This comprehensive survey and catalog of techniques in the field of biological sequence alignment covers a large amount of relevant literature ranging from theoretical proposals to practical implementations. This is an excellent read for a practitioner in the bioinformatics space who would like a comprehensive survey as a starting point for either further research or choosing a solution to deploy for sequence alignment problems.

The paper is fairly detailed and does a good job of categorizing the solutions along a set of dimensions--this makes it easy to digest the paper and zoom in on areas of interest. The authors categorize the systems studied along a number of dimensions grouped as problem (DNA, RNA, protein; size of query; global or local alignment; gap penalties); algorithm; and platform (central processing unit (CPU), application-specific integrated circuit (ASIC)/field-programmable gate array (FPGA), graphics processing unit (GPU), and hybrid approaches). The authors focus on solutions with a dynamic programming (DP) approach (that compute a DP matrix of optimal alignments) and then broaden the discussion to various adaptations of the basic DP core to sequence types, problems, and platforms. Although the paper does not cover indexing techniques like suffix trees, I believe the treatment is still thorough given its high-performance computing focus. The paper also uses a uniform comparison standard (problem size and cell updates per second achieved) that makes it easy to use the tables provided to quickly compare various solutions and approaches.

Overall, the paper is an enjoyable read and should be accessible to people at various levels of abstraction. As the authors point out, much work remains to be done in the area--this paper is a good place to start.