Is your feature request related to a problem? Please describe.
In order complex simulations with large neural networks it is common practice to distribute python code across multiple files. Here as an example files can handle various subclasses, commonly used functions or only parameter values. Currenlty, only a single brain file can be loaded into NEST Server, which also is executed directly.
Describe the solution you'd like
As a user I would like to make use of an additional REST call that only uploads one or multiple "resources" files, which can be either python or yml files into NEST Server. Then with the "exec" call I only want upload the main file and kick of the simulation.
Describe alternatives you've considered
Currently I am just mounting my files directly into NEST server, so that they appear inside the docker container in /opt/data. This has been working for a long time, but might be not possible anymore. Anyhow, this is not a valid solution for using the NEST Server in the framework of the Neurorobotics Platform, where we currently try to run an emobided Million Neuron brain, that is distributed across multiple files. In use with the Neurorobotics Platform all file import/execution needs to be done via REST calls.
Additional context
This feature is necessary to run large brains inside the Neurorobotics Platform with NEST Server on PizDaint supercomputer.
Is your feature request related to a problem? Please describe. In order complex simulations with large neural networks it is common practice to distribute python code across multiple files. Here as an example files can handle various subclasses, commonly used functions or only parameter values. Currenlty, only a single brain file can be loaded into NEST Server, which also is executed directly.
Describe the solution you'd like As a user I would like to make use of an additional REST call that only uploads one or multiple "resources" files, which can be either python or yml files into NEST Server. Then with the "exec" call I only want upload the main file and kick of the simulation.
Describe alternatives you've considered Currently I am just mounting my files directly into NEST server, so that they appear inside the docker container in /opt/data. This has been working for a long time, but might be not possible anymore. Anyhow, this is not a valid solution for using the NEST Server in the framework of the Neurorobotics Platform, where we currently try to run an emobided Million Neuron brain, that is distributed across multiple files. In use with the Neurorobotics Platform all file import/execution needs to be done via REST calls.
Additional context This feature is necessary to run large brains inside the Neurorobotics Platform with NEST Server on PizDaint supercomputer.