TAPA: a scalable task-parallel dataflow programming framework for modern FPGAs with co-optimization of HLS and physical design

L Guo, Y Chi, J Lau, L Song, X Tian, M Khatti… - ACM Transactions on …, 2023 - dl.acm.org
In this article, we propose TAPA, an end-to-end framework that compiles a C++ task-parallel
dataflow program into a high-frequency FPGA accelerator. Compared to existing solutions …

TAPA-CS: Enabling Scalable Accelerator Design on Distributed HBM-FPGAs

N Prakriya, Y Chi, S Basalama, L Song… - Proceedings of the 29th …, 2024 - dl.acm.org
Despite the increasing adoption of FPGAs in compute clouds, there remains a significant
gap in programming tools and abstractions which can leverage network-connected, cloud …

Across Time and Space: Senju's Approach for Scaling Iterative Stencil Loop Accelerators on Single and Multiple FPGAs

E Del Sozzo, D Conficconi, K Sano - ACM Transactions on …, 2024 - dl.acm.org
Stencil-based applications play an essential role in high-performance systems as they occur
in numerous computational areas, such as partial differential equation solving. In this …

Scheduling and Physical Design

J Cong - Proceedings of the 2024 International Symposium on …, 2024 - dl.acm.org
In a typical integrated circuit electronic design automation (EDA) flow, scheduling is a key
step in high-level synthesis, which is the first stage of the EDA flow that synthesizes a cycle …

Automatic multi-dimensional pipelining for high-level synthesis of dataflow accelerators

K Majumder, U Bondhugula - arxiv preprint arxiv:2309.03203, 2023 - arxiv.org
In recent years, there has been a surging demand for edge computing of image processing
and machine learning workloads. This has reignited interest in the development of custom …

[КНИГА][B] Enabling Heterogeneous Computing for Software Developers

J Lau - 2024 - search.proquest.com
The slowing of CMOS technology scaling mismatches the ever-increasing demand for
computational power, leading to a rise in the use of heterogeneous systems, which pair …

[КНИГА][B] Co-optimizing High-Level Synthesis and Physical Design for Rapid Timing Closure of Large-Scale FPGA Designs

L Guo - 2022 - search.proquest.com
High-level synthesis (HLS) tools simplify the FPGA design processes by allowing users to
express their designs in high-level languages such as C/C++ or OpenCL. In this way, users …

[PDF][PDF] EMANUELE DEL SOZZO, RIKEN Center for Computational Science, Japan DAVIDE CONFICCONI, Politecnico di Milano, Italy

K SANO - 2023 - scholar.archive.org
Stencil-based computations are ubiquitous, for they embody a prevalent means that applies
to wildly diferent and distant scientiic ields. For instance, inance [33], image processing [14 …

[ЦИТАТА][C] Towards software-defined FPGA acceleration for big data analytics

FA Lu - 2024 - Simon Fraser University