Hi Sumaya,

About ParaView MPI, what I meant is that, if the memory is the issue, you 
can consider running a pvserver on a potentially multi-node, high-RAM 
server, then using something like paraview --client to connect to the 
server to view the remote-rendered scenes. This way, you are not bound by 
your local memory. You have to figure out this workflow though.

I am not familiar with rendering using CSV input in a Houdini environment, 
supposing this is what you are contemplating. But I wouldn't imagine any 
problems for CSV, as it is just a general-purpose format.

Thank you,
Ruochun

On Wednesday, October 29, 2025 at 4:18:18 AM UTC+8 sumaya wrote:

> Hi Ruochun,
>
> Thank you for your response. Is it possible to elaborate your answer in 
> terms of using a partitioned view or MPI?
> Have the files (CSV / VTK) ever been tested in Houdini environment?
>
> Thanks
>
>
> On Wednesday, October 15, 2025 at 8:17:56 AM UTC-4 Ruochun Zhang wrote:
>
>> Hi Sumaiya,
>>
>> Apparently I meant GRC-1 not GRC-3 in the previous message. It would be 
>> interesting to develop a GRC-3 representation though!
>>
>> Ruochun
>>
>> On Wednesday, October 15, 2025 at 8:09:45 PM UTC+8 Ruochun Zhang wrote:
>>
>>> Hi Sumaiya,
>>>
>>> 1. I don't know about the 58GB thing, sounds a bit larger than I expect 
>>> but you could be right. When you visualize it in ParaView, consider 
>>> visualizing the points only without generating gpyph; or if you would like 
>>> glyph, keep the mode being "All Points", but reduce the Theta and Phi 
>>> resolution, to maybe 3 or something like that, and that will greatly reduce 
>>> the memory used. However, if the output size keeps growing, at some point 
>>> you'll have to use some sort of MPI or partitioned view.
>>>
>>> 2. Sorry for the lack of comments. Part3 is still about preparing a 
>>> GRC-3 particle bed. It requires finishing Part2. It makes several copies of 
>>> the results from Part2, put them side by side to make a larger material 
>>> patch, then let the gravity do the work. After settling, it compresses the 
>>> resultant material bed so the top is a bit more even/flat. Then it saves 
>>> the settled material to a file. Note that this file is very large, 
>>> representing a 4m × 2m soil bin which would allow a rover running on it. If 
>>> you don't need a test environment this big, you can safely ignore Part3.
>>>
>>> Thank you,
>>> Ruochun
>>>
>>> On Wednesday, October 15, 2025 at 1:48:33 AM UTC+8 sumaya wrote:
>>>
>>>> Dear Chrono Users,
>>>>
>>>> *Part 1: ParaView*
>>>>
>>>> I would like to ask about your experience running DEM simulations in 
>>>> Chrono. I have successfully generated files for a wheel drawbar pull 
>>>> simulation using the Chrono DEM engine. The simulation contains close to a 
>>>> million particles, as mentioned in the comments of the wheel_DP.exe file, 
>>>> and the total file size is around 58 GB.
>>>>
>>>> However, when I try to visualize the results in ParaView, after 
>>>> applying all the recommended filters from the GitHub documentation, 
>>>> ParaView becomes unresponsive. I have also tried running ParaView on the 
>>>> compute node cluster, but I encounter the same issue. The crash 
>>>> specifically occurs when I change the glyph mode to “All Points.”
>>>>
>>>> Do you have any suggestions on how to fix this issue, or can you 
>>>> recommend an alternative software that can handle and visualize millions 
>>>> of 
>>>> particles without crashing?
>>>>
>>>> *Part 2: GRCPrep Files*
>>>>
>>>> I noticed that the DEMdemo_GRCPrep_Part3.cpp file does not contain 
>>>> comments. I would like to better understand its purpose. Currently, the 
>>>> WheelDP executable takes DEMdemo_GRCPrep_Part2 as input. Could someone 
>>>> explain the difference between using Part 2 and Part 3?
>>>>
>>>> Thank you,
>>>> Sumaiya
>>>>
>>>

-- 
You received this message because you are subscribed to the Google Groups 
"ProjectChrono" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion visit 
https://groups.google.com/d/msgid/projectchrono/90003699-ec83-4236-9471-202d656a743an%40googlegroups.com.

Reply via email to