Accelerated BLAST Performance with Tera-BLASTTM: a comparison of FPGA versus GPU and CPU BLAST implementations

A number of technologies have emerged for accelerating similarity search algorithms in bioinformatics, including the use of field programmable gate arrays (FPGA), graphics processing units (GPU), and clusters of standard multicore CPUs. Here we present Tera-BLASTTM, an FPGA-accelerated implementation of the BLAST algorithm, and compare the performance to GPU-accelerated BLAST and the industry standard NCBI BLAST+ on high performance computers. Our results show that Tera-BLAST, running on the TimeLogic J-series FPGA Similarity Search Engine, performs 100’s of times faster than BLAST running on generic NVIDIA Tesla M2090 GPU cards or standard Intel Xeon multi-core CPU’s.