SAURIA is a Convolutional Neural Network (CNN) accelerator based on an output stationary (OS) systolic array with on-chip, on-the-fly convolution lowering, written entirely in SystemVerilog.

The accelerator can natively compute convolution and general matrix-matrix multiplication (GEMM) operations of any shape and size. The architecture is parametric and can be configured to use different systolic array shapes and sizes, local memory shapes, and arithmetic formats, but it has been tested extensively using FP16 arithmetic and an array of shape 8x16 (128 PEs).

Maintainer: Jordi Fornt Mas (jordi.fornt@bsc.es)

Check out the project's Wiki (https://github.com/bsc-loca/sauria/wiki) for some in-depth explanations on how the accelerator is built.

• Python (to generate the test stimuli)
• Gtkwave (for visualization of verilator simulations)

After cloning the repository and selecting the branch, run the following command to update all the submodules:

git submodule update --init --recursive

SAURIA can be emulated using Verilator v4.224. We have observed issues with older and newer versions, so we recommend using a local installation as described below. A testbench for simulation with commercial RTL simulators (e.g. Synopsys VCS or Questa) is also provided.

Before running any experiments, make sure to execute the setup script. The first time, it will automatically install the Python virtual environment with all dependencies needed to use the repo.

source setup.sh

To install Verilator v4.224 locally, run:

cd tools/
source install_verilator.sh

We provide an example Jupyter Notebook explaining step by step how to perform simulations with the repo, which you can find here (Python/notebooks/example_basic.ipynb).

If you use SAURIA in your work, you can cite the following paper:

@ARTICLE{sauria2023,
  author={Fornt, Jordi and Fontova-Musté, Pau and Caro, Martí and Abella, Jaume and Moll, Francesc and Altet, Josep and Studer, Christoph},
  journal={IEEE Transactions on Very Large Scale Integration (VLSI) Systems}, 
  title={An Energy-Efficient GeMM-Based Convolution Accelerator With On-the-Fly im2col}, 
  year={2023},
  volume={31},
  number={11},
  pages={1874-1878},
  doi={10.1109/TVLSI.2023.3286122}}

If you would like to contribute to SAURIA, please contact jordi.fornt@bsc.es.

SAURIA is released under the Solderpad v2.1 license.