The Supercomputing Conference (SC) is taking place in Denver, USA, this year. DEEP-SEA and the sister SEA projects will also be represented at the world’s largest HPC event. 

DEEP-SEA will be at the SC together with its two sister projects IO-SEA and RED-SEA. The SEA projects will share a booth with the ACROSS project (booth no. 214). The three projects will also have a presence at the Jülich booth (booth no. 401). Visitors at the booths will be able to inform themselves about the three projects and talk to experts about the current HPC developments and the latest news in the three projects. In addition, information about the SEA projects will also be displayed at the EuroHPC JU’s booth (booth no. 204). The SC floor plan can be accessed here.

In addition to the booths, DEEP-SEA researchers will take part in the technical programme:

Workshops/Panels:

• Third International Symposium on Quantitative Codesign of Supercomputers

Co-organiser: Estela Suarez, Jülich Supercomputing Centre

• Runtimes and Workflow Systems for Extreme Heterogeneity: Challenges and Opportunities

Panelist: Antonio J. Peña, Barcelona Supercomputing Center (BSC)

• 4th Workshop on Heterogeneous Memory Systems (HMEM 2023)

Co-organiser: Antonio J. Peña, Barcelona Supercomputing Center (BSC)

Birds of a feather sessions:

• Power Consumption and Exascale Computing: Toward a “Short Production Circuit” Model 

Session leader: Philippe Deniel, Atomic Energy and Alternative Energies Commission (CEA)

• Go with the (Energy) Flow: Adaptive Capacity Computing

Session leader: Felix Wolf, Technical University of Darmstadt

• SmartNICs : Exploring the Future of In-Network Computation with the HPC Community

Additional Session leader: Estela Suarez, Jülich Supercomputing Centre

• Introducing MPI 4.1, the Newest Version of the Message Passing Interface Standard

Session leader: Martin Schulz, Technical University of Munich

• Enabling I/O and Computation Malleability in High-Performance Computing

Session leader: Jesus Carretero, University Carlos III of Madrid

• MPICH: A High Performance Open-Source MPI Implementation

Workshop Presentations:

• Filtering and Ranking of Code Regions for Parallelization via Hotspot Detection and OpenMP Overhead Analysis
• Extra-Deep: Automated Empirical Performance Modeling for Distributed Deep Learning
• Efficient Data Redistribution for Malleable Applications
• GPUscout: Locating Data Movement-Related Bottlenecks on GPUs

Tutorials:

• Hands-On Practical Hybrid Parallel Application Performance Engineering

Poster presentations:

• sys-sage: A Fresh View on Dynamic Topologies and Attributes of HPC Systems