Re: [gmx-users] mdrun outputs incorrect resnames

[Tsjerk Wassenaar]

No, the residue names are the those from the .top file. But that's not the
same as the moleculetypes. You have to change the residue names in the [
atoms ] section.

Cheers,

Tsjerk


On Sun, May 26, 2013 at 12:57 AM, Mark Abraham wrote:

> AFAIK, the residue names in the mdrun output .gro file are those of the
> structure file you gave to grompp.
>
> Mark
>
>
> On Sun, May 26, 2013 at 12:31 AM, Reid Van Lehn 
> wrote:
>
> > Hello,
> >
> > I am simulating a lipid bilayer and wish to apply position restraints to
> > only a subset of the lipids in the bilayer. Since position restraints are
> > applied to all molecules of the same molecule type, I defined a new
> > molecule type (DOPR) which is identical to my lipid species (DOPC) by
> > copying the lipid itp file, renaming it and renaming the corresponding
> > molecule type. I then manually edited a starting .gro file to change a
> > subset of the DOPC molecules to DOPR, edited my topology,
> > renumbered/reordered, etc. I also recreated the index file to account for
> > the new molecules so that temperature coupling could be used correctly.
> >
> > Everything seemed ok when I ran the mdrun - grompp didn't complain, the
> > program ran normally, the output trajectory clearly used the correct
> > position restraints, etc. The weird part, though, is that the output .gro
> > file at the end of the simulation only had DOPC molecules in it - the
> DOPR
> > molecules that I had renamed by hand had somehow been output as DOPC
> > instead. Positions, number of atoms, everything else was fine, just the
> > name of the residues was different. I can't figure out why this is
> > happening. It was reproducible across both GMX 4.5.5 / 4.6 and multiple
> > different starting files.
> >
> > It's not a huge issue since the trajectories themselves are fine, I'm
> just
> > worried this issue might indicate other, less obvious problems. A snippet
> > of the topol file is below if that is helpful.
> >
> > Any suggestions / advice would be appreciated!
> >
> > 
> > #include "forcefield.itp
> > #include "dopc.itp"
> >
> > #ifdef POSRES
> > #include "dopc-posre.itp"
> > #endif
> >
> > ; Always include DOPR restraints for restrained lipids
> > #include "dopc_restrained.itp"
> > #include "dopr-posre.itp"
> >
> > #include "spc.itp"
> > #include "ions.itp"
> >
> > [ system ]
> > ; Name
> > frame t= 1.000 in water
> >
> > [ molecules ]
> > ; Compound#mols
> > DOPC 398
> > DOPR   2
> > SOL63882
> > NA   179
> > CL   179
> > *
> >
> > Thanks,
> > Reid
> > --
> > gmx-users mailing listgmx-users@gromacs.org
> > http://lists.gromacs.org/mailman/listinfo/gmx-users
> > * Please search the archive at
> > http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
> > * Please don't post (un)subscribe requests to the list. Use the
> > www interface or send it to gmx-users-requ...@gromacs.org.
> > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> >
> --
> gmx-users mailing listgmx-users@gromacs.org
> http://lists.gromacs.org/mailman/listinfo/gmx-users
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
> * Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-users-requ...@gromacs.org.
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>



-- 
Tsjerk A. Wassenaar, Ph.D.
-- 
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
* Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.
* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

Re: [gmx-users] mdrun outputs incorrect resnames

[Mark Abraham]

AFAIK, the residue names in the mdrun output .gro file are those of the
structure file you gave to grompp.

Mark


On Sun, May 26, 2013 at 12:31 AM, Reid Van Lehn  wrote:

> Hello,
>
> I am simulating a lipid bilayer and wish to apply position restraints to
> only a subset of the lipids in the bilayer. Since position restraints are
> applied to all molecules of the same molecule type, I defined a new
> molecule type (DOPR) which is identical to my lipid species (DOPC) by
> copying the lipid itp file, renaming it and renaming the corresponding
> molecule type. I then manually edited a starting .gro file to change a
> subset of the DOPC molecules to DOPR, edited my topology,
> renumbered/reordered, etc. I also recreated the index file to account for
> the new molecules so that temperature coupling could be used correctly.
>
> Everything seemed ok when I ran the mdrun - grompp didn't complain, the
> program ran normally, the output trajectory clearly used the correct
> position restraints, etc. The weird part, though, is that the output .gro
> file at the end of the simulation only had DOPC molecules in it - the DOPR
> molecules that I had renamed by hand had somehow been output as DOPC
> instead. Positions, number of atoms, everything else was fine, just the
> name of the residues was different. I can't figure out why this is
> happening. It was reproducible across both GMX 4.5.5 / 4.6 and multiple
> different starting files.
>
> It's not a huge issue since the trajectories themselves are fine, I'm just
> worried this issue might indicate other, less obvious problems. A snippet
> of the topol file is below if that is helpful.
>
> Any suggestions / advice would be appreciated!
>
> 
> #include "forcefield.itp
> #include "dopc.itp"
>
> #ifdef POSRES
> #include "dopc-posre.itp"
> #endif
>
> ; Always include DOPR restraints for restrained lipids
> #include "dopc_restrained.itp"
> #include "dopr-posre.itp"
>
> #include "spc.itp"
> #include "ions.itp"
>
> [ system ]
> ; Name
> frame t= 1.000 in water
>
> [ molecules ]
> ; Compound#mols
> DOPC 398
> DOPR   2
> SOL63882
> NA   179
> CL   179
> *
>
> Thanks,
> Reid
> --
> gmx-users mailing listgmx-users@gromacs.org
> http://lists.gromacs.org/mailman/listinfo/gmx-users
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
> * Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-users-requ...@gromacs.org.
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
-- 
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
* Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.
* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

