An Energy-Efficient Domain-Specific Architecture for Regular Expressions

Regular Expressions (REs) are a computational kernel widely used for finding patterns in data compute-intensive tasks such as genomic markers research, signature-based detection, and database query. Although flexible on the set of searched REs, software-based solutions cannot fulfill latency or throughput requirements to analyze massive volumes at given power budget. For this reason, many approaches exploit hardware accelerators an offloading engine REs matching. Indeed, various rely FPGA reconfigurability embed automata into reconfigurable fabric. However, approach leads time-consuming updates search. This work exploits sequences basic instructions builds Domain-Specific Architecture (DSA), called \tirex, RE matching FPGAs. Our enables user change desired run-time, providing software programmability, flexibility, specialized mechanisms. DSA delivers performance line with other state-of-the-art approaches, while remarkable flexibility we underline importance energy efficiency these computations. We compared multiple obtaining achieving noticeable results better that ranges from 3x 490x our multi-core.