SciML / DiffEqGPU.jl

GPU-acceleration routines for DifferentialEquations.jl and the broader SciML scientific machine learning ecosystem
https://docs.sciml.ai/DiffEqGPU/stable/
MIT License
274 stars 28 forks source link

Out of dynamic GPU memory in EnsembleGPUKernel for higher number of threads #206

Closed utkarsh530 closed 1 year ago

utkarsh530 commented 1 year ago

Due to new feature integrations in EnsembleGPUKernel, currently, out-of-dynamic GPU memory error might occur. Some optimizations need to be done to work on higher trajectory counts (>1000). This issue is known and being worked on.

If you need to use EnsembleGPUKernel with a relatively higher number of threads, please use v1.20.0 of DIffEqGPU.jl.