Birds of a Feather

Advances in FPGA Programming and Technology for HPC

Authors: Martin Herbordt (Boston University), Kentaro Sano (RIKEN), Christian Plessl (Universitat Paderborn), Tong Geng (University of Rochester), Taisuke Boku (University of Tsukuba), Jeffrey Vetter (Oak Ridge National Laboratory), Andreas Koch (TU Darmstadt), Mathias Jacquelin (Cerebras Systems), Dawei Huang (SambaNova Systems Inc), Ilan Tayari (NextSilicon Inc), Nicolas Agostini (Pacific Northwest National Laboratory (PNNL))

Abstract: FPGAs have gone from niche components to being a central part of many data centers worldwide. The last year has seen tremendous advances in FPGA programmability and technology, especially in the shift to reconfigurable architectures that are heterogeneous and/or based on CGRAs or other AI engines. This BoF has two parts. The first is a series of lightning talks presenting advances in tools, technologies, and use-cases for these emerging architectures. The second part of the BoF will be a general discussion driven by the interests of the attendees, potentially including additional topics.

Long Description: FPGAs have gone from niche components to a central part of many data centers worldwide and are now being considered for integration into future-generation core HPC installations. But while current use cases often match those of traditional accelerators (e.g., AWS instances), most of the millions of newly deployed FPGAs are in other configurations such as smart NICs (DPUs/IPUS) and in storage architectures.

The topic of this BOF is the latest progress and trends in HPC/FPGA, including architectures, programming methods, run-time systems, performance modeling, benchmarks, algorithms, and applications. Special emphasis will be on the shift to reconfigurable architectures that are heterogeneous and/or based on CGRAs or other AI engines. These include offerings from Intel and AMD, but also from emerging companies such as Samba Nova, Cerebras, Next Silicon, and Groq.

There are multiple goals of this BOF. First is communication of advances to a broad HPC audience. Second is obtaining feedback from the same. Third is expert discussion and evaluation. Fourth is to identify best practices that could become the basis for standards. Fifth is to provide a forum for the HPC/FPGA community to review status and present breaking news. Finally, we will continue the highly successful effort to introduce new researchers and researchers from underrepresented groups.

The relevance to a general HPC audience is filling the knowledge/experience gap with respect to HPC/FPGA. While most SC attendees are familiar with the basics of FPGAs, few have recently used them in production. Since FPGA and other configurable architectures are undergoing revolutionary changes, and these are currently being evaluated by multiple national labs, bridging this gap is vital.

We are extremely proud that this BOF has run continuously since 2010 (minus 2020). In the early years the emphasis was on outreach, novel architectures, and niche applications, and was popular with 50-70 attendees per year, depending on timeslot. More recently there has been additional emphasis on the emerging large deployments and attendance now regularly approaches 100.

The expected outcomes are in satisfying the goals stated, especially in showcasing heterogeneous FPGA/HPC to a broader audience and identifying best practices.

Since its inception, this BOF has been the go-to meeting place for the FPGA-in-HPC community, not just at SC, but in the yearly calendar. The community has grown tremendously and there is now a series of international workshops on FPGAs for HPC. This BOF is complementary to other major FPGA event at SC--the H2RC workshop--in that it focuses on systems and standards rather than research or use cases.

All session leaders have strong experience in HPC and reconfigurable computing. We will select several speakers from session leaders and leading experts in academia, government, and industry. There will also continue to be special emphasis on including as many new researchers and researchers from under-represented groups as possible.

Audience participation is a key success metric, and half the time has been allocated for discussion. As always, we expect the conversation to be lively and informative.



Back to Birds of a Feather Archive Listing