Using TeukolskyPointParticleMode with ParallelTable

[bleather]

I asked Niels last week about a problem I have been having with the new Teukolsky package. If, for instance, you wanted to compute a large number of point particle modes and wanted to parallelise this with ParallelTable, I assumed you would have to pass ParallelNeeds[“Teukolsky`”] for this to distribute the definitions to all of the Kernels.

This doesn’t seem to work for me for the latest version of the Teukolsky package on GitHub. I attempted to PacletInstall the new package on each Kernel, but this gives error messages that I cannot access the TeukolskyRadial or ConvolveSource definitions, e.g. Set::shape : Lists {TeukolskyConvolveSourcePrivate[CapitalDelta]tr$2566,TeukolskyConvolveSourcePrivate\[CapitalDelta]\[Phi]r$2566} and {\[CapitalDelta]tr,\[CapitalDelta]\[Phi]r}/. Missing[KeyAbsent,TrajectoryDeltas] are not the same shape.

Is there currently a workaround for this or have I missed something?

[barrywardell]

The attached example notebook uses parallel evaluation with the Teukolsky package. Have you tried that to see if it works? ParallelTeukolskyCalculation.nb.zip

[barrywardell]

There's also another example from the Scattering and Absorption tutorial at ICERM.

[znasipak]

The new Teukolsky package also relies on the latest main branch of KerrGeodesics. So you need to make sure those are both installed and are being distributed to the kernels via ParallelNeeds

[barrywardell]

Closing this, since the example notebook seems to work.