Denver, Colorado
Colorado Convention Center, Denver, USA
Conference
November 12 - 17, 2017
Exhibition
November 13 - 16 2017
Supercomputing 2017
The International Conference for High Performance Computing, Networking, Storage and Analysis.
The world's largest high performance computing (HPC) event, featuring the latest technologies, trends and innovations, provides a platform for sharing key questions and the latest developments in HPC, networking, storage and analysis, and big data. The German HPC community, especially the members of the Gauß-Allianz, will once again cover a broad spectrum from software development, parallel programming, runtime support and error analysis to computer architecture, clusters and distributed computing, to embedded systems.
Find out more about the exhibition presence of German HPC centers as well as the activities of the German HPC community at SC17 in Workshops and Tutorials, Paper sessions, Birds of a Feathers, Poster sessions, Panel discussions and Doctoral showcases.
Find out more about the exhibition presence of German HPC centers as well as the activities of the German HPC community at SC17 in Workshops and Tutorials, Paper sessions, Birds of a Feathers, Poster sessions, Panel discussions and Doctoral showcases.
Deutsches Klimarechenzentrum GmbH (DKRZ)
Das DKRZ präsentiert auf seinem Stand (Nr. 262) seine Infrastruktur sowie seine Services für modellbasierte Klimasimulationen. Ein Monitor zeigt aktuelle Visualisierungen, darunter mögliche zukünftigen Änderungen von Temperatur- und Niederschlag, Landnutzung oder in der atmosphärischen Zirkulation oder Meeresströmungen. An einem interaktiven Touchtisch kann erkundet werden, wie sich das Klima ändern würde, wenn sich die Erde anders herum drehen würde.Visit the DKRZ at booth #262 .
Fraunhofer ITWM
The Fraunhofer Center for HPC is focused on the development of highly parallel applications, corresponding frameworks and tools to enable high scalability. Now we present novel algorithms for Large Scale Machine Learning. The basis for scalable systems are scalable programming models and scalable I/O. GPI2 (Global Address Space Programming Interface) and the parallel filesystem BeeGFS (delivered by the Fraunhofer spin-off ThinkParQ) are 2 scalable and proven HPC core tools for this market. We demonstrate the power of our parallel programming and cloud framework GPI-Space that enables scalable data flow driven applications. The Fraunhofer Center for HPC cooperates with research and industry.Visit the Fraunhofer ITWM at booth #2143 .
FAU Erlangen-Nürnberg - RRZE
An der diesjährigen Student Cluster Competition nimmt erstmals ein gemeinsames Team „TUMany SegFAUlts“ aus sechs Studierenden der beiden größten bayerischen Hochschulen TUM und FAU mit einem erfreulich großen Frauenanteil von 50% teil. Das Gauss Centre for Supercomputing (GCS) agiert für das Team als Co-Sponsor. Das RRZE führt auf dieser SC17 wieder das Ganztags-Tutorial „Node-Level Performance Engineering“ durch.Gauss Centre for Supercomputing (GCS)
Die drei GCS-Zentren HLRS, JSC und HLRS werden auf der SC ebenfalls vertreten sein. Repräsentanten aller drei Zentren wirken im technischen Konferenzprogramm in Tutoials, BoFs oder Workshops mit, zudem präsentieren sich die GCS-Zentren in der Ausstellung auf jeweils eigenen Messeständen:Das HLRS legt bei seinen Exponaten den Schwerpunkt auf immersive parallele Visualisierung. Anwendungsgebiete werden u. a. der Maschinenbau (z.B. Luftströmung in Verpackungsanlagen), die Stadtplanung (Feinstaub- und Verkehrssimulationen), und die Architektur und das Bauwesen sein (Building Information Modeling).
Das LRZ zeigt Komponenten des neuen CoolMUC-3 Systems, einer zu 100% warmwassergekühlten KNL-basierten Maschine von Megware. Zudem können Messebesucher eine Virtual Reality (VR) Demo erleben, die ihnen die Verbindung zwischen HPC und VR näherbringt.
