-
Integrate the Juelich Benchmark Environment JUBE (http://www.fz-juelich.de/jsc/jube) to have a controlled way to execute and to track tests, benchmarks, production runs etc. It's plain python (2.7 and…
-
[This is a good SE post for explaining the difference between the major groups](http://scicomp.stackexchange.com/questions/7714/whats-the-state-of-the-art-in-parallel-ode-methods).
[Here is another…
-
Implement tests for
- Sweeper
- imex_1st_order
and make sure that the derived update matrix gives the same result as performing k many sweeps.
-
I'm trying to specify a custom datetime format via a config file:
```
## Default configuration for easyloggingpp
* GLOBAL:
ENABLED = true
To_File = false
To_Standard_O…
-
Hello, everyone!
I am a newer on pySDC. I notice that pySDC works with [FEniCS](https://fenicsproject.org/), [mpi4py-fft](https://mpi4py-fft.readthedocs.io/en/latest/) and [PETSc](http://www.mcs.anl…
-
Hello!
I am working on a prototyping framework on spectral deferred corrections called [pySDC](https://github.com/Parallel-in-Time/pySDC). Our main focus is parallel-in-time methods such as [diagonal…
-
About the `spread` initialization in the base Sweeper class. Why should the evaluation of the RHS depend on the node time ? cf :
https://github.com/Parallel-in-Time/pySDC/blob/2f72ea3b923bfd58b2ede…
-
The function PolyInterpMixin::fas(...) in file poly_interp_impl.hpp is the function to compute FAS correction 'tau' iterm. In this function, the matrix fmat seems to have rows like [ -1 -1 -1 1 | 1 1 …
-
Hi @JensHahne,
following our discussion, here is the image with the correct task factorization for PFASST :
![image](https://user-images.githubusercontent.com/12183839/221248864-9381646f-87ee-45…
-
Hey @JensHahne, just to check with you : I would like to make the `speedup` method return a tuple ` (speedup / efficiency)` (and eventually, rename it `getPerformance`, so this would simplify the impl…