Hardware acceleration of complex HEP algorithms with HLS and FPGAs: Methodology and preliminary implementation

Article Instruments & Instrumentation

Automated firmware generation and continuous testing for the CMS HGCAL trigger primitive generator

F. Beaujean et al.

Summary: The prototype version of the trigger primitive generator firmware for the Phase-2 CMS endcap calorimeter upgrade is being implemented, utilizing data-driven design flow and automated testing to assess the resource requirements and dimension the system.

JOURNAL OF INSTRUMENTATION (2022)

添加到收藏夹

Article Computer Science, Hardware & Architecture

FPGA HLS Today: Successes, Challenges, and Opportunities

Jason Cong et al.

Summary: The article evaluates the success of FPGA high-level synthesis technology in various application domains and discusses the challenges faced by the current technology and future research directions.

ACM TRANSACTIONS ON RECONFIGURABLE TECHNOLOGY AND SYSTEMS (2022)

添加到收藏夹

Article Computer Science, Information Systems

An Efficient HW/SW Design for Text Extraction from Complex Color Image

Mohamed Amin Ben Atitallah et al.

Summary: In this paper, an efficient HW/SW design using the GCM method for text extraction is proposed, aiming to study the complexity of the GCM method on Xilinx ZCU102 FPGA board and improve the quality of text extraction. Experimental results showed significant reduction in processing time compared to SW implementation.

CMC-COMPUTERS MATERIALS & CONTINUA (2022)

添加到收藏夹

Article Instruments & Instrumentation

Trigger primitive generation algorithm in the CMS barrel muon chambers during HL-LHC

N. Trevisani

Summary: This contribution presents an update on the Analytical Method (AM) algorithm for trigger primitive (TP) generation in the CMS Drift Tube (DT) chambers during the High Luminosity LHC operation, which has been developed and validated both in software with an emulation approach, and through hardware implementation tests. The algorithm is mainly divided into three steps: grouping, fitting, and correlation, and the software emulation and firmware implementation have been verified to be consistent. The update of the grouping step using a pseudo-Bayes classifier will be discussed in this contribution.

JOURNAL OF INSTRUMENTATION (2021)

添加到收藏夹

Article Computer Science, Theory & Methods

Efficient Methods for Mapping Neural Machine Translator on FPGAs

Qin Li et al.

Summary: Neural machine translation is a critical application in natural language processing, aiming to convert text from one language to another using deep neural networks. Recent developments in NMT have integrated emerging technologies like bidirectional GRUs, attention mechanisms, and beam search algorithms to improve translation quality. However, the increasing complexity of real-life NMT models poses challenges for hardware implementations to accelerate performance.

IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS (2021)

添加到收藏夹

Article Computer Science, Hardware & Architecture