Jared Frazier

Results 33 comments of Jared Frazier

I would like to add an example for a geometric approach to DD, though admittedly I'm still quite new to domain decomposition.~ In trying to get an idea for DD...

> For completeness I want to mention that, as an alternative to Schwarz methods, there is also FETI as paradigm. Another relevant package I found is [AMfeti](https://github.com/AppliedMechanics/AMfeti), which implements FETI...

> I would lean towards agreeing with Wolfgang Bangerth on this for linear problems and I highly favor AMG over DD methods myself. However, DD methods are still an active...

> Okay, maybe we need to go one step back here. What exactly is the problem you are trying to solve here? Maybe I am missing something, but if you...

Thank you for working with me on this. It is my first proper open-source contribution beyond trivial documentation PRs, so I appreciate your patience! I think the style in https://ferrite-fem.github.io/Ferrite.jl/stable/tutorials/hyperelasticity/...

Two comments that are unrelated to this direct PR but which might warrant opening **separate** issues related to the Navier-Stokes tutorial: 1. Maybe I open a separate issue based on...

- [X] need to update parenthesis in derivation for `dot(v, gradv)`, should be `dot(v, grad)v`

@termi-official Any thoughts on my comments here https://github.com/Ferrite-FEM/Ferrite.jl/pull/1138#issuecomment-2763286050 ? Regarding opening another issue/PR for > Maybe an extra details section on the NS weak form derivation would be useful? if...

Commit a73aaa9cce2eb9f618a729e761a763c4304635a5, particularly `src/tmp.jl` is an attempt to handle the master-slave problem introduced by 3D unit cell geometries. The image below is a screenshot of `outputs/debug_3D_periodic_puzzlepiece.vtu`, which can be reproduced...

@fverdugo I have updated the `src/geometry.jl` function in commit 63d431814ab66d801514b4b71a1157f5c468a6d7 to use the corrected version you provided. Corresponding tests in `example.jl` all pass; however, I think we need to still...