Module Num | Param Num | Design Param |
0 | 0 | Batch | 0 | 1 | LineDef | 0 | 2 | WordDef | 0 | 3 | Shift | 0 | 4 | Output | 0 | 5 | Sort | 0 | 6 | Sequence | 1 | 0 | ReadLines | 1 | 1 | Store | 1 | 2 | Format1 | 1 | 3 | Index1 | 1 | 4 | Batch | 2 | 0 | Load/Store | 2 | 1 | Format1 | 2 | 2 | Index1 | 2 | 3 | Index2 | 2 | 4 | PairStore | 3 | 0 | Format1 | 3 | 1 | Index1 | 3 | 2 | Format2 | 3 | 3 | Index2 | 3 | 4 | Index3 | 3 | 5 | Batch | 4 | 0 | Format1 | 4 | 1 | Format2 | 4 | 2 | Index1 | 4 | 3 | Index2 | 4 | 4 | Index3 | 4 | 5 | Batch | 5 | 0 | Sequence | 5 | 1 | Batch |
The reason "might" is used in that the actual change will be determined by a probability distribution, so we needed to analyse the possible connectivity in the decomposition. For example, in looking at the Input Module, if there was a requirement to alter the Index1 design parameter, what other modules might be affected? Clearly, any module which relies on Index1, might be affected, and so dependencies would be established from (1, 3) to {(2,2), (3,1), (4,2)}.
In addition, it might be possible for a change in a design parameter to affect a functional requirement. For example, if something in Shift Module precluded the use of batch mode (admittedly, hard to imagine), this would cause a change to the functional requirement (0,0), and all other modules which depend on batch mode.
Using this type of analysis the following connectivity set was produced:
Source | Destination Set |
(0,0) | { (0,6), (1,1), (1,2), (1,4), (2,0), (2,1), (2,4), (3,5), (4,0), (5,1) } |
(0,1) | { (1,0), (1,1), (1,3), (2,2), (3,1), (4,2) }) |
(0,2) | { (1,3), (3,3), (4,3) } |
(0,3) | { (2,3), (3,3), (4,3) } |
(0,4) | { (0,0), (1,4), (2,4), (3,5), (4,4), (4,5), (5,1) } |
(0,5) | { (0,0), (1,4), (2,4), (3,4), (3,5), (4,5) } |
(0,6) | { (5,0) } |
Last modified: Tue Dec 15 16:47:39 1998