NVIDIA: High-Level Synthesis in Agile System-on-Chip Flows: Overview and Techniques
Catapult High-Level Synthesis has been successfully employed within NVIDIA for multiple generations of system-on-chip research demonstrations. This talk provides a brief overview of NVIDIA Research’s use of Catapult HLS and highlights some useful features and flows of the Connections library, such as the ability to back-annotate SystemC simulations to identify and fix performance bugs.
발표자 소개
Nathaniel Pinckney
Senior Research Scientist
Nate Pinckney received his B.S. degree in engineering from Harvey Mudd College in 2008, and his Ph.D. in electrical engineering from the University of Michigan in 2015. He has authored or coauthored over 40 publications in the areas of high-level synthesis methodologies, low-power VLSI design, and cryptographic accelerators. In 2015, he joined NVIDIA in Austin, TX.
관련 자료
블루닷: Catapult-HLS를 이용한 NN기반의 DeepField-PQO 설계 가속화
고화질 영상에 대한 시장의 수요 증가로 인해 높은 인코딩 비용에 대한 비디오 서비스 제공 업체의 부담이 커지고 있습니다. 이 문제를 해결하기 위해 블루닷은 비디오 인코딩 효율을 향상시키는 AI 기반의 CODEC용 전처리 DeepField-PQO 필터를 개발하였습니다. 블루닷은 필터를 FPGA용 IP로 빠르게 구현하기 위해 HLS를 활용하고, ASIC용으로는 Catapult HLS를 활용하였습니다.
Xperi®: A Designer’s Life with HLS
This webinar will discuss two aspects of their experience going from RTL to HLS. The first topic is using HLS for algorithms such as Face Detection th