Supercomputing Frontiers is an annual international conference that provides a platform for thought leaders from both academia and industry to interact and discuss visionary ideas, important visionary trends and substantial innovations in supercomputing.
The meeting objectives are the following ones:
• High-level F2F technical meeting involving workpackage leaders (WPLs), task leaders (TLs) and flagship code lead developers (LDs) to coordinate the different tasks in WP2, WP3 and WP4.
• Explain the first code audit round strategy (T2.1).
• Specify the flagship code requirements for T2.2 (single-node optimization), T2.3 (multi-node optimization) and T2.4 (I/O bottlenecks), and identify who and how will perform each optimization task.
• Specify the requirements of flagship codes and Pilot Demonstrators for T3.4 (workflow manager). This includes the ChEESE definition of workflow(s) and the identification of generic workflow components shared by different pilots and future services (e.g. data acquisition and pre-process, interaction between multiple binaries within a job, task/job dependencies, scheduling and allocation of computational resources, fault tolerance, code restarting, post-processing, etc.).
The meeting will take place at HLRS in Stuttgart, Germany, on the “Vaihingen Campus” of the University of Stuttgart.