Our lab has just started analyzing Apple’s M4 chip. It turns out that M4 supports the Scalable Matrix Extension (SME) of the Arm Architecture. This opens the way for open source developments that support M4’s matrix accelerator(s). We plan to add SME support to the JITter LIBXSMM in the next few weeks, which will allow us to integrate SME into upstream software. We are documenting these efforts on a dedicated homepage. Check it out!

Paul Springer a senior software developer at NVIDIA will give a guest lecture in the class Parallel Computing. The lecture will take place on Thursday, January 11, 2024, from 12:00PM - 02:00PM. The location is HS 235 (Fürstengraben 1). Interested students outside of the class are cordially invited to join. Please write a brief email to alex.breuer@uni-jena.de or in Matrix if you would like to attend and are not part of the Parallel Computing class.

About the speaker: Dr. Paul Springer is a senior software developer at NVIDIA with a strong interest in low-level kernel development and their applications to quantum circuit simulations, computational physics and machine learning. Before joining NVIDIA in 2018, he received his Ph.D. in computer science from RWTH Aachen university where he focused on the development of high-performance tensor operations and dense linear algebra. Nowadays he is primarily tasked with the design of CUDA math libraries—most noticeably cuTENSOR.

Our lab co-organizes the AI Summer School 2023 which will take place from September 4-8 in Bad Kösen. The program is tailored to bachelor’s and master’s students with previous experience in artificial intelligence and programming. Accommodation and meals of all participants are covered. Additionally, travel support is available for selected participants.

The application process opens on April 1 and closes on May 15. Students at Friedrich Schiller University Jena and Technische Universität Ilmenau are invited to apply. A few spots are available for students enrolled at any other German university.