Exchange data between servers

In this example, two DPF servers are started, and a workflow is created with a part on both servers. This example shows how you can read data from a given machine and transform this data on another machine without any more difficulties than working on a local computer.

Note

This example requires the Premium ServerContext. For more information, see Server context.

from ansys.dpf import core as dpf
from ansys.dpf.core import examples
from ansys.dpf.core import operators as ops


dpf.set_default_server_context(dpf.AvailableServerContexts.premium)

Create two servers

Use the start_local_server() method to start two servers on your local machine. If you have another server, you can use the connect_to_server() method to connect to any DPF server on your network.

The as_global attributes allows you to choose whether a server is stored by the module and used by default. This example sets the first server as the default.

server1 = dpf.start_local_server(as_global=True, config=dpf.AvailableServerConfigs.GrpcServer)
server2 = dpf.start_local_server(as_global=False, config=dpf.AvailableServerConfigs.GrpcServer)

# Check that the two servers are listening on different ports.
print(
    server1.port if hasattr(server1, "port") else "",
    server2.port if hasattr(server2, "port") else "",
)

50054 50055

Send the result file

The result file is sent to the temporary directory of the first server. This file upload is useless in this case because the two servers are local machines.

file_path_in_tmp = examples.find_complex_rst(server=server1)

Create a workflow on the first server

Create the model

model = dpf.Model(file_path_in_tmp)

# Read displacement
disp = model.results.displacement()
disp.inputs.time_scoping(len(model.metadata.time_freq_support.time_frequencies))

Create a workflow on the second server

# Change the Cartesian coordinates to cylindrical coordinates cs
coordinates = ops.geo.rotate_in_cylindrical_cs_fc(server=server2)

# Create the Cartesian coordinate cs
cs = dpf.fields_factory.create_scalar_field(12, server=server2)
cs.data = [1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0]

coordinates.inputs.coordinate_system(cs)

# Choose the radial component to plot
comp = dpf.operators.logic.component_selector_fc(coordinates, 0, server=server2)

Pass data from one server to another

fc_disp = disp.outputs.fields_container()
fc_copy = fc_disp.deep_copy(server=server2)

mesh_copy = model.metadata.meshed_region.deep_copy(server=server2)
# give a mesh to the field
fc_copy[0].meshed_region = mesh_copy
fc_copy[1].meshed_region = mesh_copy

coordinates.inputs.field(fc_copy)

Plot the output

out = comp.outputs.fields_container()

# real part
mesh_copy.plot(out.get_field({"complex": 0}))

# imaginary part
mesh_copy.plot(out.get_field({"complex": 1}))

• 
•