Release of FDS 6.7.0, SMV 6.7.1 bundle

[Randy McDermott]

The FDS 6.7.0, SMV 6.7.1 bundle is now available.

https://pages.nist.gov/fds-smv/

Release notes can be found here.

One key aspect of this release is the introduction of a SIMULATION_MODE parameter to control the appropriate level of physics submodels used for the calculation.  This parameter can take on the string values 'DNS', 'LES', 'VLES', or 'SVLES'.  The default is 'VLES', which stands for "Very Large-Eddy Simulation" (believe it or not, this is recognized terminology in the LES community).  'SVLES' invokes a "Simplified VLES' strategy that, when appropriate, may be used to speed up calculations.  Details are provided in the FDS User Guide.

It is our hope that with this minor release the development team can start focussing on the release of FDS 7, which will enable curvilinear geometry capabilities.  So, please test this latest bundle at your earliest convenience and provide feedback.  If we need to do a maintenance release for 6.7.0, we would like to do it as soon as possible.  Thanks.

[fde]

Download is slow and fails, is it only me or a common issue on the website?

[Ben Ralph]

I downloaded okay

--
You received this message because you are subscribed to the Google Groups "FDS and Smokeview Discussions" group.
To unsubscribe from this group and stop receiving emails from it, send an email to fds-smv+u...@googlegroups.com.
To post to this group, send email to fds...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/25df5754-9b8e-4148-9443-0062a98fb493%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[fde]

Thank you, I should look into my network status.

[Bastian]

I noticed that the set up of the geometry takes longer in 6.7. In a precise case (750.000 cells in 1 Mesh) FDS 6.6 needs less than 1 second to set up the geometry. In FDS 6.7. the "Checking mesh alignment. ..." needs about 10 seconds although it is just one mesh. In another case (2 Meshes, each 800.000 cells) the "set-up only" procedure need about 15 seconds. Against 1 or 2 seconds in FDS 6.6.

Since i check a lot during building the geometry, it slows down the workflow significantly. Is there a chance to reduce the time of the "set-up only" procedure?

Thank you and best regards,

Bastian

[Bastian]

Hi!

I found a malfunction in smokeview. I can't render the current frame. If I choose that option, the setting stays on the prior setting. I can choose "all" or "Every 2nd"... but not "Current".

rgs

Bastian

[Randy McDermott]

OK, I have reproduced your problem with Smokeview.  Although, you know you can just hit the "r" key to render the current frame, right?  That still works.  Glenn will test other issue.

To unsubscribe from this group and stop receiving emails from it, send an email to fds-smv+unsubscribe@googlegroups.com.

To post to this group, send email to fds...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/a70e6bae-35b2-4be0-bad4-c39676d9e968%40googlegroups.com.

[Bastian]

The shortcut works. Thank you!

[Randy McDermott]

This was put in place because as we run bigger jobs we were finding that geometry setups would pass and then we would launch a big MPI run assuming all was find, only to find out that the meshes were not aligned.

We will discuss and see if there is a way to have both options.  Thanks for the feedback.

To unsubscribe from this group and stop receiving emails from it, send an email to fds-smv+unsubscribe@googlegroups.com.

To post to this group, send email to fds...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/486162f7-1ebd-4a28-a7d2-c18b3d09e5ce%40googlegroups.com.

[Randy McDermott]

Follow setup issue here: https://github.com/firemodels/fds/issues/6579

Follow SMV issue here: https://github.com/firemodels/fds/issues/6578

To unsubscribe from this group and stop receiving emails from it, send an email to fds-smv+unsubscribe@googlegroups.com.

To post to this group, send email to fds...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/5aeedd86-3ecb-42dd-9ba1-6ba1093949f5%40googlegroups.com.

[o...@aquacoustics.biz]

I have just completed installation of FDS 6.7.0 using the NIST bundle on my Infiniband networked Linux Cluster.  The installation was painless and faultless.  Verification was completed in accordance with the Technical Guide.  Many of the pass margins appear to have improved over FDS 6.6.0.  My speed metrics show almost identical processing time to FDS 6.6.0.  You can read the Verification Report including opemmp_test128h metric at http://fire.aquacoustics.biz/html/publications.html.  Click on the 'FDS Verification Report' link.

