These are chat archives for Fortran-FOSS-Programmers/General-Discussion

12th
Jan 2017
Izaak "Zaak" Beekman
@zbeekman
Jan 12 2017 01:36
Well it depends how the grid was generated... I have done some composition of analytic functions (conformal mapping which is investable, and 1-D stretching functions) or finite difference if the grid was elliptic or hyperbolic etc numerically generated. Obviously the difficulty becomes proving > 2nd order accuracy if you then also have a numerically generated grid and are using FD etc. to compute the jacobians or grid metrics... But the case that Stefano describes sounds very complicated... In particular with overset/chimera grids you lose conservation in the interpolated region... There is some work being done to ensure stable, conservative interpolation, but it's a real pain....
As for the expense of the smoothness coefficients, yes that is a big problem. I haven't tried or researched it, but I wonder if you could switch your basis from lagrange polynomials to something else that would yield a simpler formulation?
A good approach might be to use a compact WENO higher order scheme... I think you only need nearest neighbor for that approach, but it's been a long time since I have read the paper: https://scholar.google.com/scholar?hl=en&q=compact+WENO+scheme+baeder&btnG=&as_sdt=1%2C7&as_sdtp=
Izaak "Zaak" Beekman
@zbeekman
Jan 12 2017 01:45
If you can increase the bandwidth resolving efficiency and order of accuracy of the scheme, sometimes having more expensive computations can be close to free, as long as you have good data locality... These days on KNL etc. data motion and memory hiearchy can often be a much bigger bottleneck than FLOPS, whose relative cost has been declining. Having a higher arethmatic intensity (i.e. not memory bound) can be cheap relative to less computationally intensive algorithms that are memory bound. I would check out compact WENO schemes from Baeder (He was a student of McCormick's, just like Graham Candler who is my ex-advisor's advisor....)
Izaak "Zaak" Beekman
@zbeekman
Jan 12 2017 01:59

A little bit late, but just FYI, a new release of OpenCoarrays is here. Here are the release notes and download links etc.:


Github Releases (by Asset) Build Status license Twitter URL

Enhancements

  • Patch gcc's contributed download_prerequisites script to detect presence of wget and try alternates like curl if it is missing

Added Regression Tests

Added GCC 7 trunk regression tests. Thanks to @egiovan for reporting segfaults with event post when num_images > 2 and the type conversion before coarray puts regressions with GCC trunk.

Installation

Please see the installation instructions for more details on how to build and install this version of OpenCoarrays


GitHub forks GitHub stars GitHub stars Twitter URL

Stefano Zaghi
@szaghi
Jan 12 2017 16:05
@zbeekman Zaak, thank you very much for your hints, they are really interesting. In the past, I studied Martin's works, I am aware of bandiwth optimized model. Simply, my time was (is) limited... surely it will be strongly considered to be added to WenOOF. Compact WENO, on the contrary, is a new concept for me! Thank you very much!
Izaak "Zaak" Beekman
@zbeekman
Jan 12 2017 19:12

I'm almost done with a mac OS X Homebrew "Formula" for OpenCoarrays, but it's not quite popular enough to meet the required "notoriety" metrics to be included in the main package repository (homebrew-core tap). We only need 3 more forks, 6 more watchers and/or 16 more stargazers. Any help would in increasing our popularity metrics would be much appreciated!


GitHub forks GitHub stars GitHub stars Twitter URL

Once in Homebrew it will get picked up into the downstream LinuxBrew as well
Stefano Zaghi
@szaghi
Jan 12 2017 21:27
@zbeekman I have already starged, watched. I have now forked it. Tomorrow I'll force my boss to do the same if still there will be the need :smile:
I start with not a my boss... @giacombum Giacomo starge, watch and fork opencoarrays please :smile:
Izaak "Zaak" Beekman
@zbeekman
Jan 12 2017 23:30
@szaghi thanks :bow: due to your fork we are over the threshold for popularity/notariety!
For thos interested the PR is here: Homebrew/homebrew-core#8790