Best way to run a sensitivity analysis for an ODE model on GPUs

Hi.

First, I am not trying to parallelize ode45 (or any other solver). I have seen this question asked a lot and I am clear why it cannot be done.

What I want to do is essentially run a sensitivity analysis on a GPU. I have a system of differential equations that depends on parameters and I need to vary those parameters across certain ranges. On the CPU, I can easily use parfor but what is the smoothest way to implement this with gpuArray?

Again, I understand the potential speed-up will depend on my problem and GPU but I am curious and would like to give this a try. Any insight or relevant example is appreciated.

Best,

Francesco

[Wang Hong]

Now I face the same problem, did you get the answer for it?

[Vipul Singhal]

Just want to echo this question. Would love to see if someone has made progress in this direction. I think doing things like parameter estimation / sensitivity analysis would be greatly accelerated if we could figure out how to run an ODE on a GPU.

[Tom Jones]

I would also like to be able to run an ODE solver on a GPU. Having the equivalent function to a parfor, but which can be executed on a GPU would be a huge help in running sensitivity analyses for my research on non-linear systems.