[Paraview] running parallel pvservers

[Jeff Becker]

Hi David and list. I'm still struggling with distributing my dataset 
(bmag) across nodes. My Python Programmable source script below produces 
correct output when run on a single node.

contr = vtk.vtkMultiProcessController.GetGlobalController()
nranks = contr.GetNumberOfProcesses()
rank = contr.GetLocalProcessId()

# read in bmag data and process

executive = self.GetExecutive()
outInfo = executive.GetOutputInformation(0)
updateExtent = [executive.UPDATE_EXTENT().Get(outInfo, i) for i in 
xrange(6)]
imageData = self.GetOutput()
imageData.SetExtent(updateExtent)
myout = ns.numpy_to_vtk(bmag.ravel(),1,vtk.VTK_FLOAT)
myout.SetName("B field magnitude")
imageData.GetPointData().SetScalars(myout)


I then used the Filters/Parallel/Testing/Python/testTransmit.py test 
code as a basis to continue the script to work on multiple nodes. 
Following the last line above, I added:

da = vtk.vtkIntArray()
da.SetNumberOfTuples(6)
if rank == 0:
    ext = imageData.GetExtent()
    for i in range(6):
       da.SetValue(i,ext[i])
contr.Broadcast(da,0)

ext = []
for i in range(6):
    ext.append(da.GetValue(i))
ext = tuple(ext)

tp = vtk.vtkTrivialProducer()
tp.SetOutput(imageData)
tp.SetWholeExtent(ext)

xmit = vtk.vtkTransmitImageDataPiece()
xmit.SetInputConnection(tp.GetOutputPort())
xmit.UpdateInformation()
xmit.SetUpdateExtent(rank, nranks, 0)
xmit.Update()

However, when I run this on two nodes, it looks like both of them get 
the same half of the data, rather than the whole dataset getting split 
between the two. Can anyone see what might be wrong? Thanks.

-jeff

On 07/23/2015 11:18 AM, David E DeMarle wrote:
> Excellent.
>
> No advice yet. Anyone have a worked out example ready? If so, post it 
> to the wiki please.
>
>
> David E DeMarle
> Kitware, Inc.
> R&D Engineer
> 21 Corporate Drive
> Clifton Park, NY 12065-8662
> Phone: 518-881-4909
>
> On Thu, Jul 23, 2015 at 2:14 PM, Jeff Becker 
> <jeffrey.c.becker at nasa.gov <mailto:jeffrey.c.becker at nasa.gov>> wrote:
>
>     Hi David,
>
>     On 07/23/2015 10:57 AM, David E DeMarle wrote:
>>     pyhon shell runs on the client side.
>>
>>     try doing that within the python programmable filter, which runs
>>     on the server side.
>
>     Yes that works. Thanks. Now I will try to use that and follow
>
>     http://www.paraview.org/pipermail/paraview/2011-August/022421.html
>
>     to get my existing Image Data producing Python Programmable Source
>     to distribute the data across the servers. Any additional advice?
>     Thanks again.
>
>     -jeff
>>
>>
>>
>>     David E DeMarle
>>     Kitware, Inc.
>>     R&D Engineer
>>     21 Corporate Drive
>>     Clifton Park, NY 12065-8662
>>     Phone: 518-881-4909 <tel:518-881-4909>
>>
>>     On Thu, Jul 23, 2015 at 1:53 PM, Jeff Becker
>>     <jeffrey.c.becker at nasa.gov <mailto:jeffrey.c.becker at nasa.gov>> wrote:
>>
>>         Hi. I do "mpirun -np 4 pvserver --client-host=xxx
>>         --use-offscreen-rendering", and connect a ParaView client
>>         viewer. I can see 4 nodes in the memory inspector, but when I
>>         start a python shell in ParaView, and do:
>>
>>         from mpi4py import MPI
>>         print MPI.COMM_WORLD.Get_size()
>>
>>         I get the answer 1. Shouldn't it be 4? Thanks.
>>
>>         -jeff
>>
>>
>>         _______________________________________________
>>         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
>>
>>         Search the list archives at:
>>         http://markmail.org/search/?q=ParaView
>>
>>         Follow this link to subscribe/unsubscribe:
>>         http://public.kitware.com/mailman/listinfo/paraview
>>
>>
>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview/attachments/20150805/a1d92521/attachment.html>