NanoMagnet Logic: An Architectural Level Overview - Field-Coupled Nanocomputing: Paradigms, Progress, and Perspectives - page 250

This kind of simulation is useful for the investigation of cellular effects of molec-
ular alterations due to pathological conditions or to extend the characterization
of other sub-cellular structures [ 62 ]. Recently a highly parallel implementation
of molecular dynamics simulation was developed for general purpose GPUs in
order to exploit the massive parallelism offered by such devices [ 63 ]. As shown
in this GPU implementation, the massive parallelism is one of the major feature
that can improve the computation in the biotechnology field.
Another biotechnology task that has recently attracted interest from many
interdisciplinary research groups is the study of biological sequences and the
relative development of tools. Biologists study the similarities between proteins
to reconstruct phylogenetic trees and to assess the presence of mutations that
lead to genetic diseases or tumors. Since proteins consist of long sequences of
amino acids, the fastest way to performs a first analysis is to align the studied
protein with the protein coming from huge databases searching for regions of
similarity. After that, the short list of proteins that share a sucient level of
similarity are investigated in detail. There are many alignment algorithms that
are chosen according to the type of analysis that the biologist wants to per-
form. One of the most used is the method developed by Smith and Waterman
that performs, in dynamic programming, the exhaustive local alignment between
two sequences [ 64 ]. We have designed a systolic architecture that accelerate the
Smith-Waterman algorithm execution [ 65 - 67 ] mapping it to NML technology
[ 68 , 69 ] demonstrating how much gain can be obtained from the use of NML
technology.
6 Conclusions
In this chapter we have presented a thorough analysis of the main problems
that arise at architectural level in Field-Coupled devices. The analysis is mainly
based on NML logic but most of the results here presented are valid for all
QCA implementations. For every problem presented a solution is proposed. The
solutions described in this chapter are designed around both logic and technology
constraints. Finally, a brief overview of applications that can fully exploit the
true potential of this technology is presented. This chapter provides researchers
and designers guidelines for the design of complex circuits with this technology,
and provides directions for future development of this technology.
Now it is important to continue the architectural analysis focusing on the
system level integration and all related problems. Important problems must
be investigated, like interfacing with the outside world and the generation of
clock signals. Particularly, in case of NML logic it is important to investigate
how to reduce the necessity of CMOS transistors in the support circuits, like
the input/output interfaces and clock waveforms generators. The necessity to
use CMOS transistors in the external circuits represents a weakness because it
reduces one of the most important advantages of this technology, the immunity
to radiations.