=-nan tibase=3.72529e-05 ti_step=-2147483648
>>>> ac=-nan
>>>> xyz=(2.05361e+11|-1.32362e+09|0) tree=(-0.999847|0.006444340)
>>>>
>>>>
>>>>
>>>> hydro-frc=(-nan|-nan|-nan)
>>>>
>>>> task 0: endrun called with an error level of 818
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> Error: A timestep of size zero was assigned on the integer timeline!
>>>>
>>>> We better stop.
>>>>
>>>> Task=1 Part-ID=0 dt=-nan tibase=3.72529e-05 ti_step=-2147483648
>>>> ac=-nan
>>>> xyz=(1.40842e+11|-1.21532e+11|0) tree=(-0.922557|0.796070)
>>>>
>>>>
>>>>
>>>> hydro-frc=(-nan|-nan|-nan)
>>>> task 1: endrun called with an error level of 818
>>>>
>>>>
>>>>
>>>>
>>>> This is Gadget, version `2.0'.
>>>>
>>>>
>>>>
>>>> Running on 12 processors.
>>>>
>>>>
>>>>
>>>> Allocated 30 MByte communication buffer per processor.
>>>>
>>>>
>>>>
>>>> Communication buffer has room for 604946 particles in gravity
>>>> computation
>>>>
>>>> Communication buffer has room for 245760 particles in density
>>>> computation
>>>>
>>>> Communication buffer has room for 196608 particles in hydro
>>>> computation
>>>>
>>>> Communication buffer has room for 196608 particles in domain
>>>> decomposition
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> Hubble (internal units) = 3.24078e-18
>>>>
>>>> G (internal units) = 6.672e-08
>>>>
>>>> UnitMass_in_g = 1
>>>>
>>>> UnitTime_in_s = 1
>>>>
>>>> UnitVelocity_in_cm_per_s = 1
>>>>
>>>> UnitDensity_in_cgs = 1
>>>>
>>>> UnitEnergy_in_cgs = 1
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> Allocated 0.53112 MByte for particle storage. 68
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> reading file `non_gas.dat' on task=0 (contains 32768 particles.)
>>>>
>>>> distributing this file to tasks 0-11
>>>>
>>>> Type 0 (gas): 0 (tot= 0000000000) masstab=0
>>>>
>>>> Type 1 (halo): 32767 (tot= 0000032767) masstab=0
>>>>
>>>> Type 2 (disk): 1 (tot= 0000000001) masstab=0
>>>>
>>>> Type 3 (bulge): 0 (tot= 0000000000) masstab=0
>>>>
>>>> Type 4 (stars): 0 (tot= 0000000000) masstab=0
>>>>
>>>> Type 5 (bndry): 0 (tot= 0000000000) masstab=0
>>>>
>>>>
>>>>
>>>> reading done.
>>>>
>>>> Total number of particles : 0000032768
>>>>
>>>>
>>>>
>>>> allocated 0.0762939 Mbyte for ngb search.
>>>>
>>>>
>>>>
>>>> Allocated 1.5753 MByte for BH-tree. 64
>>>>
>>>>
>>>>
>>>> domain decomposition...
>>>>
>>>> NTopleaves= 1450
>>>>
>>>> work-load balance=1.01807 memory-balance=1.01807
>>>>
>>>> exchange of 0000030112 particles
>>>>
>>>> domain decomposition done.
>>>>
>>>> begin Peano-Hilbert order...
>>>>
>>>> Peano-Hilbert done.
>>>>
>>>> Begin Ngb-tree construction.
>>>>
>>>> Ngb-Tree contruction finished
>>>>
>>>>
>>>>
>>>> Setting next time for snapshot file to Time_next= 0
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> writing snapshot file...
>>>>
>>>> done with snapshot.
>>>>
>>>>
>>>>
>>>> Setting next time for snapshot file to Time_next= 500
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> Begin Step 0, Time: 0, Systemstep: 0
>>>>
>>>> domain decomposition...
>>>>
>>>> NTopleaves= 1450
>>>>
>>>> work-load balance=1.01807 memory-balance=1.01807
>>>>
>>>> domain decomposition done.
>>>>
>>>> begin Peano-Hilbert order...
