An execution plan may be executed on an arbitrary node set, but is typically generated in such a way that it will perform best on a given configuration (for example it might have been created with the intention of being used on {@code X} calculation nodes, each with {@code Y} available execution threads).
|
|
|
|