Performance (in millions of cells per second) of the various Smith-Waterman implementations, including a regular implementation (not vectorized), Wozniak's diagonal implementation with memory lookups, Farrar's method and our diagonal approach without score lookups.

We inserted each into SHRiMP, and used SHRiMP to align 50 thousand reads to a reference genome with default parameters. The improvements of the Core 2 architecture for vectored instructions lead to a significant speedup for our approach and Farrar's, while Wozniak's algorithm slight improvement is due to the slow match/mismatch lookups.

我们将每条序列读段（reads）导入SHRiMP工具，并以默认参数借助SHRiMP将50,000条序列读段与参考基因组进行序列比对。酷睿2（Core 2）架构针对向量指令的优化，为本研究方法与法拉尔算法（Farrar's algorithm）带来了显著的运算提速；而沃兹尼亚克算法（Wozniak's algorithm）仅实现小幅提速，其原因在于该算法的匹配/错配查找操作效率偏低。