[gmx-users] mdrun outputs incorrect resnames

[Reid Van Lehn]

Hello,

I am simulating a lipid bilayer and wish to apply position restraints to
only a subset of the lipids in the bilayer. Since position restraints are
applied to all molecules of the same molecule type, I defined a new
molecule type (DOPR) which is identical to my lipid species (DOPC) by
copying the lipid itp file, renaming it and renaming the corresponding
molecule type. I then manually edited a starting .gro file to change a
subset of the DOPC molecules to DOPR, edited my topology,
renumbered/reordered, etc. I also recreated the index file to account for
the new molecules so that temperature coupling could be used correctly.

Everything seemed ok when I ran the mdrun - grompp didn't complain, the
program ran normally, the output trajectory clearly used the correct
position restraints, etc. The weird part, though, is that the output .gro
file at the end of the simulation only had DOPC molecules in it - the DOPR
molecules that I had renamed by hand had somehow been output as DOPC
instead. Positions, number of atoms, everything else was fine, just the
name of the residues was different. I can't figure out why this is
happening. It was reproducible across both GMX 4.5.5 / 4.6 and multiple
different starting files.

It's not a huge issue since the trajectories themselves are fine, I'm just
worried this issue might indicate other, less obvious problems. A snippet
of the topol file is below if that is helpful.

Any suggestions / advice would be appreciated!


#include "forcefield.itp
#include "dopc.itp"

#ifdef POSRES
#include "dopc-posre.itp"
#endif

; Always include DOPR restraints for restrained lipids
#include "dopc_restrained.itp"
#include "dopr-posre.itp"

#include "spc.itp"
#include "ions.itp"

[ system ]
; Name
frame t= 1.000 in water

[ molecules ]
; Compound#mols
DOPC 398
DOPR   2
SOL63882
NA   179
CL   179
*

Thanks,
Reid
-- 
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
* Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.
* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

Re: [gmx-users] Why the shape of the molecules changed?

[Dr. Vitaly Chaban]

Of course, you can simulate what you like, but I would personally use an
all-atom representation for long hydrocarbons. I think explicit hydrogens
matter for conformations.


V.V. Chaban




On Thu, May 23, 2013 at 5:16 PM, Bao Kai  wrote:

> Hi, all,
>
> I finally wrote a decane topology file. I used it to some simple
> simulation.
>
> The energy minimization works, while after a simple nvt equilibration,
> the shape of the molecules changed.
>
> Please find the images before and after the nvt equilibration from the
> following links.
> https://www.dropbox.com/s/qcfgipchnus3vcb/before_nvt.png
> https://www.dropbox.com/s/nvbxbs6tofdc0ol/after_nvt.png
>
>
>
> The following is the topology part for the decane molecule.   Please
> help to check if there is anything wrong.   C2 and C3 are CH2 and CH3
> respectively.
>
> [ defaults ]
> ; nbfunccomb-rule   gen-pairs   fudgeLJ fudgeQQ
> 1  3  yes 0.50.5
>
> [ atomtypes ]
> ;   type  at.nummasschargeptypesigmaepsilon
>   C2   6 14.02700.0  A0.3740.4301
>   C3   6 15.035  0.0  A0.3740.6243
> [ moleculetype ]
> ; Namenrexcl
> D10 3
>
> [ atoms ]
> ;   nr   type  resnr residue  atom   cgnr charge   mass
> typeBchargeB  massB
>  1C3  1D10 1C3  1   0.0 15.035
>  ; qtot -0.18
>  2C2  1D10 1C2  2  0.0  14.027
>   ; qtot -0.3
>  3C2  1D10 2C2  3  0.0  14.027
>   ; qtot -0.42
>  4C2  1D10 3C2  4  0.0  14.027
>   ; qtot -0.54
>  5C2  1D10 4C2  5  0.0  14.027
>   ; qtot -0.66
>  6C2  1D10 5C2  6  0.0  14.027
>   ; qtot -0.78
>  7C2  1D10 6C2  7  0.0  14.027
>   ; qtot -0.9
>  8C2  1D10 7C2  8  0.0  14.027
>   ; qtot -1.02
>  9C2  1D10 8C2  9  0.0  14.027
>   ; qtot -1.14
> 10C3  1D10 2C3  10 0.0  15.035
>   ; qtot -1.32
>
>
> [ bonds ]
> ;  aiaj functc0c1c2c3
> 1 2 1  1.53e-01  543  1.53e-01  543
> 2 3 1  1.53e-01  543  1.53e-01  543
> 3 4 1  1.53e-01  543  1.53e-01  543
> 4 5 1  1.53e-01  543  1.53e-01  543
> 5 6 1  1.53e-01  543  1.53e-01  543
> 6 7 1  1.53e-01  543  1.53e-01  543
> 7 8 1  1.53e-01  543  1.53e-01  543
> 8 9 1  1.53e-01  543  1.53e-01  543
> 910 1  1.53e-01  543  1.53e-01  543
>
> [ pairs ]
> ;  aiaj functc0c1c2c3
> 1 4 1
> 2 5 1
> 3 6 1
> 4 7 1
> 5 8 1
> 6 9 1
> 710 1
>
> [ angles ]
> ;  aiajak functc0c1c2
>c3
> 1 2 3 2  1.11e+02  4.00e+02  1.11e+02
>  4.00e+02
> 2 3 4 2  1.11e+02  4.00e+02  1.11e+02
>  4.00e+02
> 3 4 5 2  1.11e+02  4.00e+02  1.11e+02
>  4.00e+02
> 4 5 6 2  1.11e+02  4.00e+02  1.11e+02
>  4.00e+02
> 5 6 7 2  1.11e+02  4.00e+02  1.11e+02
>  4.00e+02
> 6 7 8 2  1.11e+02  4.00e+02  1.11e+02
>  4.00e+02
> 7 8 9 2  1.11e+02  4.00e+02  1.11e+02
>  4.00e+02
> 8 910 2  1.11e+02  4.00e+02  1.11e+02
>  4.00e+02
>
> [ dihedrals ]
> ;  aiajakal functc0c1
> c2c3c4c5
> 1 2 3 4 3  0.00e+00 5.00e+00  3.00e+00
>  0.00e+00 5.00e+00  3.00e+00
> 2 3 4 5 3  0.00e+00 5.00e+00  3.00e+00
>  0.00e+00 5.00e+00  3.00e+00
> 3 4 5 6 3  0.00e+00 5.00e+00  3.00e+00
>  0.00e+00 5.00e+00  3.00e+00
> 4 5 6 7 3  0.00e+00 5.00e+00  3.00e+00
>  0.00e+00 5.00e+00  3.00e+00
> 5 6 7 8 3  0.00e+00 5.00e+00  3.00e+00
>  0.00e+00 5.00e+00  3.00e+00
> 6 7 8 9 3  0.00e+00 5.00e+00  3.00e+00
>  0.00e+00 5.00e+00  3.00e+00
> 7 8 910 3  0.00e+00 5.00e+00  3.00e+00
>  0.00e+00 5.00e+00  3.00e+00
>
>
> [system]
> sparkling water
> [molecules]
> D10 208
>
> Thank you very much.
>
> Best Regards,
> Kai
> --
> gmx-users mailing 

