| 1 |
A denotes MPGrav and Ac denotes MPGrav_travel
|
| 2 |
Case when i compared to i is not needed as particles dont interact with themselves
|
| 3 |
At step k, interact particle i with particle j= 1 + mod((i+k-1),N)
|
| 4 |
Accumulate force on i due to j in fixed Ai
|
| 5 |
Accumulate negative of this as force on j due to i in circulating Acj
|
| 6 |
At the end of the algorithm, add Ai and Aci
|
| 7 |
In parallel version, Note that Ai will be calculated in "home processor for particle i but Aci will travel around the machine being accumulated in processor holding particle j
|
| 8 |
Thus this violates the owner computes rule and so this parallel algorithm must be implemented by hand -- the compiler will not find it automatically
|
