CGRA4ML: A Hardware/Software Framework to Implement Neural Networks for Scientific Edge Computing
This provides a modular hardware/software solution for scientists deploying neural networks in edge computing, though it is incremental as it builds on existing CGRA concepts.
The authors tackled the need for high-performance, programmable accelerators for machine learning in scientific edge computing by developing cgra4ml, an open-source framework that generates customizable CGRA accelerators in SystemVerilog RTL, demonstrating its effectiveness with ASIC and FPGA design flows.
The scientific community increasingly relies on machine learning (ML) for near-sensor processing, leveraging its strengths in tasks such as pattern recognition, anomaly detection, and real-time decision-making. These deployments demand accelerators that combine extremely high performance with programmability, ease of integration, and straightforward verification. We present cgra4ml, an open-source, modular framework that generates parameterizable CGRA accelerators in synthesizable SystemVerilog RTL, tailored to common ML compute patterns found in scientific applications. The framework supports seamless system integration through AXI-compliant interfaces and open-source DMA components, and it includes automatic firmware generation for programming the accelerator. A comprehensive verification suite and a runtime firmware stack further support deployment across diverse SoC platforms. cgra4ml provides a modular, full-stack infrastructure, including a Python API, SystemVerilog hardware, TCL toolflows, and a C runtime, which facilitates easy integration and experimentation, allowing scientists to focus on innovation rather than dealing with the intricacies of hardware design and optimization. We demonstrate the effectiveness of cgra4ml to implement common scientific edge neural networks using ASIC and FPGA design flows.