Re: [gmx-users] Re: Have your ever got a real NVE simulation (good energy conservation) in gromacs?

[Szilárd Páll]

With the verlet cutoff scheme (new in 4.6) you get much better control
over the drift caused by (missed) short range interactions; you just
set a maximum allowed target drift and the buffer will be calculated
accordingly. Additionally, with the verlet scheme you are free to
tweak the neighbor search frequency (nstlist) to tune for performance.
As you anyway need a buffer, you should get similar if not better
performance as with the default group scheme with the added benefit of
and far better scalability - I.e. you'll be able to run the same
system on much larger number of cores.

For more details check the manual or the following wiki page:
http://www.gromacs.org/Documentation/Cut-off_schemes

Cheers,
--
Szilárd


On Fri, May 24, 2013 at 5:13 PM, cyberjhon  wrote:
> Hi Guys
>
> Thanks for your reply
>
> 1.  Ok guys I will use the following
>
> Double precision gromacs
>
> dt=1fs
> pbc  = xyz
>
> nstlist  = 5
> rlist= 1.7
> rlist_long   = 2.0
>
> coulombtype  = PME-Switch
> rcoulomb_switch   = 1.2
> rcoulomb= 1.4
>
> fourierspacing   = 0.12
> pme_order= 6
> ewald_rtol   = 1e-5
> ewald_geometry   = 3d
> epsilon_surface  = 0
> optimize_fft = yes
>
> vdw-type = Shift
> rvdw-switch= 0
> rvdw= 1.7
>
> tcoupl   = no
> pcoupl   = no
>
> constraints  = h-bonds
> constraint-algorithm = Lincs
> lincs-order  = 4
> lincs-warnangle  = 30
>
> 2. Any other suggestion ??
>
> Thanks
>
> John Michael
>
>
>
>
>
> --
> View this message in context: 
> http://gromacs.5086.x6.nabble.com/Have-your-ever-got-a-real-NVE-simulation-good-energy-conservation-in-gromacs-tp5008501p5008530.html
> Sent from the GROMACS Users Forum mailing list archive at Nabble.com.
> --
> gmx-users mailing listgmx-users@gromacs.org
> http://lists.gromacs.org/mailman/listinfo/gmx-users
> * Please search the archive at 
> http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
> * Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-users-requ...@gromacs.org.
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
--
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
* Please don't post (un)subscribe requests to the list. Use the
www interface or send it to gmx-users-requ...@gromacs.org.
* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

Aw: Re: Re: [gmx-users] GPU-based workstation

[lloyd riggs]

 

You can also look at profilling on varied web sites, the high end Nvidia run only slightly better than the 2 year old ones, from an individual point not worth the money yet, but if you have the money? as I've been browsing.

 

Also, the sim I did on the cluster was 180-190,000 atoms so the exact same performance the other person had.

 

Stephan


Gesendet: Samstag, 25. Mai 2013 um 15:19 Uhr
Von: "James Starlight" 
An: "Discussion list for GROMACS users" 
Betreff: Re: Aw: Re: [gmx-users] GPU-based workstation

Richard,

thanks for suggestion!

Assuming that I'm using 2 high end GeForce's what performance be better

1) in case of one i7 (4 or 6 nodes ) ?

2) in case of 8 core Xeon like CPU Intel Xeon E5-2650 2.0 GHz / 8core

What properties of MB should take into account primarily for such
Xenon-based system. Does such MBs support multi-GPU ( I noticed that many
such MBs lack for PCI)?

