Questions tagged [parallel-computing]
The study of solving computational problems by exploiting the use of multiple processors simultaneously.
325
questions
0
votes
2
answers
75
views
Source for scalability challenge for number of finite element nodes per process
Context
In distributed simulation of a finite element mesh with $N$ nodes and $P$ processes, a professor stated to me that "achieving good scaling for more than 25,000 finite element nodes per ...
0
votes
0
answers
53
views
Parallel vs Serial computing using Python Qutip. Why is serial faster?
I am trying to learn parallel computing using qutip's parallel_map function. I've tried to write a basic and simple code to understand the differences between parallel and serial calculation of ...
1
vote
1
answer
126
views
Is it possible to run a Metropolis Monte Carlo simulation in parallel?
Is it possible to run a Metropolis Monte Carlo simulation in parallel?
Suppose I perform a Metropolis Monte Carlo simulation using four threads.
Suppose, the programming source code divides a ...
0
votes
1
answer
82
views
Non-dimensionalizing the Ideal MHD System
Non-dimensionalization is a really frustrating topic for me, and I imagine many others, because in school it was glossed over while being really important to implementing a simulation.
I'm writing a ...
-1
votes
1
answer
107
views
What is the name of the theory that combines 3d discretized surfaces and distributed numerical algebra
I've looked into distributed numerical computation work before, but I've realized that 3D applications are all about 3d surfaces and meshes.So I'd like to look into related work on mesh and parallel ...
0
votes
1
answer
101
views
row and column based distributed LU factorization
The LU parallel computation theories I've seen are based on $M\times N$ mesh computations, is there a theory for one dimensional device mesh LU parallel decomposition?
For example, $A$ is a matrix. we ...
2
votes
2
answers
483
views
Iteration counts of AMG solver changes in parallel
I am solving the linear elasticity equation within a FEM library with a complex 3D geometry. The resulting linear system is solved with CG, preconditioned by AMG (Algebraic Multigrid). The computed ...
2
votes
0
answers
80
views
Diagonalization of large sparse matrix, computational programme recommendation and methods
According to this link, All eigenpairs of large sparse symmetric matrix. The guy @Baranas seems to have given a very confident answer about solving the whole Eigen spectrum. May I know if anyone has ...
0
votes
2
answers
714
views
PhD in scientific computing to be a scientific programmer
Intro and disclaimer: this question concerns developing a career in Scientific Computing in industry, starting from an (applied) mathematics background, say an MSc. It definitely arises from my ...
0
votes
1
answer
157
views
Lattice Boltzmann method parallelization
I'm learning about the lattice Boltzmann method (LBM) and how it can be parallelized. On Wikipedia, they say, "The LBM was designed from scratch to run efficiently on massively parallel ...
2
votes
0
answers
83
views
OpenMP Fortran array operations in target regions
I am asking this here for the possibility that someone here is familiar with OpenMP offloading. Is there some other way to parallelise Fortran array operations outside of unrolling the loops of the ...
0
votes
0
answers
30
views
Parallel Block-Structured class abstraction for FDM
I’m currently developing a FDM/FVM (using contravariant coordinates) code using Fortran and Co-Arrays (SIMD, in general), and so far I have all sparse matrix (BiCGStab, working on AMG) solvers and ...
4
votes
1
answer
341
views
Under what circumstances is parallel scaling of the finite element method not "solved"?
I see things like this deal.II example, (ctrl + f for "superMUC") which seems to show some pretty impressive scaling (nearly twice as fast for twice as many CPU cores for a wide range of ...
1
vote
0
answers
42
views
Where to find datasets to research the optimal parameter search space for HPC systems? [closed]
I am a student researcher who is new to the HPC domain.
I have recently taken a project where I am working on optimizing the parameter search space (both application level and hardware level ...
2
votes
0
answers
171
views
Why GPU scaling (speedup) is usually worse than CPU scaling?
Let's define scaling, as how linear the speedup of using more than one GPU or CPU is. For example, having 2 GPUs gives you 2x faster execution time.
I have noticed that in many software (e.g. ...