Grouped vs. ungrouped

Grouping allows you to run incremental ringfence calculations on groups of projects instead of on individual projects (it does not apply to standard ringfencing). This increases the speed of incremental ringfence calculations for large hierarchies because superfluous temporary Result Set iterations are eliminated (instead of one calculation pass per project there is only one pass per group).

Typically this is used when there is a group of projects that represent the "base case". On top of these there are one or more incremental projects. The base group should be treated as an ordinary ringfence with back-allocation whereas the incremental project should receive the incremental ringfence treatment.

To see how it works, consider seven projects: A, B, C, D, E, F and G, divided into three groups: α, β and γ where

α = A + B,

β = C + D + E

γ = F + G

With this grouping, the incremental dependent grouped ringfence will be organized as follows:

α = α

β = (α + β) – α

γ = (α + β + γ) - (α + β)

and calculated as follows:

α = (A + B)

β = (A + B + C + D + E) – (A + B)

γ =(A + B + C + D + E + F + G) – (A + B + C + D + E)

While the incremental independent grouped ringfence will be organized as follows:

α = α

β = (α + β) – α

γ = (α + γ) – α

and calculated as follows:

α = (A + B)

β = (A + B + C + D + E) – (A + B)

γ = (A + B + F + G) – (A + B)