Im Mittelpunkt des JSC-Auftritts steht das Konzept Modulares Supercomputing und die Vorstellung des neuen JURECA-Booster-Moduls. Präsentiert werden Aktivitäten der Sicherheitsforschung mit den Schwerpunkten Fußgängerdynamik, Verkehr und Brandschutz, die Beteiligung an der Infrastruktur für das Human-Brain-Projekt sowie die Kooperation mit der RWTH Aachen in der Jülich Aachen Research Alliance (JARA). Außerdem werden Ergebnisse wissenschaftlicher Simulationen und im JSC entwickelte HPC-Softwarewerkzeuge wie z. B. Scalasca, LLview, und SIONlibvorgestellt. Mit Dr. Bernd Mohr vom JSC ist erstmals ein Nicht-Amerikaner Vorsitzender der SC.
Visit the GCS centers at the booths #417 (HLRS@GCS), 401 (JSC@GCS), 875 (LRZ@GCS).
Zentrum für Informationsdienste und Hochleistungsrechnen (ZIH)
The ZIH will present its know-how on the topics of HPC, Data Intensive Computing as well as Grid and Cloud Computing in talks, tutorials and discussions. In addition to current research work and software developments, activities in the field of Big Data and data analysis will be presented. Two special topics are the online performance analysis and the analysis of the I/O behavior of applications.At the ZIH booth, visitors can try out the interactive online analysis prototype Vampir Live and experience new I/O analysis capabilities of Vampir. In particular, the I/O behavior of applications is becoming a bottleneck on the way to exascale high-performance computers.
Novel I/O systems based on non-volatile memory technologies address this bottleneck. Together with the EU Horizon202020 project NEXTGenIO (http://www.nextgenio.eu/ ), the capabilities of the Score-P and Vampir tools have been significantly enhanced for I/O analysis. The online analysis approach presents strategies for dealing with limited I/O capacities on the tool side as well.
Visit the ZIH at booth #1881 .
Tutorials and Workshops
12. November 2017
Advanced OpenMP: Performance and 4.5 Features
Parallel Programming Languages, Libraries, Models and Notations
8:30am - 5pm
404
Christian Terboven (RWTH Aachen University)
Hands-On Practical Hybrid Parallel Application Performance Engineering
Performance
8:30am - 5pm
405
Christian Feld (Juelich Supercomputing Center)
Markus Geimer (Juelich Supercomputing Center)
Ronny Tschüter (Technical University Dresden)
Brian Wylie (Juelich Supercomputing Center)
Node-Level Performance Engineering
Performance
8:30am - 5pm
402
Georg Hager (University of Erlangen-Nuremberg)
Gerhard Wellein (University of Erlangen-Nuremberg)
Eighth Annual Workshop for the Energy Efficient HPC Working Group (EE HPC WG)
9am - 5:30pm
704-706
Daniel Hackenberg (Technical University Dresden)
Torsten Wilde (Leibniz Supercomputing Centre)
13. November 2017
Mastering Tasking with OpenMP
Parallel Programming Languages, Libraries, Models and Notations
8:30am - 12pm
210-212
Christian Terboven (RWTH Aachen University)
Application Porting and Optimization on GPU-Accelerated POWER Architectures
GPU
Performance
8:30am - 5pm
405
Dirk Pleiter (Forschungszentrum Juelich)
Andreas Herten (Forschungszentrum Juelich)
Fourth Workshop on Accelerator Programming Using Directives (WACCPD)
Machine Learning in HPC Environments
SIGHPC Workshop
9am - 5:30pm
503-504
Janis Keuper (Fraunhofer Institute for Industrial Mathematics)
Paper
14. November 2017
A Configurable Rule-Based Classful Token Bucket Filter Network Request Scheduler for the Lustre File System
File Systems
QoS
Storage
11:30am - 12pm
405-406-407
Lingfang Zeng (Johannes Gutenberg University Mainz)
Jürgen Kaiser (Johannes Gutenberg University Mainz)
Tim Süß (Johannes Gutenberg University Mainz)
André Brinkmann (Johannes Gutenberg University Mainz)
