Gahls: An Optimized Graph Analytics Based High Level Synthesis Framework

This article proposes a novel comprehensive graph analytics based high level synthesis (GAHLS) framework that efficiently analyzes complex high level programs and synthesizes them into message passing domain-specific accelerators. The GAHLS framework first constructs a compiler-assisted dependency graph (CaDG) from low level virtual machine (LLVM) intermediate representation (IR) of high level programs and converts it into a hardware friendly description representation. Evaluations of the GAHLS framework on several real-life applications demonstrate that it provides 14.27× performance improvements compared to state-of-the-art approaches.