Performance Modeling, Benchmarking and Simulation 
of High Performance Computer Systems (PMBS17) 

held as part of ACM/IEEE Supercomputing 2017 (SC17), Denver, CO, USA

8th IEEE International Workshop on

Session 1: HPC Interconnects
Chair: Simon Hammond

9:00 - 9:30
Evaluating On-Node GPU Interconnects for Deep Learning Workloads
  [abstract] [paper] [presentation]

Nathan R. Tallent, Nitin A. Gawande, Charles Siegel, Abhinav Vishnu
Pacific Northwest National Laboratory, Richland, WA

Adolfy Hoisie
Brookhaven National Laboratory, Upton, NY

9:30 - 10:00
Modeling UGAL on the Dragonfly Topology
  [abstract] [paper] [presentation]

Md Atiqul Mollah, Peyman Faizian, Md Shafayat Rahman, Xin Yuan
Flordia State University, Tallahassee, FL

Scott Pakin, Michael Lang
Los Alamos National Laboratory, Los Alamos, NM

10:00 - 10:30 Coffee Break

Session 2: Performance Optimization and Portability
Chair: John Pennycook

10:30 - 11:00
A performance study of Quantum ESPRESSO's PWscf code on multi-core and GPU systems
  [abstract] [paper] [presentation]

Joshua Romero, Everett Phillips, Gregory Ruetsch, Massimiliano Fatica
NVIDIA Corporation, Santa Clara, CA

Filippo Spiga
University of Cambridge, Cambridge, UK

Paolo Giannozzi
University of Udine, Udine, Italy

11:00 - 11:30
Comparison of Parallelisation Approaches, Languages, and Compilers for Unstructured Mesh Algorithms on GPUs
  [abstract] [paper] [presentation]

Gábor D. Balogh, István Z. Reguly
Pazmany Peter Catholic University, Budapest, Hungary

Gihan R. Mudalige
University of Warwick, Coventry, UK

Session 3: Late-breaking Research and Preliminary Techniques
Chair: Steven Wright

11:30 - 11:50
Path-synchronous Performance Monitoring in HPC Interconnection Networks with Source-code Attribution
  [abstract] [paper] [presentation]

Adarsh Yoga, Milind Chabbi
Hewlett Packard Labs, Palo Alto, CA

11:50 - 12:10
Performance and Energy Usage of Workloads on KNL and Haswell Architectures
  [abstract] [paper] [presentation]

Tyler Allen
Clemson University, Clemson, SC

Christopher S. Daley, Douglas Doerfler, Brian Austin, Nicholas J. Wright
Lawrence Berkeley National Laboratory, Berkeley, CA

12:10 - 12:30
A Survey of Application Memory Usage on a National Supercomputer: An Analysis of Memory Requirements on ARCHER
  [abstract] [paper] [presentation]

Andy Turner
Edinburgh Parallel Computing Centre, University of Edinburgh, Edinburgh, UK

Simon McIntosh-Smith
University of Bristol, Bristol, UK

12:30 - 14:00 Lunch

Session 4: Performance Modeling I
Chair: Doug Doerfler

14:00 - 14:30
Resilient N-Body Tree Compuations with Algorithm-Based Focused Recovery: Model and Performance Analysis
  [abstract] [paper] [presentation]

Aurélien Cavalan
University of Basel, Switzerland
Laboratoire LIP, ENS Lyon and Inria, France

Aiman Fang
University of Chicago, Chicago, IL

Andrew A. Chien
University of Chicago, Chicago, IL
Argonne National Laboratory, Argonne, IL

Yves Robert
Laboratoire LIP, ENS Lyon and Inria, France
University of Tennessee, Knoxville, TN

14:30 - 15:00
Multi-fidelity Surrogate Modeling for Application/Architecture Co-design
  [abstract] [paper] [presentation]

Yiming Zhang, Aravind Neelakantan, Nalini Kumar, Chanyoung Park, Raphael T. Haftka, Nam H. Kim, Herman Lam
University of Florida, Gainsville, FL

15:00 - 15:30 Coffee Break

Session 5: Performance Modeling II
Chair: Jeremiah Wilke

15:30 - 16:00
Modeling Large Compute Nodes with Heterogeneous Memories with Cache-Aware Roofline Model
  [abstract] [paper] [presentation]

Nicolas Denoyelle, Brice Goglin, Emmanuel Jeannot
Inria, University of Bordeaux, Bordeaux, France

Aleksandar Ilic, Leonel Sousa
Universidade de Lisboa, Lisbon, Portugal

16:00 - 16:30
A Scalable Analytical Memory Model for CPU Performance Prediction
  [abstract] [paper] [presentation]

Gopinath Chennupati, Nandakishore Santhi, Robert Bird, Sunil Thulasidasan, Stephan Eidenbenz
Los Alamos National Laboratory, Los Alamos, NM

Abdel-Hameed A. Badawy, Satyajayant Misra
New Mexico State University, Las Cruces, NM

Session 6: HPC Scheduling
Chair: Stephen Jarvis

16:30 - 17:00
A Slurm Simulator: Implementation and Parametric Analysis
  [abstract] [paper] [presentation]

Nikolay A. Simakov, Martins D. Innus, Matthew D. Jones, Robert L. DeLeon, Joseph P. White, Steven M. Gallo, Abani K. Patra, Thomas R. Furlani
State University of New York, University at Buffalo, Buffalo, NY

17:00 - 17:30
Periodic I/O Scheduling for Super-computers
  [abstract] [paper] [presentation]

Guillaume Aupy
Inria, University of Bordeaux, Bordeaux, France

Ana Gainaru
Vanderbilt University, Nashville, TN

Valentin Le Fèvre
Inria, University of Bordeaux, Bordeaux, France
École Normale Supérieure de Lyon, Lyon, France

Program Committee
Submit Paper