Re: Re: slow NAMD simulations on a AMD athlon64 in SUSE linux

From: Fam Eklund (areklund_at_yahoo.se)
Date: Thu Jun 15 2006 - 02:19:34 CDT

When I try two namd simulations (in order to get an extra load on the cpu) the wall time (clock) drops and approaches the cpu time ..

-----------------------------------------------------------------------------------------
 WRITING COORDINATES TO DCD FILE AT STEP 251500
WRITING COORDINATES TO DCD FILE AT STEP 251600
WRITING COORDINATES TO DCD FILE AT STEP 251700
WRITING COORDINATES TO DCD FILE AT STEP 251800
WRITING COORDINATES TO DCD FILE AT STEP 251900
TIMING: 252000 CPU: 1724.01, 0.86291/step Wall: 2623.38, 1.23378/step, 0 hours remaining, 209549 kB of memory in use.
ENERGY: 252000 32.2695 100.0721 101.3504 0.0000 -59688.1245 -2718.9017 0.0000 0.0000 10082.3029 -52091.0313 302.3897 -52069.5866 -52071.4109 301.0572 95.3632 174.6694 115542.0837 2.5037 2.2962

WRITING EXTENDED SYSTEM TO RESTART FILE AT STEP 252000
WRITING COORDINATES TO DCD FILE AT STEP 252000
WRITING COORDINATES TO RESTART FILE AT STEP 252000
FINISHED WRITING RESTART COORDINATES
WRITING VELOCITIES TO RESTART FILE AT STEP 252000
FINISHED WRITING RESTART VELOCITIES
WRITING EXTENDED SYSTEM TO OUTPUT FILE AT STEP 252000
WRITING COORDINATES TO OUTPUT FILE AT STEP 252000
CLOSING COORDINATE DCD FILE
WRITING VELOCITIES TO OUTPUT FILE AT STEP 252000
==========================================
WallClock: 2624.351074 CPUTime: 1724.499756 Memory: 209419 kB
Program finished.
-----------------------------------------------------------------------------------------------

compared to the system without any load

-----------------------------------------------------------------------------------------------
WRITING COORDINATES TO DCD FILE AT STEP 250600
WRITING COORDINATES TO DCD FILE AT STEP 250700
WRITING COORDINATES TO DCD FILE AT STEP 250800
WRITING COORDINATES TO DCD FILE AT STEP 250900
TIMING: 251000 CPU: 882.651, 0.890888/step Wall: 6591.91, 6.02607/step, 1.67391 hours remaining, 205083 kB of memory in use.
ENERGY: 251000 24.0551 86.1352 96.3483 0.0000 -59801.6600 -2750.7151 0.0000 0.0000 9971.6492 -52374.1872 299.0710 -52353.7634 -52353.2645 300.0882 285.0440 349.3006 115043.4255 3.9893 4.3952

WRITING COORDINATES TO DCD FILE AT STEP 251000
roberte_at_c-8d44e455:~/download/namd_filer> tail test33.out
WRITING COORDINATES TO RESTART FILE AT STEP 252000
FINISHED WRITING RESTART COORDINATES
WRITING VELOCITIES TO RESTART FILE AT STEP 252000
FINISHED WRITING RESTART VELOCITIES
WRITING EXTENDED SYSTEM TO OUTPUT FILE AT STEP 252000
WRITING COORDINATES TO OUTPUT FILE AT STEP 252000
CLOSING COORDINATE DCD FILE
WRITING VELOCITIES TO OUTPUT FILE AT STEP 252000
==========================================
WallClock: 8818.390625 CPUTime: 1803.148682 Memory: 209523 kB
[1]+ Done ../NAMD_2.6b1_Linux-amd64/charmrun ++local ../NAMD_2.6b1_Linux-amd64/namd2 okta_2_dyn1_cc.conf >test33.out
-------------------------------------------------------------------------------------------------------------

Note the difference which is rahter large.......

Which of the times should be taken as the actual computation time ???

Sincerely
Robert Eklund PhD Organic chemistry. Stockholm University.

Jim Phillips <jim_at_ks.uiuc.edu> skrev: Hi,

The reported wallclock time for NAMD 2.6b1 and earlier is confused by
machines with variable clockspeeds. (It checks the CPU speed at startup,
and then multiplies by the number of cycles to get walltime cheaply.)
This is fixed in newer versions of Charm++ (and hence NAMD). A workaround
is to have some kind of load on the machine when starting the benchmark,
or to just multiply by the ratio of the resting and loaded clockspeeds.

-Jim

On Wed, 14 Jun 2006, Fam Eklund wrote:

> Hello.
> I have resently upgraded my computer to an athlon64 +3500 1GB ram, using an ASUS A8n5x motherboard. And I am trying to run NAMD in linux mode. I have installed the new SUSE 10.1 (for 64 bit computers).
> I have both tried to run NAMD from the 64bit binary that one can download from their homepage, as well as from the compiled one from source, and get almost the same result.
> The program starts and runs pretty smoothly however when I check the benchmark for the run I get the following :
> CPU TIME 0.87s/step WALL time 1.99s/step
>
> Why is the total time more then twice the cpu time, is the computer .
>
> (I have also tried to run the same simulation on my windows partition using the windows binary (32bit?) from the NAMD homepage, and run it on windows XP home edition. I get the following results CPU TIME is 1.3 s/step and wall time is roughly the same.)
>
> Shouldn't the linux version be faster then the windows version ???...
>
> If someone has encountered the same problem, or has any idea about what could slow the process down please let me know. Because if we are going to expand the system and connect more computers we want to know if we should continue with the athlon64 type or if it is this slow try something else.
>
> Thanks
> Robert Eklund PhD Organic chemistry Stockholm University.
>
>

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:42:11 CST