SODA Synthesizer
Accelerating Data Science Applications with an end-to-end Silicon Compiler
HPCA 2023 Tutorial - Room: Outremont 2
Nicolas Bohm Agostini (PNNL and Northeastern University), Serena Curzel (PNNL and Politecnico di Milano),
Michele Fiorito (Politecnico di Milano), Vito Giovanni Castellana (PNNL),
Fabrizio Ferrandi (Politecnico di Milano), Antonino Tumeo (PNNL)
Abstract
Data Science applications (machine learning, graph analytics) are among the main drivers for the renewed interests in designing domain specific accelerators, both for reconfigurable devices (Field Programmable Gate Arrays) and Application-Specific Integrated Circuits (ASICs). Today, the availability of new high-level synthesis (HLS) tools to generate accelerators starting from high-level specifications provides easier access to FPGAs or ASICs and preserves programmer productivity. However, the conventional HLS flow typically starts from languages such as C, C++, or OpenCL, heavily annotated with information to guide the hardware generation, still leaving a significant gap with respect to the (Python based) data science frameworks.
This tutorial will discuss HLS to accelerate data science on FPGAs or ASICs, highlighting key methodologies, trends, advantages, benefits, but also gaps that still need to be closed. The tutorial will provide a hands-on experience of the SOftware Defined Accelerators (SODA) Synthesizer, a toolchain composed of SODA-OPT, an opensource front-end and optimizer that interface with productive programming data science frameworks in Python, and Bambu, the most advanced open-source HLS tool available, able to generate optimized accelerators for data-intensive kernels.
Schedule
|
8:15 – 8:45am |
Antonino Tumeo |
Agile Hardware Design for Complex Data Science Applications: Opportunities and Challenges. |
|
8:45 – 9:30am |
Fabrizio Ferrandi |
Bambu: An Open-Source Research Framework for the High-Level Synthesis of Complex Applications. |
|
9:30 - 10:00am |
Nicolas Bohm Agostini |
SODA-OPT: Enabling System-Level Design in MLIR for High-Level Synthesis and Beyond |
|
10:00 - 10:20 |
|
Coffee Break |
|
10:20 – 11:05 |
Claudio Barone |
Hands-on: Productive High-Level Synthesis with Bambu, Compiler Based Optimizations, Tuning and Customization of Generated Accelerators |
|
11:05 - 11:50 |
Nicolas Bohm Agostini |
Hands-on: From DNN Models to ASIC Devices with SODA-OPT |
|
11:50 - 12:20 |
Antonino Tumeo |
Svelto: High-Level Synthesis of Multi-Threaded Accelerators for Graph Analytics |
The time zone for all times mentioned at the HPCA website is EST – Eastern Standard Time (UTC-5).
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).
✔ 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://gitlab.pnnl.gov/sodalite/soda-opt
Tutorial setup
✔ Have a computer with access to internet and a valid Google account
o We will use two Jupiter Notebooks hosted on Google COLAB
▪ Notebook and material for Bambu: https://github.com/ferrandi/PandA-bambu/tree/doc/hpca23/documentation/tutorial_hpca_2023
▪ Notebook and material for SODA-OPT: https://gitlab.pnnl.gov/sodalite/soda-opt/-/blob/main/docs/tutorials/hpca2023