Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Explore TornadoVM, an open-source technology that enables Java developers to harness GPU and hardware accelerator power without requiring deep GPU programming expertise. Learn how this seamlessly integrated solution works with major JDK distributions including Amazon Corretto, GraalVM, OpenJDK, Red Hat Mandrel, Microsoft JDK, and Azul Zulu by extending the Graal compiler with GPU code generation and introducing dynamic reconfiguration and multi-device execution capabilities. Begin with a crash introduction to GPU programming fundamentals and data parallelism concepts, then master the TornadoVM API to annotate and structure Java code for expressing parallelism and enabling transparent offloading to accelerators. Discover the comprehensive tool ecosystem including the TornadoInsight IntelliJ plugin for profiling and debugging, and explore the TornadoVM Bytecode Visualizer to understand Java code transformation for heterogeneous targets. Witness a live demonstration of LLM optimization through integration with a separate Java project that showcases various optimizations applied to large language model pipelines, revealing real-world GPU acceleration techniques and performance insights while examining critical performance factors including data movements and code generation strategies.
