Exclusive licence for meshing and solving

[OmkarJ]

Hi

It is puzzling that while we are meshing, we have to give away a licence in solving, forcing us to stop the jobs running in the queue, just to be able to mesh. In most commercial CFD (or FEA) software packages that I know of, I see that meshing and solving are completely exclusive of each otherin a single licence.

Majority of time of every CFD engineer is occuppied in meshing at his workstation. At the same time, he can't finish his job unless these meshes are run, often in the remote cluster etc. Sorry if I sound rude, but is it me, or has Autodeksk really employed this unfair and unjust tactic to engage a solver licence in meshing, thereby indirectly forcing users to go for additional solver licences, so that they can continue both meshing and solving uninterrupted at the same time?

I know this may call for a major overhaul in code (or maybe not), but is it not reasonable to demand that  "meshing" and "solving" be treated exclusive of each other?

OJ

[hartogj]

Appreciate the open and honest feedback.

The mesher/solver arrangement is designed to support the workflow most users want to follow. By meshing on the remote machine, you get to leverage its CPU/RAM for meshing and you reduce the outbound data transfer. It also keeps things simple when you want to run with mesh adaptation, which involves both the mesher and the solver concurrently.

I understand the above may not be what you feel is optimal but am curious to hear what others have to say.

Keep in mind this has been the arrangement since way back in the CFdesign days so shouldn't be new. Actually under Autodesk we have been able to open things up a little, for example giving each solver license full HPC capability.

Jon

Product Manager

[OmkarJ]

Thanks for noting. I understand that this arrangement is since CFDesign and Autodesk adopted it in legacy. Nevertheless, every new iteration comes (and should come) with welcome changes!

To be honest, when I try meshing with my own workstation versus remote cluster in office, I do see little improvement in time required for meshing. The configurations are:

• My workstation: i7-2600, 3.4GHz processor and 16 GB RAM
• Cluster: Four boxes, each with i7-2600, 3.4GHz processor and 16 GB RAM, connected to each other using  Infiniband SDR 4X with RDMA, 10 Gbits/s, latency ~5 microseconds

So if you have a reasonable workstation, would using cluster as a mesher remotely any beneficial over using the workstation locally? I have observed that even multithreading upto 4 processors when used (without extruded meshes) doesn't yield significant improvement, indicating little improvement while using cluster for meshing.  

It is understood that while using mesh adaptation, you really need to solve for the results to have gradients, and it is fair that while solving (not while meshing) the solver licence should be engaged. At the same time, many users (including me) rely on judgment and prefer do their own mesh sensitivity studies insted of relying on mesh adaptation. And this arrangement is an inconvinience for them 🙂

Regards

OJ

[napoleonm]

That would be a nice enhancement for the software..

a much practical work flow even for people or companies with just one lic.

[Royce_adsk]

Jon,

Maybe we should consider using 0 iterations as a non-license check operation similar to what we have now with Sim360 Pro.

User would still need the DSE license to initiate running 0 iterations.

-Royce

[hartogj]

Thanks for the ideas guys. We can't promise a change at this point but we can look into it.