James

2013/5/25 Broadbent, Richard 

> I've been running on my Universities GPU nodes these are one E5-xeon
> (6-cores 12 threads) and have 4 Nvidia 690gtx's. My system is 93 000 atoms
> of DMF under NVE. The performance has been a little disappointing
> ~10ns/day. On my home system using a core i5-2500 and a nvidia 560ti I get
> 5.4ns/day for the same system. On our HPC system using 32 nodes each with 2
> quad-core xeon processors I get 30-40ns/day.
>
> I think that to achieve reasonable performance the system has to be
> balanced between CPU's and GPU's probably getting 2 high end GPU's and a
> top end xeon E5 or core i7 would be a good choice.
>
>
> Richard
>
> From: lloyd riggs >
> Reply-To: Discussion users 
> gmx-users@gromacs.org>>
> Date: Saturday, 25 May 2013 12:02
> To: Discussion users >
> Subject: Aw: Re: [gmx-users] GPU-based workstation
>
> More RAM the better, and the best I have seen is 4 GPU work station. I
> can use/have used 4. The GPU takes 2 slots though, so a 7-8 PCIe board is
> really 3-4 GPU, except the tyan mentioned (there designed as blades so an 8
> or 10 slot board really holds 8 or 10 GPU's). There's cooling problems
> though with GPU's, as on a board there packed, so extra cooling things may
> help not blow a GPU, but I would look for good ones (ask around), as its a
> video game market and they go for looks even though its in casing? The
> external RAM (not onboard GPU RAM) helps if you do a larger sim, but I dont
> know performance wise, the onboard GPU, the more RAM the marrier...so yes,
> normal work stations you can get 4 GPU's for a 300 US$ board, but then the
> price goes way up (3-4000 US$ for an 8-10 gpu board). RAM ordered abroad
> is also cheep, 8 or 16 MB Vs. Shop...I have used 4 GPU's but only on tests
> software, not Gromacs, so would be nice to see performance...for a small
> 100 atom molecule and 500 solvent, using just the CPU I get it to run 5-10
> minutes real for 1 ns sim, but tried simple large 800 amino, 25,000
> solvent eq (NVT or NPT) runs and they clock at around 1 hour real for say
> 50 ps eq's
>
> Stephan
>
> Gesendet: Samstag, 25. Mai 2013 um 07:54 Uhr
> Von: "James Starlight" 
> jmsstarli...@gmail.com>>
> An: "Discussion list for GROMACS users" 
> gmx-users@gromacs.org>>
> Betreff: Re: [gmx-users] GPU-based workstation
> Dear Dr. Watkins!
>
> Thank you for the suggestions!
>
> In the local shops I've found only Core i7 with 6 cores (like Core
> i7-39xx) and 4 cores. Should I obtain much better performance with 6 cores
> than with 4 cores in case of i7 cpu (assuming that I run simulation in
> cpu+gpu mode )?
>
> Also you've mentioned about 4 PCeI MD. Does it means that modern
> work-station could have 4 GPU's in one home-like desktop ? According to my
> current task I suppose that 2 GPU's would be suitable for my simulations
> (assuming that I use typical ASUS MB and 650 Watt power unit). Have
> someone tried to use several GPU's on one workstation ? What attributes of
> MB should be taken into account for best performance on such multi-gpu
> station ?
>
> James
>
> 2013/5/25 lloyd riggs >
>
> > There's also these, but 1 chip runs 6K US, they can get performance up to
> > 2.3 teraflops per chip though double percission...but have no clue about
> > integration with GPU's...Intell also sells their chips on PCIe
> cards...but
> > get only about 350 Gflops, and run 1K US$.
> >
> > http://en.wikipedia.org/wiki/Field-programmable_gate_array and vendor
> > http://www.xilinx.com/
> >
> > They can design them though to fit a PCIe slot and run about the same,
> but
> > still need the board, ram etc...
> >
> > Mostly just to dream about, they say you can order them with radiation
> > shielding as well...so...
> >
> > Stephan Watkins
> >
> > *Gesendet:* Freitag, 24. Mai 2013 um 13:17 Uhr
> > *Von:* "James Starlight" 
> jmsstarli...@gmail.com>>
> > *An:* "Discussion list for GROMACS users" 
> gmx-users@gromacs.org>>
> > *Betreff:* [gmx-users] GPU-based workstation
> > Dear Gromacs Users!
> >
> >
> > I'd like to build new workstation for performing simulation on GPU with

Aw: Re: Re: [gmx-users] GPU-based workstation

[lloyd riggs]

 

Id go for the i7 6 core,

 

To the other message, funny.  I bought ATI's as they clock faster and cost 1/3 the price of Nvidia's but then the software all went to Nvidia.  The new ATI with twice the shaders runs at the same speed (around 1-1.3 terflops ) due to the same problems the Nvidias ran into with IO (or maybe onboard RAM does solve the problem if they went up to 16 or 32 MB)  Gromacs, etc...doesn't run on ATI's, and I've been hoping they, AMD,  catch up, but all I ever see is the constant in 6 months then nothing.

 

I ran around 40 4 ns simulations on University blades with 8 AMD quad cores, using 3 blades I only was able to get 1 ns/day, but never pressed it as far as why so slow, as I needed to finish.  With the Nvidia at  even 5 ns/day I or alot of people could do some really nice work as far as publishing, with raw data in 2 weeks time, so now I feel a bit saddened...

 

I also just found openCL profilling with CUDA 5 that will take any C or C++ software, and mark all sections you need to convert to openCL, but the trial software is 30 day, then 250 US$...