15. November 2017
Towards Fine-Grained Dynamic Tuning of HPC Applications on Modern Multi-Core Architectures
Performance
3:30pm - 4pm
301-302-303
Daniel Hackenberg (Technical University Dresden)
Robert Schöne (Technical University Dresden)
Poster
14. November 2017
Comprehensive Visualization of Large-Scale Simulation Data Linked to Respiratory Flow Computations on HPC Systems
Visualization
8:30am - 5pm
Mile High Prefunction
Andreas Lintermann (Juelich Aachen Research Alliance)
Sonja Habbinga (Forschungszentrum Juelich)
Jens Henrik Goebbert (Forschungszentrum Juelich)
15. November 2017
Comprehensive Visualization of Large-Scale Simulation Data Linked to Respiratory Flow Computations on HPC Systems
Visualization
8:30am - 5pm
Mile High Prefunction
Andreas Lintermann (Juelich Aachen Research Alliance)
Sonja Habbinga (Forschungszentrum Juelich)
Jens Henrik Goebbert (Forschungszentrum Juelich)
16. November 2017
Comprehensive Visualization of Large-Scale Simulation Data Linked to Respiratory Flow Computations on HPC Systems
Visualization
8:30am - 5pm
Mile High Prefunction
Andreas Lintermann (Juelich Aachen Research Alliance)
Sonja Habbinga (Forschungszentrum Juelich)
Jens Henrik Goebbert (Forschungszentrum Juelich)
Birds of a Feather
14. November 2017
OpenACC API User Experience, Vendor Reaction, Relevance, and Roadmap
Usability, Scalability and Productivity on Many-Core Processors: Intel Xeon Phi
15. November 2017
Contemporary Design of Supercomputer Experiments
OpenHPC Community BoF
State of the Practice: Energy and Power Aware Job Scheduling and Resource Management (EPA-JSRM)
16. November 2017
Machine Learning for Parallel Performance Analytics
Panel
14. November 2017
How Serious Are We About the Convergence Between HPC and Big Data?
Data management
Deep Learning
Hot Topics
10:30am - 12pm
201-203
Bastian Koller (High Performance Computing Center Stuttgart)
15. November 2017
Best Practices for Architecting Performance and Capacity in the Burst Buffer Era
Accelerators
File Systems
I/O
Runtime Systems
10:30am - 12pm
201-203
Dirk Pleiter (Forschungszentrum Juelich)
Doctoral showcase
14. November 2017
Productivity and Software Development Effort Estimation in HPC
Accelerators
Algorithms
Applications
Architectures
Clouds and Distributed Computing
GPU
Graph Algorithms
Heterogeneous Systems
Networks
Programming Systems
Runtime Systems
1:37pm - 2:08pm
Four Seasons Ballroom
Sandra Wienke (RWTH Aachen University)
Productivity and Software Development Effort Estimation in HPC
Accelerators
Algorithms
Applications
Architectures
Clouds and Distributed Computing
GPU
Graph Algorithms
Heterogeneous Systems
Networks
Programming Systems
Runtime Systems
3:30pm - 3:48pm
210-212
Sandra Wienke (RWTH Aachen University)
Productivity and Software Development Effort Estimation in HPC
Accelerators
Algorithms
Applications
Architectures
Clouds and Distributed Computing
GPU
Graph Algorithms
Heterogeneous Systems
Networks
Programming Systems
Runtime Systems
5:15pm - 7pm
Four Seasons Ballroom
Sandra Wienke (RWTH Aachen University)
15. November 2017
Productivity and Software Development Effort Estimation in HPC
Accelerators
Algorithms
Clouds and Distributed Computing
GPU
Graph Algorithms
Heterogeneous Systems
Networks
Programming Systems
Runtime Systems
10am - 5:15pm
t.b.a.
Sandra Wienke (RWTH Aachen University)
16. November 2017
Productivity and Software Development Effort Estimation in HPC
Accelerators
Algorithms
Applications
Architectures
Clouds and Distributed Computing
GPU
Graph Algorithms
Heterogeneous Systems
Networks
Programming Systems
Runtime Systems
10am - 3pm
Four Seasons Ballroom
Sandra Wienke (RWTH Aachen University)