Parallella Community

Hello,

I'm new in this area (molecular dynamics simulation, programming, Parallella etc) and I would like to know if Parallella is recommended to MD simulations (using GROMACS, for example).

Thank you for your attention (and sorry if I posted in the wrong area).

Hi Gustavo-

Being somewhat familiar with this area with my own research, I don't think parallella would be ideal for something like GROMACS. It's certainly possible, but there's also the matter of adapting the code to use the Epiphany as an accelerator. As far as I know, there's isn't a MM or MD software which has even touched the idea of using the Parallella.

Just thinking about it, you can probably get better performance out of GROMACS using commodity desktop hardware than multiple Parallellas, and for less money. But if you're doing serious MD simulations, you might as well just get connected with supercomputing center.

That said, if you want to port GROMACS to the Parallella platform, be my guest. You might just prove me wrong.

Opti

Hi, optimaler.

Thank you for the help. Now I have just two questions.

First: If the GROMACS code be adapted to use the Epiphany as an accelerator, can Parallella be a better alternative than these supercomputing centers in the future?

Second: I'm not doing serious MD simulations yet, because I'm trying to learn how to do it before advancing. Is there any hardware that you recommend to my learning case?

Thank you again.