Michalis Papadimitriou

Michalis Papadimitriou, a compiler engineer at TornadoVM, discusses GPU acceleration for LLMs in Java. He covers the evolution from SIMD optimizations to enhanced GPU memory management. Key insights include the hybrid approach that blends CPU and GPU tasks, and the introduction of a persist/consume API to optimize data handling. Michalis highlights the performance trade-offs between TornadoVM and CUDA, along with the increasing role of LLMs in kernel optimization. He also hints at future support for Apple Silicon and new models, showcasing TornadoVM's expanding capabilities.

Michalis Papadimitriou, an expert in GPU acceleration and compiler optimizations for AI and Java, shares fascinating insights from his journey in tech. He discusses how he achieved up to 20x speedups in Java applications by leveraging OpenCL and TornadoVM. Hear about his work at Huawei and how he is optimizing AI frameworks like Llama 3, emphasizing the importance of standardizing ML model formats. With a focus on enhancing GPU processing in Java, he highlights kernel fusion techniques and the exciting potential of Graal VM in the modern developer landscape.