When: Thursday 6th of February, 1:00pm AEDT
Where: This seminar will be partially presented at the ACFR seminar area, J04 lvl 2 (Rose St Building) and partially online via Zoom. RSVP
Speaker: Jerome Justin
Title: ACFR’s HPC for ML research
Abstract:
ACFR maintains a High-Performance Computing (HPC) cluster featuring multiple GPUs. A central node manages this cluster, acting as a job scheduler to efficiently distribute tasks across the available resources. In this technical talk, we will demonstrate how to leverage this HPC to schedule your Machine Learning (ML) jobs within Docker containers. We’ll cover best practices, discuss the system’s limitations, and explore potential avenues for improvement. The talk will conclude with a Q&A session.
Bio:
Jerome Justin is a highly experienced Robotics Engineer with a diverse industry background spanning mining, medical, defense, and aerospace. Throughout his career, he has successfully worked in private sector companies to develop innovative robotic solutions for intricate challenges. Currently, Jerome contributes his expertise as a Senior Robotics Software Engineer at ACFR, where he is involved in a variety of projects, including the development of the HPC. His core specializations lie in robot kinematics, control systems, autonomous navigation, and motion planning.