ERROR: LoadError: ArgumentError: Package Wflow does not have Distributed

[BastianT22]

When I was trying to use Distributed package to accelerate wflow model, this error has occured. I have tried Pkg.instantiate() and Pkg.resolve(), but it still doesn't work. How can I fix this problem or try another way to accelerate wflow? Thank you!

[verseve]

Wflow supports the use of multiple threads, see also;
https://deltares.github.io/Wflow.jl/dev/model_docs/lateral/kinwave/#Multi-Threading
https://docs.julialang.org/en/v1/manual/multi-threading/#Starting-Julia-with-multiple-threads

Distributed is a dependency of the Graphs package, so it should be available. Is Distributed part of the Manifest.toml file?
You could try to run Pkg.update() to check if this resolves the issue.

Note that we do not make use of Distributed in Wflow directly (it is not an explicit dependency).

[BastianT22]

Since calibration of model parameters requires hundreds of simulations, the speed of the model is critical. I am using the wflow_hbv model. After setting nthreads=10, the speed and cpu usage of the model is not significantly improved. So I try to modify the update function in the surface_routing function to parallelize the operation to improve the speed. But I got an error when using the distributed package. I can find distributed in the Manifest.toml file, but not in the Graphs package, and I still can't use distributed after trying Pkg.update().

[verseve]

If you want to use Distibuted as part of your Wflow environment, you have to add this package:

julia> import Pkg
julia> Pkg.add("Distributed")

and then also add it to the Wflow.jl file (using Distributed).