>>>>
>>>> Peano-Hilbert done.
>>>>
>>>> Start force computation...
>>>>
>>>> Tree construction.
>>>>
>>>> Tree construction done.
>>>>
>>>> Begin tree force.
>>>>
>>>> tree is done.
>>>>
>>>> Begin tree force.
>>>>
>>>> tree is done.
>>>>
>>>> force computation done.
>>>>
>>>>
>>>>
>>>> Begin Step 1, Time: 4.88281, Systemstep: 4.88281
>>>>
>>>> domain decomposition...
>>>>
>>>> NTopleaves= 1450
>>>>
>>>> work-load balance=1.02153 memory-balance=1.38647
>>>>
>>>> exchange of 0000004772 particles
>>>>
>>>> domain decomposition done.
>>>>
>>>> begin Peano-Hilbert order...
>>>>
>>>> Peano-Hilbert done.
>>>>
>>>> Start force computation...
>>>>
>>>> Tree construction.
>>>>
>>>> Tree construction done.
>>>>
>>>> Begin tree force.
>>>>
>>>> tree is done.
>>>>
>>>> force computation done.
>>>>
>>>>
>>>>
>>>> Begin Step 2, Time: 9.76562, Systemstep: 4.88281
>>>>
>>>> domain decomposition...
>>>>
>>>> NTopleaves= 1457
>>>>
>>>> work-load balance=1.03279 memory-balance=1.36597
>>>>
>>>> exchange of 0000000226 particles
>>>>
>>>> domain decomposition done.
>>>>
>>>> begin Peano-Hilbert order...
>>>>
>>>> Peano-Hilbert done.
>>>>
>>>> Start force computation...
>>>>
>>>> Tree construction.
>>>>
>>>> Tree construction done.
>>>>
>>>> Begin tree force.
>>>>
>>>> tree is done.
>>>>
>>>> force computation done.
>>>>
>>>> Regards
>>>>
>>>> Phil Sutton
>>>> Physics Department
>>>> Loughborough University
>>>>
>>>>
>>>
>>>
>>>
>>>
>>> -----------------------------------------------------------
>>>
>>> If you wish to unsubscribe from this mailing, send mail to
>>> minimalist_at_MPA-Garching.MPG.de with a subject of: unsubscribe
>>> gadget-list A web-archive of this mailing list is available here:
>>> http://www.mpa-garching.mpg.de/gadget/gadget-list
>>>
>>>
>>>
>>>
>>> -----------------------------------------------------------
>>>
>>> If you wish to unsubscribe from this mailing, send mail to
>>> minimalist_at_MPA-Garching.MPG.de with a subject of: unsubscribe
>>> gadget-list A web-archive of this mailing list is available here:
>>> http://www.mpa-garching.mpg.de/gadget/gadget-list
>>>
>>
>>
>>
>>
>> -----------------------------------------------------------
>>
>> If you wish to unsubscribe from this mailing, send mail to
>> minimalist_at_MPA-Garching.MPG.de with a subject of: unsubscribe
>> gadget-list A web-archive of this mailing list is available here:
>> http://www.mpa-garching.mpg.de/gadget/gadget-list
>
>
>
>
> -----------------------------------------------------------
>
> If you wish to unsubscribe from this mailing, send mail to
> minimalist_at_MPA-Garching.MPG.de with a subject of: unsubscribe
> gadget-list A web-archive of this mailing list is available here:
> http://www.mpa-garching.mpg.de/gadget/gadget-list




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

If you wish to unsubscribe from this mailing, send mail to minimalist_at_MPA-Garching.MPG.de with a subject of: unsubscribe gadget-list A web-archive of this mailing list is available here:
http://www.mpa-garching.mpg.de/gadget/gadget-list
Received on 2015-01-21 10:13:41

This archive was generated by hypermail 2.3.0 : 2023-01-10 10:01:32 CET