Changes between Version 4 and Version 5 of experiments


Ignore:
Timestamp:
10/01/08 09:32:02 (10 years ago)
Author:
daley
Comment:

more explanation

Legend:

Unmodified
Added
Removed
Modified
  • experiments

    v4 v5  
    1616 
    1717 
     18t_evo is the total evolution time 
    1819Rows 1,3,4 constitute the weak scaling curve to be. 
    1920Rows 1,2 test strong scaling. 
     
    2122 
    2223 
     24 
    2325Setup syntax: ./setup WD_def -3d +cube16 -maxblocks=70 +noio -auto +pm4dev -objdir=PM4dev -parfile=[specific flash.par] 
    2426Here, "specific flash.par" is: 
    25 scaling5_intrepid.par - Row 1 (4416 procs) & Row 2 (8192 procs) 
    26 scaling5h_intrepid.par - Row 3 (6144 procs) 
    27 scaling6_intrepid.par - Row 4 (8192 procs) 
     27scaling5_intrepid.par - Row 1 (4416 procs) & Row 2 (8192 procs) in table. 
     28scaling5h_intrepid.par - Row 3 (6144 procs) in table. 
     29scaling6_intrepid.par - Row 4 (8192 procs) in table. 
    2830}}} 
    2931 
    30 t_evo is the total evolution time 
     32For the weak scaling we attempt to keep the number of blocks per processor  
     33constant as we vary the number of processors. 
     34 
     35The way to experiment is to use the setup as is, and change the 
     36lrefine_max in flash.par to get to the next problem size. Keep adjusting the number of 
     37processors until you get roughly the same number of blocks per processor. You may have less of a problem 
     38doing this on XT4 since there is lots more memory per processor. Also, eliminate IO 
     39from the runs, that should give you more memory to play around in. 
     40