QuantumATK Forum

QuantumATK => General Questions and Answers => Topic started by: esp on September 4, 2012, 21:08

Title: runtime with asymmetric doping
Post by: esp on September 4, 2012, 21:08
I am wondering, I made a graphene transistor device with symmetrically doped source and drain, runtime for a transmission spectrum calculation was something like 40 minutes per Vds, Vgs point ... then I made another of same size with different doping in source and drain, and it has run i think for like 10 hours and has not completed the first point yet ... it is still running, no errors, just taking a long time .. does this sound reasonable or does this make sense to you?  Even if the symmetric case re-used calculations, still i would not think it would take longer than 2x the time ...
Title: Re: runtime with asymmetric doping
Post by: esp on September 5, 2012, 02:50
attached is the log currently, 12 hours running .. maybe you can tell if it is running ok? it seems to be, but 40 minutes to over 8 hours seems strange
Title: Re: runtime with asymmetric doping
Post by: Anders Blom on September 5, 2012, 15:41
No, it doesn't make sense.
Could you send me the entire structure by email, please?
Title: Re: runtime with asymmetric doping
Post by: esp on September 5, 2012, 22:45
Just ran an LDOS calculation .. ran on 10 nodes, 8 processors per node, mpi, cluster setup with 2.5GB memory per ran for 24 hours then died .. the last lines in the log are below .. it seems it did the initial part yesterday only took 3 hours, then hung on one of the ldos calcs ... is there any log file or atk file i can look up to see what happened? there were no errors on the system, it stopped because i can only request 24 hours at a time:

or any other idea what might have happened?

Code
|  287   H   [  12.000 ,  24.028 ,  48.149 ]    0.79415  -0.20585              |
+------------------------------------------------------------------------------+
|  14 E = -552.716 dE =  1.364983e-06 dH =  3.032989e-06                       |
+------------------------------------------------------------------------------+
| Calculation Converged in 14 steps                                            |
+------------------------------------------------------------------------------+
+------------------------------------------------------------------------------+
|                                                                              |
| Device DFT Calculation  [Finished Thu Sep 20 23:21:46 2012]                  |
|                                                                              |
+------------------------------------------------------------------------------+

                            |--------------------------------------------------|
Calculating LDOS           : ==================================================
doLDOS [vg-0.2V][vds0.1V] -3.0 eV 09-20-12 11:22 PM

                            |--------------------------------------------------|
Calculating LDOS           : ==================================================
doLDOS [vg-0.2V][vds0.1V] -2.8 eV 09-20-12 11:22 PM

                            |--------------------------------------------------|
Calculating LDOS           : ==================================================
doLDOS [vg-0.2V][vds0.1V] -2.6 eV 09-20-12 11:22 PM

                            |--------------------------------------------------|
Calculating LDOS           : ==================================================
doLDOS [vg-0.2V][vds0.1V] -2.4 eV 09-20-12 11:22 PM

                            |--------------------------------------------------|
Calculating LDOS           : ==================================================
doLDOS [vg-0.2V][vds0.1V] -2.2 eV 09-20-12 11:22 PM

                            |--------------------------------------------------|
Calculating LDOS           : ==================================================
doLDOS [vg-0.2V][vds0.1V] -2.0 eV 09-20-12 11:23 PM

                            |--------------------------------------------------|
Calculating LDOS           : =

the LDOS function I am running I have posted before, pretty basic and comes down to this:

Code
			for energy in energies:
				LDOS = LocalDeviceDensityOfStates(
					configuration=device_configuration,
					energy=energy,
					kpoints=MonkhorstPackGrid(1,1),
					contributions=All,
					energy_zero_parameter=AverageFermiLevel,
					infinitesimal=1e-06*eV,
					self_energy_calculator=KrylovSelfEnergy(),
					spin=Spin.Sum,
				)
Title: Re: runtime with asymmetric doping
Post by: esp on September 22, 2012, 02:31
actually does it signify an issue that the LDOS took 0 time to calculate? i thought it should take some time
Title: Re: runtime with asymmetric doping
Post by: Anders Blom on September 22, 2012, 21:18
It doesn't take zero time, but less than 1 minute, afaIcs. So, it's a small task. About the hanging I have no immediate answer, seems spurious, making it hard to troubleshoot.
Title: Re: runtime with asymmetric doping
Post by: esp on September 22, 2012, 21:48
ok thank you that is what i was hoping