Application Extreme-scaling Experience of Leading Supercomputing Centres (aXXLs)

Start
16th July 2015 07:00 AM
End
16th July 2015 16:00 PM
Location
Workshop on ISC-HPC 2015, Frankfurt, Germany

Workshop on ISC-HPC, Frankfurt, 16 July 2015

Scope of the workshop

Recent editions of the Top500 list of supercomputers have shown little change in the top systems, perhaps a global pause as everyone prepares for exascale. Well-established IBM BlueGene/Q and x86_64-based clusters from Cray and others are strongly represented, as well as newer architectures incorporating GPGPU accelerators and MIC coprocessors, and experience porting and scaling applications on these systems is varied and scattered around America, Asia and Europe. This workshop will bring together speakers from the leading supercomputing centres to share their experience with running and tuning highly-scalable applications on current leadership computer systems and open discussion on how they may look in future. In particular, observed gaps between current and the "ideal" solution arising from technology limitations, programming models, file I/O, tools for debugging and analysis, etc., will be noted. This workshop is organized by employees of the Jülich Supercomputing Centre (JSC).

Targeted audience

Developers of HPC applications; HPC system providers, purchasers, administrators and support staff

Agenda

09:00-11:00

Session 1: Homogeneous HPC Systems

B. Wylie (JSC)

Introduction

D. Brömmel (JSC)

Extreme-scaling applications 24/7 on JUQUEEN Blue Gene/Q

M. Tsuji (RIKEN)

"K-scale" applications on the K computer and co-design effort for "post-K"

J.-P. Nominé (CEA)

Application Extreme-scaling Experience on Bullx Curie, Helios & Tera100

B. de Supinski (LLNL)

Programming Future Large Scale Systems

 

Discussion

11:30-13:00

Session 2: Accelerated Systems (MIC)

Y. Du (NSCC-GZ)

Scaling applications from six application domains on Tianhe-2

C. Rosales-Fernandez (TACC)

Application Extreme-scaling Experience on Stampede

F. Jamitzky (LRZ)

SuperMUC Extreme Scale-out Phase2, lessons learned

 

Discussion

14:00-16:00

Session 3: Accelerated Systems (GPU)

S. Andersson (Cray)

A Strategy for Developing a Performance Portable Highly Scalable Application

W. Sawyer (CSCS)

Can large clusters of GPUs lead us to ExaScale?

A. Nomura (TITech)

Performance measurement and monitoring in TSUBAME2.5 towards next generation supercomputers

N. Stringfellow (Pawsey)

From Tailored Systems to Scalable Machines, Multiple Architectures to Meet Competing Demands in a Data-Intensive World

 

Discussion

16:30-18:00

Discussion Session

 

Open discussion and wrap-up

Workshop material

Workshop organizers and contact

Dr. Dirk Brömmel
Dr. Wolfgang Frings
Dr. Brian J. N. Wylie

Jülich Supercomputing Centre
Institute for Advanced Simulation
Forschungszentrum Jülich GmbH
52425 Jülich, Germany

For any questions concerning the workshop please send an e-mail to Dr. Brian J. N. Wylie.

Venue
Last Modified: 18.11.2022