MPI-HMMER-Boost: Distributed FPGA Acceleration

Abstract

HMMER, based on the profile Hidden Markov Model (HMM) is one of the most widely used sequence database searching tools, allowing researchers to compare HMMs to sequence databases or sequences to HMM databases. Such searches often take many hours and consume a great number of CPU cycles on modern computers. We present a cluster-enabled hardware/software-accelerated implementation of the HMMER search tool hmmsearch. Our results show that combining the parallel efficiency of a cluster with one or more high-speed hardware accelerators (FPGAs) can significantly improve performance for even the most time consuming searches, often reducing search times from several hours to minutes.

Multicore Architectures

“…Hardware support for bioinformatics has been explored in the form of fully custom accelerators, most of them using FPGAs [42,47,68,69,95] and some ASICs [35,85]. Although capable of achieving very significant speedups, FPGA solutions suffer from low power efficiency and programmability issues.…”

Improving MPI-HMMER's scalability with parallel I/O

“…Other optimizations, including several FPGA implementations, have been demonstrated in the literature [15,13,10,16]. FPGAs can achieve excellent performance, at the cost of exceptionally long development times.…”

Accelerating HMMer on FPGAs using systolic array based architecture

“…Our approach is simpler: our design can dynamically decide the operation of processing elements (PEs) and parallelize the calculations that don't involve the feedback loop; the states number of Plan7 HMM can be assigned at start up. The work in [8] combines both MPI [7] and FPGA [14] strategies together, and the proposed multi-grained acceleration achieves a reasonable speed-up.…”

“…Many solutions [1] [11] [15] are based on the simplified HMM without a rare feedback loop in HMM, while other implementations accelerates full Plan7 HMM [14] [13]. FPGA and MPI are combined into an MPI-HMMER-Boost to achieve a better speedup [8].…”