March 2014 - Nektar-users - imperial.ac.uk

Tetrahedron bug
by David Moxey 31 Mar '14

31 Mar '14

Dear all, Apologies for the spam, but just wanted to give a quick notification for anyone running curvilinear tetrahedra. I have noticed that at sufficiently high order (NUMMODES > 10) the code crashes with a memory corruption error. This is due to an incorrectly sized workspace as outlined in the bug report (http://www.nektar.info/ticket/136) Since this is unlikely to affect anyone in the short term, the workaround in the report should temporarily fix the issue until a more permanent solution is found. Thanks, Dave

1 0

ARCHER compilation instructions
by David Moxey 31 Mar '14

31 Mar '14

Dear all, Just a quick note for anyone using ARCHER. I noticed that boost 1.55 is now available as a module file, which means that you no longer need to compile boost using THIRDPARTY_BUILD_BOOST. The (internal) compilation instructions have been updated to reflect this. Unfortunately there is still a need to have a copy of the code in $WORK as opposed to $HOME, because it doesn't seem the compute nodes have access to the filesystem and therefore runtime linking fails. Since it appears compilation can be quite slow on $WORK sometimes, I have updated the instructions to guide installation into $WORK but with $HOME as the compilation directory. Additionally since HECToR has now been decommissioned, I have removed these instructions from the wiki. Thanks, Dave

1 0

FieldConvert update
by Sherwin, Spencer J 27 Mar '14

27 Mar '14

Hi, We have now pushed the fix/Fieldconvert branch into master. This update has added C0Projection (perform a C0 projection) and interpfield (interpolate one field to another) options to the master branch. I have put some preliminary instructions on how to use these options under the page: https://www.nektar.info/wiki/3.4.1/UserGuide/Tutorial/FieldConvertInterpPoi… which allow one to interpolate a set of points. There is also the ability to add fields in this branch. Once I have tested this option I will update the Wiki with some instructions. Cheers, Spencer. I now have a feature/FieldConvert Spencer Sherwin Professor of Computational Fluid Mechanics, Department of Aeronautics, Imperial College London South Kensington Campus London SW7 2AZ s.sherwin(a)imperial.ac.uk +44 (0) 20 759 45052

1 0

Welcome
by David Moxey 24 Mar '14

24 Mar '14

Dear all, Welcome to the Nektar++ users mailing list! As Nektar++ is now being used by a number of users outside of ICL and Utah, it's become apparent that we need a new list so that we can inform everybody of: - critical bugs and errors in the code - details of new releases - some means of providing support/help from a larger pool of developers However it's hoped this can also be of wider use to everyone. There's been quite a few instances recently where we've discovered bugs only to find that they have been solved previously in another branch or solver, but that information never got communicated to everyone. Please feel free to use this mailing list as frequently as you see fit to let people know about features you're developing/interesting in developing, bugs you find and/or fixes you've made in branches - *as well as* submitting tickets for bugs when you encounter them! Registration is (hopefully) open to everyone, so I would encourage you to pass on the details of this list to anyone you know that is currently using Nektar++. Many thanks, Dave -- David Moxey Research Associate Department of Aeronautics Imperial College London South Kensington Campus Room 253, Roderic Hill Building London, SW7 2AZ d.moxey(a)imperial.ac.uk www.imperial.ac.uk/people/d.moxey

1 0