Re: [AMBER] Fwd: increase speed in amber using implcit solvent model

From: Carlos Simmerling <carlos.simmerling.gmail.com>
Date: Sun, 24 Jun 2018 07:18:56 -0400

Compare your timings to the benchmarks on the amber web site. Gb
calculations for large proteins on cpus can be slow. The comparison will
let you know if your timings are expected, and also show you computers that
may be faster (such as GPUs).

On Sun, Jun 24, 2018, 3:27 AM Chhaya Singh <chhayasingh014.gmail.com> wrote:

> ---------- Forwarded message ----------
> From: Chhaya Singh <chhayasingh014.gmail.com>
> Date: 24 June 2018 at 12:47
> Subject: increase speed in amber using implcit solvent model
> To: amber.ambermd.org
>
>
> Hi,
> I am trying to perform a simulation having a protein using implicit solvent
> model using force field ff14sbonlysc with igb = 8.
> I am getting a very low speed using 2 nodes. the speed i get now is less a
> ns/ day.
> is there a way to increase speed in amber.
> please let me know.
> my min.in file has :
> Stage 1 - minimisation of fibril
> &cntrl
> imin=1, maxcyc=1000, ncyc=500,
> cut=999., rgbmax=999.,igb=8, ntb=0,
> ntpr=100
> /
>
> my heat1.out file has:
> Final Performance Info:
> | -----------------------------------------------------
> | Average timings for last 50 steps:
> | Elapsed(s) = 17.41 Per Step(ms) = 348.14
> | ns/day = 0.12 seconds/ns = 696286.68
> |
> | Average timings for all steps:
> | Elapsed(s) = 3487.17 Per Step(ms) = 348.72
> | ns/day = 0.12 seconds/ns = 697434.94
> | -----------------------------------------------------
>
> | Master Setup CPU time: 0.26 seconds
> | Master NonSetup CPU time: 3461.87 seconds
> | Master Total CPU time: 3462.13 seconds 0.96 hours
>
> | Master Setup wall time: 1 seconds
> | Master NonSetup wall time: 3487 seconds
> | Master Total wall time: 3488 seconds 0.97 hours
>
>
>
>
>
> can you please help. how can I increase the speed using implicit solvent in
> amber.
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Sun Jun 24 2018 - 04:30:03 PDT
Custom Search