I haven't run the weak and strong scaling tests but I expect that, on the basis of other metrics, these will remain almost identical to previous performance.

A further note is that my previously reported problem with CHID.stop is no longer an issue.  

I could probably improve processing performance slightly with processor-specific Intel FORTRAN compiler optimizations, but the anticipated improvement will be small.  I'll experiment with this in due course.

My thanks to NIST for continued support and development of FDS.

[Kevin]

We have received a few discussion and issue tracker posts from users having trouble running FDS 6.7.0 on Windows 10 using mpiexec and MPI. For those of you reading this post who have run FDS 6.7.0 under Windows 10 (or even Windows 7), could you tell us

1. Did mpiexec -n ... work successfully for a job either running on one computer using multiple cores or across a network.

2. What version of Windows

[dr_jfloyd]

Windows 7 Enterprise SP 1

It runs on multiple cores on my computer, but I have to make sure I run a cmd window without the Intel compiler path variables set.

[Salah Benkorichi]

2 Desktops with OS Windows 10: it runs on multiple cores.

2 Laptops with OS Windows 10: mpi doesn't work.

1 laptop  with OS Windows 8: mpi doesn't work.

--

You received this message because you are subscribed to the Google Groups "FDS and Smokeview Discussions" group.
To unsubscribe from this group and stop receiving emails from it, send an email to fds-smv+unsubscribe@googlegroups.com.
To post to this group, send email to fds...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/f8276ff4-f172-44c8-8cd1-9f3f6f049fb9%40googlegroups.com.

[Mikkel Lund Mortensen]

Single desktop with multiple cores, Windows 10 - mpiexec works with "-localonly", fails without it.

[Rusbe]

A curvilinear geometry capability in FDS7 would be great. But that means the users must invest more efforts on the grid generation than before? That would be great if one can also import meshes from other mesh generation software in some standard formats such as CGNS etc. Anyway, I like the current fearture of FDS that the obstacles are forced to conform to the grid lines. In this way one can change the geometry (some simple parts, like an array of the cars in a tunnel) without getting worried for the grid.

On Wednesday, June 27, 2018 at 4:00:24 PM UTC+2, Randy McDermott wrote:

[Randy McDermott]

FDS generally uses an immersed boundary method, which means the solid obstructions are immersed in a Cartesian grid.  This will continue to be true for FDS 7.  But the option to use curvilinear geometries will then be available.  In that case, FDS will automatically generate an unstructured finite-volume "cut cell" mesh around the surface of the solid for gas phase transport near the solid.  The user will not be responsible for generating this mesh.  But the user will be responsible for generating the geometry.  Third-party tools like Pyrosim or BlenderFDS will usually be needed.  Just to be clear, though, the old Lego block geometry method via (OBST) will still be available.

--

You received this message because you are subscribed to the Google Groups "FDS and Smokeview Discussions" group.

To unsubscribe from this group and stop receiving emails from it, send an email to fds-smv+u...@googlegroups.com.

To post to this group, send email to fds...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/cd4230a0-b63b-4cb8-be77-5db6eaa89a1b%40googlegroups.com.

[Rkj]

The download page of FDS-SMV is not accessible. It shows an error 404: Page not found. Has anyone else faced this issue?

[Salah Benkorichi]

Yes, it's down. 

There is a way around to this for now. You can download fds from here : https://github.com/firemodels/fds/releases/tag/FDS6.7.0

On Thu, 27 Dec 2018 at 10:30, Rkj <rahul...@gmail.com> wrote:

The download page of FDS-SMV is not accessible. It shows an error 404: Page not found. Has anyone else faced this issue?

--
You received this message because you are subscribed to the Google Groups "FDS and Smokeview Discussions" group.
To unsubscribe from this group and stop receiving emails from it, send an email to fds-smv+u...@googlegroups.com.
To post to this group, send email to fds...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/51b37d70-6e09-4586-b74c-83c9942efc93%40googlegroups.com.

[Rkj]

I see, thank you for the link.