SODA Synthesizer: Agile Accelerator Design from Python to Silicon

ISCA 2026 – Sunday, June 28, Morning, Room 305A

Nicolas Bohm Agostini (PNNL), Serena Curzel (Politecnico di Milano), Vito Giovanni Castellana (PNNL),

Fabrizio Ferrandi (Politecnico di Milano), Antonino Tumeo (PNNL)

Abstract

Artificial intelligence (AI) workloads—including machine learning and graph analytics—are a primary driver behind the renewed interest in domain-specific accelerators targeting both reconfigurable platforms (e.g., field programmable gate arrays – FPGAs) and applicationspecific integrated circuits (ASICs). The rapid evolution of algorithms and models increasingly outpaces conventional hardware design cycles, motivating the need for agile design methodologies that can translate high-level algorithmic descriptions into efficient hardware implementations while enabling rapid exploration across competing architectural trade-offs. This tutorial presents the state of the art in compiler-driven and high-level synthesis (HLS) approaches for agile accelerator design, discussing current methodologies, architectural trends, benefits, and remaining gaps that limit widespread adoption. The tutorial provides a hands-on introduction to the Software Defined Archiectures (SODA) Synthesizer, an open-source, compiler-based, end-to-end accelerator generation framework. SODA combines SODA-OPT, an MLIR-based front-end and optimizer that interfaces directly with productive Python-based data science and machine learning frameworks, with Bambu, a state-of-the-art open-source HLS engine capable of generating optimized RTL for data-intensive kernels targeting both FPGA and ASIC implementations. Through guided examples, participants will explore how compiler abstractions enable rapid hardware–software co-design and architectural specialization from Python to silicon.

 

Tentative Schedule

 

 

Reading list

✔    Papers:

o   Zhang, J.J., Agostini, N.B., Song, S., Tan, C., Limaye, A., Amatya, V., Manzano, J., Minutoli, M., Castellana, V.G., Tumeo, A. and Wei, G.Y., 2021, July. Towards Automatic and Agile AI/ML Accelerator Design with End-to-End Synthesis. In 2021 IEEE 32nd International Conference on Application-specific Systems, Architectures and Processors (ASAP) (pp. 218-225).

o   Ferrandi, F., Castellana, V.G., Curzel, S., Fezzardi, P., Fiorito, M., Lattuada, M., Minutoli, M., Pilato, C. and Tumeo, A., 2021, December. Bambu: an Open-Source Research Framework for the High-Level Synthesis of Complex Applications. In 2021 58th ACM/IEEE Design Automation Conference (DAC) (pp. 1327-1330).

o   Minutoli, M., Castellana, V.G., Saporetti, N., Devecchi, S., Lattuada, M., Fezzardi, P., Tumeo, A. and Ferrandi, F., 2021. Svelto: High-level synthesis of multi-threaded accelerators for graph analytics. IEEE Transactions on Computers.

o   Agostini, N.B., Curzel, S., Kaeli, D., and Tumeo, A., 2022, May, SODA-OPT an MLIR based flow for co-design and high-level synthesis. In 2022 Proceedings of the 19th ACM International Conference on Computing Frontiers (CF) (pp. 201–202).

o   Agostini, N.B., Curzel, S., Amatya, V., Tan, C., Minutoli, M., Castellana, V.G., Manzano, J., Kaeli, D., and Tumeo, A. 2022. An MLIR-based Compiler Flow for System-Level Design and Hardware Acceleration. In Proceedings of the 41st IEEE/ACM International Conference on Computer-Aided Design (ICCAD '22), Article 6, (pp. 1-9)

o   Agostini, N.B., Curzel, S., Zhang, J.J., Limaye, A., Tan, C., Amatya, V., Minutoli, M., Castellana, V.G., Manzano, J., Brooks, D. and Wei, G.Y., 2022, June. Bridging Python to Silicon: The SODA Toolchain. In IEEE Micro, 42(5), (pp. 78–88). BEST PAPER FOR 2022

o   Serena Curzel, Nicolas Bohm Agostini, Vito Giovanni Castellana, Marco Minutoli, Ankur Limaye, Joseph B. Manzano, Jeff Zhang, David Brooks, Gu-Yeon Wei, Fabrizio Ferrandi, Antonino Tumeo: End-to-End Synthesis of Dynamically Controlled Machine Learning Accelerators. IEEE Trans. Computers 71(12): 3074-3087 (2022)

o   Agostini, N.B., Haris, J., Gibson, P., Jayaweera, M., Rubin, N., Tumeo, A., Abellán, J.L., Cano, J., Kaeli, D.R., 2024, March, AXI4MLIR: User-Driven Automatic Host Code Generation for Custom AXI-Based Accelerators. In proceedings of CGO 2024 (pp. 143-157)

o   Giovanni Gozzi, Michele Fiorito, Serena Curzel, Claudio Barone, Vito Giovanni Castellana, Marco Minutoli, Antonino Tumeo, Fabrizio Ferrandi: SPARTA: High-Level Synthesis of Parallel Multi-Threaded Accelerators. ACM Trans. Reconfigurable Technol. Syst. 18(1): 9:1-9:30 (2025)

✔    Additional material:

o   Other Bambu publications are listed on https://panda.dei.polimi.it/?page_id=177

o   Code repositories: https://github.com/ferrandi/PandA-bambu and https://github.com/pnnl/soda-opt

Tutorial setup

✔    Have a computer with access to internet and a valid Google account for the bambu part and access to a x86 machine with docker installed and 22GB free for the soda-opt part.

o   We will use two Jupiter Notebooks hosted on Google COLAB

▪       Notebook and material for Bambu: https://github.com/ferrandi/PandA-bambu/tree/dev/panda/documentation/bambu101

o   We will download a docker image with all tools installed: SODA docker image

▪       Notebook and material for SODA-OPT: https://github.com/pnnl/soda-benchmarks/tree/main/tutorials