| | |
Summary: Robust SAT-Based Search Algorithm
for Leakage Power Reduction
Fadi A. Aloula, Soha Hassounb, Karem A. Sakallaha, David Blaauwa
aDepartment of Electrical Engineering and Computer Science
University of Michigan - Ann Arbor
{faloul, karem, blaauw}@eecs.umich.edu
bDepartment of Electrical Engineering and Computer Science
Tufts University
soha@eecs.tufts.edu
Abstract. Leakage current promises to be a major contributor to power dissi-
pation in future technologies. Bounding the maximum and minimum leakage
current poses an important problem. Determining the maximum leakage
ensures that the chip meets power dissipation constraints. Applying an input
pattern that minimizes leakage allows extending battery life when the circuit
is in stand-by mode. Finding such vectors can be expressed as a satisfiability
problem. We apply in this paper an incremental SAT solver, PBS [1], to find
the minimum or maximum leakage current. The solver is called as a post-pro-
cess to a random-vector-generation approach. Our results indicate that using a
such a generic SAT solver can improve on previously proposed random
approaches [7].
|
