Using a programmable network switch TCAM to find the best alignment of two DNA sequences

With the latest advances in molecular biology, a huge amount of data is being generated each day at the labs. The need to process this data in an efficient way poses a challenge for the modern computational infrastructure. One of the most common operations in computational biology is sequence comparison. In this paper we present a method and implementation prototype to find the best alignments between two DNA sequences, using a ternary context addressable memory (TCAM) and modern switch programming paradigms. These elements allows utilizing the existing processing power of high performance switch processors to perform a variety of general purpose computations. We present a simple framework and an implementation to find DNA alignments involving substrings of sequences.