Why Does MCNP5 Behave Differently Across Operating Systems?

  • Thread starter Thread starter junaidarshad
  • Start date Start date
  • Tags Tags
    Loops Mcnp5 Stuck
AI Thread Summary
MCNP5 is experiencing issues with simulations getting stuck in a loop on a CentOS 6.5 cluster using mpich2-1.4.1p, while the same simulations run successfully on Windows with mpich-1.2.4. The problem is attributed to how MCNP divides tasks for parallel processing, where only one process may become stuck while others complete. Some users have reported that allowing the simulation to run for an extended period can eventually resolve the issue. However, others have noted that even after a long wait, the simulations do not return to normal operation. Understanding the differences in parallel processing between the operating systems may help address these discrepancies.
junaidarshad
Messages
7
Reaction score
0
I recently installed MCNP5 on rocks cluster running centOS 6.5, with mpich2-1.4.1p.
While running simulations, MNCP5 is stuck in loop in 2 of the simulations.

When I tried to perform one of the stuck simulations on windows with mpich-1.2.4, it didn't stuck. Can anyone guide me why MNCP5 is behaving differently on different operating systems.

Regards,
 
junaidarshad said:
I recently installed MCNP5 on rocks cluster running centOS 6.5, with mpich2-1.4.1p.
While running simulations, MNCP5 is stuck in loop in 2 of the simulations.

When I tried to perform one of the stuck simulations on windows with mpich-1.2.4, it didn't stuck. Can anyone guide me why MNCP5 is behaving differently on different operating systems.

Regards,
I got the error too, mcnp5 stucked at 2 of simulations loop.
Did you solved it ?
 
wangm23456 said:
I got the error too, mcnp5 stucked at 2 of simulations loop.
Did you solved it ?

There was nothing to solve. The problem was because of the way MCNP divides the problem for parallel solution. Only one of the Processes was stuck, while rest of the processes had done their job. I let it run and after some time it solved the problem.
 
junaidarshad said:
There was nothing to solve. The problem was because of the way MCNP divides the problem for parallel solution. Only one of the Processes was stuck, while rest of the processes had done their job. I let it run and after some time it solved the problem.
I let it run but it can't return normal after a very long time.
 

Thread 'Energy Spectrum Attenuation'

Hello, I'm currently trying to compare theoretical results with an MCNP simulation. I'm using two discrete sets of data, intensity (probability) and linear attenuation coefficient, both functions of energy, to produce an attenuated energy spectrum after x-rays have passed through a thin layer of lead. I've been running through the calculations and I'm getting a higher average attenuated energy (~74 keV) than initial average energy (~33 keV). My guess is I'm doing something wrong somewhere...
View full post »

Similar threads

Why Does My Varactor Diode Test Circuit Behave Differently on a Breadboard?
2
Replies
93
Views
7K
Two operating systems or a virtual machine
Replies
4
Views
2K
Requesting suggestions for languages, libraries, and architectures for parallel (and sometimes non parallel) numerical and scientific computations
Replies
8
Views
4K
Sending drive Pulse in CQED and visualize the state by QuTip
Replies
6
Views
3K
Investigation of hydrogen-on-demand systems (HODS)
Replies
4
Views
31K
Build a New System Some A/V and Moderate Gaming Requirements
Replies
7
Views
3K

Hot Threads

Recent Insights

Back
Top