|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ABSTRACT
Pulse-encoded buses, (i.e., in which a transition is encoded as a pulse) have recently emerged as an effective solution to solve crosstalk issues in global interconnects, since they suppress transitions in opposite directions by construction. As a side effect, this also reduces energy, since coupling capacitances in deep-submicron technologies are larger than ground capacitances. Furthermore, a single pulse consumes less energy than a conventional transition, because the limited length does not fully implies extra transitions since all input transitions are encoded with a pulse. This tradeoff leaves the issue of energy efficiency of pulsed buses an open problem, which we address in this work. We present an explorative analysis of the energy efficiency bounds of pulse encoded bus with respect to conventional level-signaled ones. More specifically, we provide a quantitative comparison of the two signaling modes under the same delay constraints, by accounting for the various bus devices (codecs, drivers, and receivers), under (i) the presence of process variations, and (ii) under different input switching activities transmitted on a bus, along with the impact of delay and pulse width. REFERENCES
Note: OCR errors may be found in this Reference List extracted from the full text article. ACM has opted to expose the complete List rather than only correct and linked references.
INDEX TERMS
Primary Classification:
General Terms:
Keywords:
Collaborative Colleagues:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||