 

Stephan


Gesendet: Samstag, 25. Mai 2013 um 15:19 Uhr
Von: "James Starlight" 
An: "Discussion list for GROMACS users" 
Betreff: Re: Aw: Re: [gmx-users] GPU-based workstation

Richard,

thanks for suggestion!

Assuming that I'm using 2 high end GeForce's what performance be better

1) in case of one i7 (4 or 6 nodes ) ?

2) in case of 8 core Xeon like CPU Intel Xeon E5-2650 2.0 GHz / 8core

What properties of MB should take into account primarily for such
Xenon-based system. Does such MBs support multi-GPU ( I noticed that many
such MBs lack for PCI)?

James

2013/5/25 Broadbent, Richard 

> I've been running on my Universities GPU nodes these are one E5-xeon
> (6-cores 12 threads) and have 4 Nvidia 690gtx's. My system is 93 000 atoms
> of DMF under NVE. The performance has been a little disappointing
> ~10ns/day. On my home system using a core i5-2500 and a nvidia 560ti I get
> 5.4ns/day for the same system. On our HPC system using 32 nodes each with 2
> quad-core xeon processors I get 30-40ns/day.
>
> I think that to achieve reasonable performance the system has to be
> balanced between CPU's and GPU's probably getting 2 high end GPU's and a
> top end xeon E5 or core i7 would be a good choice.
>
>
> Richard
>
> From: lloyd riggs >
> Reply-To: Discussion users 
> gmx-users@gromacs.org>>
> Date: Saturday, 25 May 2013 12:02
> To: Discussion users >
> Subject: Aw: Re: [gmx-users] GPU-based workstation
>
> More RAM the better, and the best I have seen is 4 GPU work station. I
> can use/have used 4. The GPU takes 2 slots though, so a 7-8 PCIe board is
> really 3-4 GPU, except the tyan mentioned (there designed as blades so an 8
> or 10 slot board really holds 8 or 10 GPU's). There's cooling problems
> though with GPU's, as on a board there packed, so extra cooling things may
> help not blow a GPU, but I would look for good ones (ask around), as its a
> video game market and they go for looks even though its in casing? The
> external RAM (not onboard GPU RAM) helps if you do a larger sim, but I dont
> know performance wise, the onboard GPU, the more RAM the marrier...so yes,
> normal work stations you can get 4 GPU's for a 300 US$ board, but then the
> price goes way up (3-4000 US$ for an 8-10 gpu board). RAM ordered abroad
> is also cheep, 8 or 16 MB Vs. Shop...I have used 4 GPU's but only on tests
> software, not Gromacs, so would be nice to see performance...for a small
> 100 atom molecule and 500 solvent, using just the CPU I get it to run 5-10
> minutes real for 1 ns sim, but tried simple large 800 amino, 25,000
> solvent eq (NVT or NPT) runs and they clock at around 1 hour real for say
> 50 ps eq's
>
> Stephan
>
> Gesendet: Samstag, 25. Mai 2013 um 07:54 Uhr
> Von: "James Starlight" 
> jmsstarli...@gmail.com>>
> An: "Discussion list for GROMACS users" 
> gmx-users@gromacs.org>>
> Betreff: Re: [gmx-users] GPU-based workstation
> Dear Dr. Watkins!
>
> Thank you for the suggestions!
>
> In the local shops I've found only Core i7 with 6 cores (like Core
> i7-39xx) and 4 cores. Should I obtain much better performance with 6 cores
> than with 4 cores in case of i7 cpu (assuming that I run simulation in
> cpu+gpu mode )?
>
> Also you've mentioned about 4 PCeI MD. Does it means that modern
> work-station could have 4 GPU's in one home-like desktop ? According to my
> current task I suppose that 2 GPU's would be suitable for my simulations
> (assuming that I use typical ASUS MB and 650 Watt power unit). Have
> someone tried to use several GPU's on one workstation ? What attributes of
> MB should be taken into account for best performance on such multi-gpu
> station ?
>
> James
>
> 2013/5/25 lloyd riggs >
>
> > There's also these, but 1 chip runs 6K US, they can get performance up to
> > 2.3 teraflops per chip though double percission...but have no clue about
> > integration with GPU's...Intell also sells their chips on PCIe
> cards...b

Re: Aw: Re: [gmx-users] GPU-based workstation

[James Starlight]

Richard,

thanks for suggestion!

Assuming that I'm using 2 high end GeForce's what performance be better

1) in case of one i7 (4 or 6 nodes ) ?

2) in case of 8 core Xeon like  CPU Intel Xeon E5-2650 2.0 GHz / 8core

What properties of MB should take into account primarily for such
Xenon-based system. Does such MBs support multi-GPU ( I noticed that many
such MBs lack for PCI)?

