FLUENT

History

FLUENT has an annual license, and it was scheduled to expire 4-Oct.  We (Cindy Twohy and me) started using FLUENT ~10 years ago.  Cindy preferred to use a different simulation package, STAR-CD, but FLUENT was my preference. It was installed on eol-flow and my laptop. On the laptop, I used it mostly to build geometry and to display results, because the laptop didn't really have enough computing horsepower for big meshes (>1 M cells) or solving the flows. A few years ago, ANSYS acquired FLUENT software as part of its engineering suite of software. NCAR is eligible for the academic discount at ANSYS, which is a great cost factor.  Our contact at ANSYS has been Brenda Elser (916-803-2347 phone, brenda.elser@ansys.com).  She handles license renewal and requests for updating license files when our license server changed.

As for eol-flow, it was a very useful shared linux workstation but its performance was eclipsed by time and Moore's law. Besides FLUENT, it also had TecPlot installed, which we used to analyze the GV flow modeling results that were performed by Gulfstream (Paul Olsen) and were provided to us as TecPlot files. Those flow files were used for the small wing store project and for the radome gust probe hole placement. All those files resided on eol-flow.  I think they were provided on CD's and were also copied to the hiaper network disk.

About 5 yr ago, I moved FLUENT off eol-flow and onto the CISL high performance cluster in order to (a) take advantage of the resources and (b) make it available to the atmos user community.  In concert with CISL's move to yellowstone last year, FLUENT was also moved from the Mesa Lab to Wyoming. CISL's involvement in the installation and licensing included Shawn Neeham (sneedham@ucar.edu) and Si Liu (siliu@ucar.edu). I was the primary NCAR FLUENT contact and submitted a number of ExtraView requests for CISL help (for example, #82142 a license update). In May, as my retirement drew near, my FLUENT contact role was passed to Shawn.

Historically, RAF has always paid for the annual license renewals ~$7.5k, and there were only a few people who used FLUENT. That always surprised me, because it is such a powerful system for flow modeling, calculating particle trajectories and thermodynamic fields, etc. The geometry building and meshing were coupled to other ANSYS products, and it was fairly easy to incorporate a SolidWorks geometry file from SteveR.  Anyway, besides me, the users at NCAR included Jim Walega, Duncan Axisa, and occasional ACD and RAL visitors. There were also a few atmos community users, such as grad students from Clarkson Univ in Suresh Dhaniyala's group and from CU and CSU.  In June, I asked ~10 users for their opinions about continuing FLUENT at NCAR's expense. Their response was piecemeal and weak. My conclusion is that people recognize the value of the resource but they have more important things to work on.  FLUENT would be very valuable for simulation studies of existing instruments, and NCAR/EOL is an appropriate host for the user community.  However, flow modeling needs to be attached to a person that has some CFD background and motivation to use it.  I think that flow modeling/trajectory studies could thrive given the right champion.

I hope these comments are useful for you.

   ..Dave..

 

Notes