ISC 2019 Research Papers

The ISC research paper sessions provide world-class opportunities for engineers and scientists in academia, industry and government to present and discuss issues, trends and results that will shape the future of high performance computing (HPC), Networking, Storage and AI/Machine Learning.

The ISC organizers as well as the German Gauss Center for Supercomputing will again sponsor the call for research papers with two awards for outstanding research papers: the Hans Meuer Award and the GCS Award.

Attendance will require a Conference Pass. 

Chair - Dipl.-Ing. Carsten Trintis

  • Dipl.-Ing. Carsten Trintis

From all Research Papers submitted to the conference, the ISC 2019 Research Papers Committee, which is headed by Dr.-Ing. Carsten Trinitis, TU Munich, Germany, with Prof. Saday Sadayappan, Ohio State University, USA, as Deputy Chair, has selected the following Research Papers for presentation at ISC High Performance 2019:

Monday, June 17

Research Paper Award Session

  • Hans Meuer Award Finalist 1: Global Task Data Dependencies in PGAS Applications
  • Hans Meuer Award Finalist 2: GPUMixer: Performance-Driven Floating-Point Tuning for GPU Scientific Applications
  • GCS Award Winning Paper: End-to-end Resilience for HPC Applications

Tuesday, June 18

Research Paper Session - Data Storage and Visualisation

  • SLOPE: Structural Locality-aware Programming Model for Composing Array Data Analysis
  • A Near-Data Processing Server Architecture and Its Impact on Data Center Applications
  • Comparing the Efficiency of In Situ Visualization Paradigms at Scale 

Research Paper Session - Machine Learning and Emerging Technologies

  • Densifying Assumed-sparse Tensors: Improving Memory Efficiency and MPI Collective Performance during Tensor Accumulation for Parallelized Training of Neural Machine Translation Models
  • Layout-Aware Embedding for Quantum Annealing Processors
  • Learning Neural Representations for Predicting GPU Performance 

Research Paper Session - Performance Modelling and Measurement & Algorithms

  • PerfMemPlus: A Tool for Automatic Discovery of Memory Performance Problems
  • Toward efficient architecture-independent algorithms for dynamic programs
  • Performance Exploration Through Optimistic Static Program Annotations 

Wednesday, June 19

Research Paper Session - HPC Applications

  • Petaflop Seismic Simulations in the Public Cloud
  • MaLTESE: Large-Scale Simulation-Driven Machine Learning for Transient Driving Cycles 

Research Paper Session - Programming Models & System Software, Architectures & Networks

  • Evaluating Quality of Service Traffic Classes on the Megafly Network
  • Resilient Optimistic Termination Detection for the Async-Finish Model
  • Finepoints: Partitioned Multithreaded MPI Communication