## apbs-users

 [Apbs-users] APBS memory demands From: Gernot Kieseritzky - 2006-03-17 13:46:11 ```Hi all! I have a question regarding the memory demand of APBS. I noticed that APBS consumes roughly 10x as much memory as e.g. Bashford's solvate program. I don't quite understand what this extra memory is spend for. In my test calculation I used a grid setup of 161x161x161 which eats up roughly one 1 GB of memory using APBS. Since on my 64-bit opteron system the size of a double value is 8 bytes, and I need to store the dielectric constant, kappa, the charge and a solution I would get the following minimum memory demand: 8 bytes x 161^3 x 4 = 127.36 MB Ok, I know APBS stores three x-/y-/z-shifted dielectric maps which makes 191 MB. The multigrid algorithm APBS will have to store coarser copies of those data structures, but, since, they are coarsened, I estimate an additional factor of 2: ~400 MB. So, what happens with the remaining 600 MB? Greetings, Gernot ```
 Re: [Apbs-users] APBS memory demands From: Nathan Baker - 2006-03-17 14:00:28 ```Hi Gernot -- You're seeting the ~20 work vectors for the nonlinear & multigrid solver (PMG) plus the 4 that I use for internal storage 20 * 161^3 * 8 =3D 636 MB 4 * 161^3 * 8 =3D 128 MB We can definitely reduce the amount of memory actually allocated for some of the calculations based on the calculation type. We're going to be doing a fair amount of optimization of APBS for speed and memory in the next year so this will be on the list. Thanks! -- Nathan On 3/17/06, Gernot Kieseritzky wrote: > Hi all! > I have a question regarding the memory demand of APBS. I noticed that > APBS consumes roughly 10x as much memory as e.g. Bashford's solvate progr= am. > > I don't quite understand what this extra memory is spend for. In my test > calculation I used a grid setup of 161x161x161 which eats up roughly one > 1 GB of memory using APBS. Since on my 64-bit opteron system the size of > a double value is 8 bytes, and I need to store the dielectric constant, > kappa, the charge and a solution I would get the following minimum > memory demand: > > 8 bytes x 161^3 x 4 =3D 127.36 MB > > Ok, I know APBS stores three x-/y-/z-shifted dielectric maps which makes > 191 MB. The multigrid algorithm APBS will have to store coarser copies > of those data structures, but, since, they are coarsened, I estimate an > additional factor of 2: ~400 MB. > > So, what happens with the remaining 600 MB? > > Greetings, > Gernot > > > ------------------------------------------------------- > This SF.Net email is sponsored by xPML, a groundbreaking scripting langua= ge > that extends applications into web and mobile media. Attend the live webc= ast > and join the prime developer group breaking into this new coding territor= y! > http://sel.as-us.falkag.net/sel?cmd=3Dlnk&kid=3D110944&bid=3D241720&dat= =3D121642 > _______________________________________________ > apbs-users mailing list > apbs-users@... > https://lists.sourceforge.net/lists/listinfo/apbs-users > > -- Assistant Professor, Dept. of Biochemistry and Molecular Biophysics Center for Computational Biology, Washington University in St. Louis Web: http://cholla.wustl.edu/ ```