| 1 |
This problem is classic SPMD with different phases of computing and communication which the four 4 MPGrav phases further subdividable into subphases which are compute only -- communicate only phases
|
| 2 |
We divide particles equally between processors
-
N Particles
-
Nproc Processors
-
N/Nproc particles per processor
|
| 3 |
As no "locality" in force, all particles are "equal" and it does not matter which particle is placed in which processor
|
| 4 |
Remember each of nine phases is a full loop over all particles in a given processor with local index running from 1 ... N/Nproc
|
| 5 |
We get "automatic" balanced parallel computation in steps 1) 3) 5) 9) with "embarassingly parallel" (i.e. no communication) operation
|