James

2013/5/25 Broadbent, Richard 

> I've been running on my Universities GPU nodes these are one E5-xeon
> (6-cores 12 threads)  and have 4 Nvidia 690gtx's. My system is 93 000 atoms
> of DMF under NVE.  The performance has been a little disappointing
> ~10ns/day. On my home system using a core i5-2500 and a nvidia 560ti I get
> 5.4ns/day for the same system. On our HPC system using 32 nodes each with 2
> quad-core xeon processors I get 30-40ns/day.
>
> I think that to achieve reasonable performance the system has to be
> balanced between CPU's and GPU's probably getting 2 high end GPU's and a
> top end xeon E5 or core i7 would be a good choice.
>
>
> Richard
>
> From: lloyd riggs mailto:lloyd.ri...@gmx.ch>>
> Reply-To: Discussion users  gmx-users@gromacs.org>>
> Date: Saturday, 25 May 2013 12:02
> To: Discussion users mailto:gmx-users@gromacs.org>>
> Subject: Aw: Re: [gmx-users] GPU-based workstation
>
> More RAM the better, and the best I have seen is 4 GPU work station.  I
> can use/have used 4.  The GPU takes 2 slots though, so a 7-8 PCIe board is
> really 3-4 GPU, except the tyan mentioned (there designed as blades so an 8
> or 10 slot board really holds 8 or 10 GPU's).  There's cooling problems
> though with GPU's, as on a board there packed, so extra cooling things may
> help not blow a GPU, but I would look for good ones (ask around), as its a
> video game market and they go for looks even though its in casing?  The
> external RAM (not onboard GPU RAM) helps if you do a larger sim, but I dont
> know performance wise, the onboard GPU, the more RAM the marrier...so yes,
> normal work stations you can get 4 GPU's for a 300 US$ board, but then the
> price goes way up (3-4000 US$ for an 8-10 gpu board).  RAM ordered abroad
> is also cheep, 8 or 16 MB Vs. Shop...I have used 4 GPU's but only on tests
> software, not Gromacs, so would be nice to see performance...for a small
> 100 atom molecule and 500 solvent, using just the CPU I get it to run 5-10
> minutes real  for 1 ns sim, but tried simple large 800 amino, 25,000
> solvent eq (NVT or NPT) runs and they clock at around 1 hour real for say
> 50 ps eq's
>
> Stephan
>
> Gesendet: Samstag, 25. Mai 2013 um 07:54 Uhr
> Von: "James Starlight"  jmsstarli...@gmail.com>>
> An: "Discussion list for GROMACS users"  gmx-users@gromacs.org>>
> Betreff: Re: [gmx-users] GPU-based workstation
> Dear Dr. Watkins!
>
> Thank you for the suggestions!
>
> In the local shops I've found only Core i7 with 6 cores (like Core
> i7-39xx) and 4 cores. Should I obtain much better performance with 6 cores
> than with 4 cores in case of i7 cpu (assuming that I run simulation in
> cpu+gpu mode )?
>
> Also you've mentioned about 4 PCeI MD. Does it means that modern
> work-station could have 4 GPU's in one home-like desktop ? According to my
> current task I suppose that 2 GPU's would be suitable for my simulations
> (assuming that I use typical ASUS MB and 650 Watt power unit). Have
> someone tried to use several GPU's on one workstation ? What attributes of
> MB should be taken into account for best performance on such multi-gpu
> station ?
>
> James
>
> 2013/5/25 lloyd riggs mailto:lloyd.ri...@gmx.ch>>
>
> > There's also these, but 1 chip runs 6K US, they can get performance up to
> > 2.3 teraflops per chip though double percission...but have no clue about
> > integration with GPU's...Intell also sells their chips on PCIe
> cards...but
> > get only about 350 Gflops, and run 1K US$.
> >
> > http://en.wikipedia.org/wiki/Field-programmable_gate_array and vendor
> > http://www.xilinx.com/
> >
> > They can design them though to fit a PCIe slot and run about the same,
> but
> > still need the board, ram etc...
> >
> > Mostly just to dream about, they say you can order them with radiation
> > shielding as well...so...
> >
> > Stephan Watkins
> >
> > *Gesendet:* Freitag, 24. Mai 2013 um 13:17 Uhr
> > *Von:* "James Starlight"  jmsstarli...@gmail.com>>
> > *An:* "Discussion list for GROMACS users"  gmx-users@gromacs.org>>
> > *Betreff:* [gmx-users] GPU-based workstation
> > Dear Gromacs Users!
> >
> >
> > I'd like to build new workstation for performing simulation on GPU with
> > Gromacs 4.6 native cuda support.
> > Recently I've used such setup with Core i5 cpu and nvidia 670 GTX video
> > and obtain good performance ( ~ 20 ns\day for typical 60.000 atom system
> > with SD integrator)
> >
> >
> > Now I'd like to build multi-gpu wokstation.
> >
> > My question - How much GPU would give me best performance on the typical
> > home-like workstation. What algorithm of Ncidia GPU integration should I
> > use (e.g

Re: Aw: Re: [gmx-users] GPU-based workstation

[Broadbent, Richard]

I've been running on my Universities GPU nodes these are one E5-xeon (6-cores 
12 threads)  and have 4 Nvidia 690gtx's. My system is 93 000 atoms of DMF under 
NVE.  The performance has been a little disappointing ~10ns/day. On my home 
system using a core i5-2500 and a nvidia 560ti I get 5.4ns/day for the same 
system. On our HPC system using 32 nodes each with 2 quad-core xeon processors 
I get 30-40ns/day.

I think that to achieve reasonable performance the system has to be balanced 
between CPU's and GPU's probably getting 2 high end GPU's and a top end xeon E5 
or core i7 would be a good choice.


