|
# Decomposition
|
|
# Decomposition
|
|
|
|
|
|
Every structure should divide the information that they store across processors, in order to do this two main approach are considered inside OpenFPM.
|
|
Every structure should divide the information that they store across processors, considering that in our context multiple distributed structures can interact with each other and dynamically change, our particular focus is in mapping structures into common domain and decomposition and distribution strategies over this domain. Once this domain is defined two main approach are considered inside OpenFPM.
|
|
|
|
|
|
* Model decomposition:
|
|
* Model decomposition:
|
|
* Domain decomposition
|
|
* Domain decomposition
|
... | @@ -20,8 +20,9 @@ The Fixed decomposition just divide the domain as the user specified and assign |
... | @@ -20,8 +20,9 @@ The Fixed decomposition just divide the domain as the user specified and assign |
|
|
|
|
|
The data driven decomposition instead consider the volume of the information the structure store and divide the volume equally them equally regardless of the communication.
|
|
The data driven decomposition instead consider the volume of the information the structure store and divide the volume equally them equally regardless of the communication.
|
|
|
|
|
|
### Model decomposition
|
|
In the following we will consider structures that map over N-dimensional spaces because at the moment is our main interest
|
|
|
|
|
|
|
|
### Model decomposition
|
|
|
|
|
|
In a distributed-memory setting, where data are scattered across
|
|
In a distributed-memory setting, where data are scattered across
|
|
processors, two factors are important: equal division of work across
|
|
processors, two factors are important: equal division of work across
|
... | | ... | |