[Paraview] Please help to understand the live mode of Catalyst

[Hong Yi]

Hi Andy,

A quick related question to confirm with you: for the Catalyst live mode to work, is pvserver process needed in between for the PV client to communicate with Catalyst enabled simulation? In other words, do I need to run two jobs on HPC, one is Catalyst enabled simulation, the other is pvserver, in order for the client to connect to the Catalyst enabled simulation via sockets? From your message below, it seems pvserver is not needed and I can directly connect a PV client to the Catalyst enabled simulation via sockets, is that right? It’d be great if that is indeed the case since it will simplify the workflow greatly. For one thing, users don’t have to worry about socket communication between two running jobs on HPC.

Thanks,

Hong

From: paraview-bounces at paraview.org [mailto:paraview-bounces at paraview.org] On Behalf Of Andy Bauer
Sent: Monday, October 21, 2013 2:04 PM
To: Junyi Han
Cc: paraview at paraview.org
Subject: Re: [Paraview] Please help to understand the live mode of Catalyst

Hi,
The Catalyst live mode is still beta functionality. The client does connect to the Catalyst enabled simulation directly. You have to keep in mind that there's a pipeline on the server and a separate pipeline on the client. Essentially the pipeline on the client uses the requested filter output from the server and uses that as the source. The pipeline on the server branches of the 'catalyst:' server (instead of the eyeball it will have a circle with an intersecting triangle on the left). You can click on that icon to download that data to your client. You should be able to add in other filters on the Catalyst server by selecting the desired filter to chain off of in the Pipeline Browser and then choosing the desired filter in the menu.
If I remember correctly, the initial pipeline helps set up the environment on the Catalyst server to work with. For big simulations, you wouldn't want to download the full data set to the client anyway.
Regards,
Andy


On Mon, Oct 21, 2013 at 12:50 PM, Junyi Han <demonmerlin at gmail.com<mailto:demonmerlin at gmail.com>> wrote:
Dear all,

I'm new to Catalyst. I'm wondering how does the live mode work.
In my understanding, the simulation is modified to call Catalyst library which can get pre-processed information from a python script file. The simulation also connects to paraview client directly. It means that there is no pvserver, instead, the simulation calls Catalyst library which based on paraview lib and VTK lib to process data like pvserver, is that correct?
But in live mode, if I modify the paraview pipeline through the Paraview client, for example, add a filter, how to inform the simulation that the pipeline has been changed? If the simulation can be noticed of the modified pipeline, why do we need to create a python script to describe the pipeline at the beginning?


Best wishes,
Junyi Han

_______________________________________________
Powered by www.kitware.com<http://www.kitware.com>

Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html

Please keep messages on-topic and check the ParaView Wiki at: http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.paraview.org/pipermail/paraview/attachments/20131021/cb5d3d5c/attachment.htm>