ToPoliNano: NanoMagnet Logic Circuits Design and Simulation - Field-Coupled Nanocomputing: Paradigms, Progress, and Perspectives - page 291

HOLD state are left untouched, while all magnets in RESET state are reset.
In case of magnets belonging to a clock zone that is in the SWITCH state, the
situation is more complex because the calculation of the state of each magnet
requires a specific algorithm.

5.2 Matrix Exploration

For each clock zone where the SWITCH state is active, the simulation matrix
must be scanned and the new value of magnetization of each magnet must be
calculated. The matrix exploration algorithm is based on two nested loops which
scan every column of a zone and for each of them scan every row.

X

012 34

Y

0

1

2

3

4

Fig. 19. Exploration matrix algorithm. Matrix columns are scanned one by one from
left to right, each row of the column is scanned from up to down and then from down
to up.

The behavior of the matrix exploration algorithm is depicted in Fig. 19 .Each
column of the matrix is scanned starting from the left border going toward the
right border. For each column rows are scanned one by one from up to down
and then from down to up. During the matrix visit for the new magnet state is
evaluated following the magnetization algorithm (Sect. 5.3 ). This matrix explo-
ration algorithm is chosen to reduce the usage of if..then...else constructs inside
the programming code of ToPoliNano, to reduce their huge negative impact on
performance when the code is executed in modern superscalar machines.