fast SIMD parallelized implementation of the Smith-Waterman algorithm

 SSW is a fast implementation of the Smith-Waterman algorithm, which uses the
 Single-Instruction Multiple-Data (SIMD) instructions to parallelize the
 algorithm at the instruction level. SSW library provides an API that can be
 flexibly used by programs written in C, C++ and other languages. The library
 can do protein and genome alignment directly.
 Current version of this implementation is ~50 times faster than an ordinary
 Smith-Waterman. It can return the Smith-Waterman score, alignment location
 and traceback path (cigar) of the optimal alignment accurately; and return
 the sub-optimal alignment score and location heuristically.