Richard

From: lloyd riggs mailto:lloyd.ri...@gmx.ch>>
Reply-To: Discussion users mailto:gmx-users@gromacs.org>>
Date: Saturday, 25 May 2013 12:02
To: Discussion users mailto:gmx-users@gromacs.org>>
Subject: Aw: Re: [gmx-users] GPU-based workstation

More RAM the better, and the best I have seen is 4 GPU work station.  I can 
use/have used 4.  The GPU takes 2 slots though, so a 7-8 PCIe board is really 
3-4 GPU, except the tyan mentioned (there designed as blades so an 8 or 10 slot 
board really holds 8 or 10 GPU's).  There's cooling problems though with GPU's, 
as on a board there packed, so extra cooling things may help not blow a GPU, 
but I would look for good ones (ask around), as its a video game market and 
they go for looks even though its in casing?  The external RAM (not onboard GPU 
RAM) helps if you do a larger sim, but I dont know performance wise, the 
onboard GPU, the more RAM the marrier...so yes, normal work stations you can 
get 4 GPU's for a 300 US$ board, but then the price goes way up (3-4000 US$ for 
an 8-10 gpu board).  RAM ordered abroad is also cheep, 8 or 16 MB Vs. Shop...I 
have used 4 GPU's but only on tests software, not Gromacs, so would be nice to 
see performance...for a small 100 atom molecule and 500 solvent, using just the 
CPU I get it to run 5-10 minutes real  for 1 ns sim, but tried simple large 800 
amino, 25,000 solvent eq (NVT or NPT) runs and they clock at around 1 hour real 
for say 50 ps eq's

Stephan

Gesendet: Samstag, 25. Mai 2013 um 07:54 Uhr
Von: "James Starlight" mailto:jmsstarli...@gmail.com>>
An: "Discussion list for GROMACS users" 
mailto:gmx-users@gromacs.org>>
Betreff: Re: [gmx-users] GPU-based workstation
Dear Dr. Watkins!

Thank you for the suggestions!

In the local shops I've found only Core i7 with 6 cores (like Core
i7-39xx) and 4 cores. Should I obtain much better performance with 6 cores
than with 4 cores in case of i7 cpu (assuming that I run simulation in
cpu+gpu mode )?

Also you've mentioned about 4 PCeI MD. Does it means that modern
work-station could have 4 GPU's in one home-like desktop ? According to my
current task I suppose that 2 GPU's would be suitable for my simulations
(assuming that I use typical ASUS MB and 650 Watt power unit). Have
someone tried to use several GPU's on one workstation ? What attributes of
MB should be taken into account for best performance on such multi-gpu
station ?

James

2013/5/25 lloyd riggs mailto:lloyd.ri...@gmx.ch>>

> There's also these, but 1 chip runs 6K US, they can get performance up to
> 2.3 teraflops per chip though double percission...but have no clue about
> integration with GPU's...Intell also sells their chips on PCIe cards...but
> get only about 350 Gflops, and run 1K US$.
>
> http://en.wikipedia.org/wiki/Field-programmable_gate_array and vendor
> http://www.xilinx.com/
>
> They can design them though to fit a PCIe slot and run about the same, but
> still need the board, ram etc...
>
> Mostly just to dream about, they say you can order them with radiation
> shielding as well...so...
>
> Stephan Watkins
>
> *Gesendet:* Freitag, 24. Mai 2013 um 13:17 Uhr
> *Von:* "James Starlight" 
> mailto:jmsstarli...@gmail.com>>
> *An:* "Discussion list for GROMACS users" 
> mailto:gmx-users@gromacs.org>>
> *Betreff:* [gmx-users] GPU-based workstation
> Dear Gromacs Users!
>
>
> I'd like to build new workstation for performing simulation on GPU with
> Gromacs 4.6 native cuda support.
> Recently I've used such setup with Core i5 cpu and nvidia 670 GTX video
> and obtain good performance ( ~ 20 ns\day for typical 60.000 atom system
> with SD integrator)
>
>
> Now I'd like to build multi-gpu wokstation.
>
> My question - How much GPU would give me best performance on the typical
> home-like workstation. What algorithm of Ncidia GPU integration should I
> use (e.g SLI etc) ?
>
>
> Thanks for help,
>
>
> James
> --
> gmx-users mailing list gmx-users@gromacs.org
> http://lists.gromacs.org/mailman/listinfo/gmx-users
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
> * Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to 
> gmx-users-requ...@gromacs.org.
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
> --
> gmx-users mailing list gm

Aw: Re: [gmx-users] GPU-based workstation

[lloyd riggs]

More RAM the better, and the best I have seen is 4 GPU work station.  I can use/have used 4.  The GPU takes 2 slots though, so a 7-8 PCIe board is really 3-4 GPU, except the tyan mentioned (there designed as blades so an 8 or 10 slot board really holds 8 or 10 GPU's).  There's cooling problems though with GPU's, as on a board there packed, so extra cooling things may help not blow a GPU, but I would look for good ones (ask around), as its a video game market and they go for looks even though its in casing?  The external RAM (not onboard GPU RAM) helps if you do a larger sim, but I dont know performance wise, the onboard GPU, the more RAM the marrier...so yes, normal work stations you can get 4 GPU's for a 300 US$ board, but then the price goes way up (3-4000 US$ for an 8-10 gpu board).  RAM ordered abroad is also cheep, 8 or 16 MB Vs. Shop...I have used 4 GPU's but only on tests software, not Gromacs, so would be nice to see performance...for a small 100 atom molecule and 500 solvent, using just the CPU I get it to run 5-10 minutes real  for 1 ns sim, but tried simple large 800 amino, 25,000 solvent eq (NVT or NPT) runs and they clock at around 1 hour real for say 50 ps eq's

 

Stephan

 

Gesendet: Samstag, 25. Mai 2013 um 07:54 Uhr
Von: "James Starlight" 
An: "Discussion list for GROMACS users" 
Betreff: Re: [gmx-users] GPU-based workstation

Dear Dr. Watkins!

