long computation time large models when going from wflow0.81 to wflow1.02

[aweerts]

Wflow version checks

• I have checked that this issue has not already been reported.

• I have checked that this bug exists on the latest version of Wflow.

Reproducible Example

For WMO report we have models setup for wflow 1.02 that take now very long. This becomes critical for large catchments such as the Danube where calculation went from 2 days to 10-15 days (on WCFi even 44.23 days). Setup of the models and forcing (chunking/compression) is the same.

I put the Danube model here p:\11212698-005-stateofwater2025\pfafstetter_Europe_12000uparea_255468

Current behaviour

For WMO report we have models setup for wflow 1.02 that take now very long. This becomes critical for large catchments such as the Danube where calculation went from 2 days to 10-15 days (on WCFi even 44.23 days). Setup of the models and forcing (chunking/compression) is the same.

Desired behaviour

Calculation times as in version 0.81

Additional Context

No response

[JoostBuitink]

As mentioned via email: this is most likely related to the model configuration itself (river lengths most likely) and not so much due to the different Wflow.jl versions. Once the model configurations are actually comparable, it would still be good to compare the runtimes, but we don't expect slowdowns of this magnitude. Some slowdowns are expected since we had to drop support for LoopVectorization, but that should not be of this magnitude.