Objectives

Dynamic resource management on large-scale HPC platforms is a key component of several past, currently running, and future projects. It poses highly interdisciplinary challenges concerning, among others, the applications that support dynamic resources, extending dynamic resource provisioning to the full system stack, I/O malleable systems, and I/O scheduling, among others.

This workshop is planned to be an amplifier of collaborations around these ideas and covers the following topics (but is not limited to):

• System and system architecture considerations in designing architectures that use dynamic resources. 
• Emerging software designs to achieve dynamic resource management in high-performance computing. 
• High-level parallel programming models and programmability techniques to improve dynamic resource utilization in applications. 
• Run-Time Techniques to provide dynamic resource execution models for computation, communication, and I/O. 
• Resource management frameworks and interfaces supporting dynamic resource management.
• Computing and I/O scheduling algorithms providing and/or exploiting static or dynamic resources. 
• Use of AI and ML techniques to steer dynamic resources in systems and applications
• Ad-hoc storage systems and I/O scheduling techniques supporting dynamic resource provisioning in I/O. 
• Support for dynamic resource execution of applications in performance, debugging, and correctness tools. 
• Experiences and use cases applying dynamic resource management to HPC applications. 
• Experience adopting dynamic resource frameworks.
• Performance monitoring and modeling in dynamic environments.

Important Dates

Registration deadline: 15 January