Penn Engineering researchers Jing (Jane) Li, Linus Y. Wong and Jialiang Zhang are recipients of the Best Paper Award at the 31st Association for Computing Machinery (ACM)/Special Interest on Design Automation (SIGDA) International Symposium on Field-Programmable Gate Arrays (FPGA 2023).
In their paper titled, “DONGLE: Direct FPGA-Orchestrated NVMe Storage for HLS,” the authors introduced a programming model and storage architecture to tackle big data challenges in fields like machine learning, data analytics and graph processing. The proposed new programming paradigm unlocks the potential of future near-storage computing by enabling hardware accelerators to directly access massive SSD storage with minimal programming efforts, without going through a host CPU. This allows for fast and efficient data processing, with minimal latency and maximum bandwidth.
Advised by Jing (Jane) Li, Eduardo D. Glandt Faculty Fellow and Associate Professor in Electrical and Systems Engineering (ESE) and in Computer and Information Science, Wong and Zhang are Ph.D. students in ESE. Wong is a 2021 honorary recipient of the Croucher Doctoral Scholarship, and Zhang worked on the record-setting supercomputer ENIAD, which was recognized for a new world record for energy efficient supercomputing in 2021.
The annual ACM/SIGDA International Symposium is the most prestigious conference in FPGA and reconfigurable computing. This year’s Best Paper Award recipient was selected from a total of 82 submissions, out of which 21% of the reviewed papers were shortlisted for regular presentation at the conference and the final winning paper was chosen.