Thank you for the suggestions!

In the local shops I've found only Core i7 with 6 cores (like Core
i7-39xx) and 4 cores. Should I obtain much better performance with 6 cores
than with 4 cores in case of i7 cpu (assuming that I run simulation in
cpu+gpu mode )?

Also you've mentioned about 4 PCeI MD. Does it means that modern
work-station could have 4 GPU's in one home-like desktop ? According to my
current task I suppose that 2 GPU's would be suitable for my simulations
(assuming that I use typical ASUS MB and 650 Watt power unit). Have
someone tried to use several GPU's on one workstation ? What attributes of
MB should be taken into account for best performance on such multi-gpu
station ?

James

2013/5/25 lloyd riggs 

> There's also these, but 1 chip runs 6K US, they can get performance up to
> 2.3 teraflops per chip though double percission...but have no clue about
> integration with GPU's...Intell also sells their chips on PCIe cards...but
> get only about 350 Gflops, and run 1K US$.
>
> http://en.wikipedia.org/wiki/Field-programmable_gate_array and vendor
> http://www.xilinx.com/
>
> They can design them though to fit a PCIe slot and run about the same, but
> still need the board, ram etc...
>
> Mostly just to dream about, they say you can order them with radiation
> shielding as well...so...
>
> Stephan Watkins
>
> *Gesendet:* Freitag, 24. Mai 2013 um 13:17 Uhr
> *Von:* "James Starlight" 
> *An:* "Discussion list for GROMACS users" 
> *Betreff:* [gmx-users] GPU-based workstation
> Dear Gromacs Users!
>
>
> I'd like to build new workstation for performing simulation on GPU with
> Gromacs 4.6 native cuda support.
> Recently I've used such setup with Core i5 cpu and nvidia 670 GTX video
> and obtain good performance ( ~ 20 ns\day for typical 60.000 atom system
> with SD integrator)
>
>
> Now I'd like to build multi-gpu wokstation.
>
> My question - How much GPU would give me best performance on the typical
> home-like workstation. What algorithm of Ncidia GPU integration should I
> use (e.g SLI etc) ?
>
>
> Thanks for help,
>
>
> James
> --
> gmx-users mailing list gmx-users@gromacs.org
> http://lists.gromacs.org/mailman/listinfo/gmx-users
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
> * Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-users-requ...@gromacs.org.
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
> --
> gmx-users mailing list gmx-users@gromacs.org
> http://lists.gromacs.org/mailman/listinfo/gmx-users
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
> * Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-users-requ...@gromacs.org.
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
--
gmx-users mailing list gmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
* Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
* Please don't post (un)subscribe requests to the list. Use the
www interface or send it to gmx-users-requ...@gromacs.org.
* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists



-- 
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
* Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.
* Can'

Re: [gmx-users] GPU-based workstation

[James Starlight]

Dear Dr. Watkins!

Thank you for the suggestions!

In the local shops I've found only Core i7 with 6 cores (like  Core
i7-39xx) and 4 cores.  Should I obtain much better performance with 6 cores
than with 4 cores in case of i7 cpu (assuming that I run simulation in
cpu+gpu mode )?

Also you've mentioned about 4 PCeI MD. Does it means that modern
work-station could have 4 GPU's in one home-like desktop ? According to my
current task I suppose that 2 GPU's would be suitable for my simulations
(assuming that I use typical ASUS MB  and 650 Watt power unit). Have
someone tried to use several GPU's on one workstation ? What attributes of
MB should be taken into account for best performance on such multi-gpu
station ?

James

2013/5/25 lloyd riggs 

> There's also these, but 1 chip runs 6K US, they can get performance up to
> 2.3 teraflops per chip though double percission...but have no clue about
> integration with GPU's...Intell also sells their chips on PCIe cards...but
> get only about 350 Gflops, and run 1K US$.
>
> http://en.wikipedia.org/wiki/Field-programmable_gate_array and vendor
> http://www.xilinx.com/
>
> They can design them though to fit a PCIe slot and run about the same, but
> still need the board, ram etc...
>
> Mostly just to dream about, they say you can order them with radiation
> shielding as well...so...
>
> Stephan Watkins
>
> *Gesendet:* Freitag, 24. Mai 2013 um 13:17 Uhr
> *Von:* "James Starlight" 
> *An:* "Discussion list for GROMACS users" 
> *Betreff:* [gmx-users] GPU-based workstation
> Dear Gromacs Users!
>
>
> I'd like to build new workstation for performing simulation on GPU with
> Gromacs 4.6 native cuda support.
> Recently I've used such setup with Core i5 cpu and nvidia 670 GTX video
> and obtain good performance ( ~ 20 ns\day for typical 60.000 atom system
> with SD integrator)
>
>
> Now I'd like to build multi-gpu wokstation.
>
> My question - How much GPU would give me best performance on the typical
> home-like workstation. What algorithm of Ncidia GPU integration should I
> use (e.g SLI etc) ?
>
>
> Thanks for help,
>
>
> James
> --
> gmx-users mailing list gmx-users@gromacs.org
> http://lists.gromacs.org/mailman/listinfo/gmx-users
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
> * Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-users-requ...@gromacs.org.
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
> --
> gmx-users mailing listgmx-users@gromacs.org
> http://lists.gromacs.org/mailman/listinfo/gmx-users
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
> * Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-users-requ...@gromacs.org.
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
-- 
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
* Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.
* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists