Dr. Thomas Steinke
Thomas is head of the HPC dept. at the Zuse Institute Berlin (ZIB). His research interest is in high-performance computing, heterogeneous systems for scientific and data analytics applications, and parallel simulation methods. Thomas co-founded the OpenFPGA initiative in 2004, and he leads the Intel® Parallel Computing Center (Intel® PCC) at ZIB. He received the doctoral degree in Theoretical Chemistry from the Humboldt-Universität zu Berlin in 1990.
Florian is part of the Distributed Algorithms and Supercomputing department at Zuse Institute Berlin (ZIB). He is interested in accelerator and many-core computing with application in Computer Science and Computational Physics. His focus is on load balancing of irregular parallel computations and on close-to-hardware code optimization. He received a Diploma degree in Physics from Humboldt Universität zu Berlin and a Bachelor degree in Computer Science from Freie Universität Berlin.
Matthias is part of the Distributed Algorithms and Supercomputing group at Zuse Institute Berlin (ZIB). His interests include parallel programming models, heterogeneous architectures, and scientific computing. He developed the Heterogeneous Active Messages (HAM) framework, which provides efficient offloading, local and over fabric, for multi- and many-cores. Matthias currently focuses on runtime compilation techniques, portable programming methods for vectorization, as well as optimization and scaling of the Hierarchical Equations of Motion (HEOM) method.
Intel Corporation and Konrad-Zuse-Zentrum für Informationstechnik Berlin (ZIB) have set up a "Research Center for Many-core High-Performance Computing" at ZIB. This Center will foster the uptake of current and next generation Intel many- and multi-core technology in high performance computing and big data analytics. The Intel® PCC at ZIB is focusing on a diverse set of codes including VASP which is targeted at atomic scale materials modelling.
The activities of the "Research Center for Many-core High-Performance Computing" are focused on enhancing selected workloads with impact on the HPC community to improve their performance and scalability on many-core processor technologies and platform architectures. The selected applications cover a wide range of scientific disciplines including materials science and nanotechnology, atmosphere and ocean flow dynamics, astrophysics, quantum physics, drug design, particle physics and big data analytics. Novel programming models and algorithms will be evaluated for the parallelization of the workloads on many-core processors.
The workload optimization for many-core processors is supported by research activities associated with many-core architectures at ZIB, where novel programming models and algorithms for many-core architectures are developed and evaluated.
Furthermore, the parallelization work is complemented by dissemination and education activities within the Northern German HPC Alliance "HLRN" to overcome the barriers involved with the introduction of upcoming highly parallel processor and platform technologies
"We are delighted to enter into a multi-year cooperation with Intel" said Prof. Alexander Reinefeld, head of the computer science department at Zuse Institute Berlin. "Our goal is to port and optimize selected HPC codes for Intel many-core processors with a special focus on maximum performance and scalability"
- Florian Wende, Guido Laubender and Thomas Steinke, May 2014, Integration of Intel Xeon Phi Servers into the HLRN-III Complex: Experiences, Performance and Lessons Learned, Cray User Group (CUG) 2014
- Olaf Krzikalla, Florian Wende, Markus Hohnerbach, June 2016, Dynamic SIMD Vector Lane Scheduling, ISC16 IXPUG Workshop
- Florian Wende, March 2015, SIMD Enabled Functions on Intel Xeon CPU & Intel Xeon Phi Coprocessor, White Paper: ZIB Website
- Rob Farber, October 2014, Concurrent Kernel Offloading, Web Article: TechEnablement
- Florian Wende, Thomas Steinke, Frank Cordes, August 2014, Concurrent Kernel Execution on Xeon Phi within Parallel Heterogeneous Workloads, Euro-Par 2014
- Florian Wende, November 2014, Enabling Manual Vectorization of Complex Code Patterns in Fortran, IXPUG
- Mattias Noack, June 2016, AVX512 vs AVX2 on KNL, IXPUG
- Olaf Krzikalla, Florian Wende, Markus Hohnerbach, June 2016, Dynamic SIMD Vector Lane Scheduling, IXPUG
- Florian Wende, November 2015, Dynamic SIMD Scheduling, IXPUG
- Florian Wende, July 2015, How Effective is SIMD in Case of Divergent Code Execution?, IXPUG
- Florian Wende, July 2015, Language Impact on Vectorization: Vector Programming in Fortran, IXPUG
- Matthisa Noack, November 2014, Hierarchical Equations of Motion: – What we can learn from OpenCL, IXPUG
- Matthisa Noack, Florian Wende, Georg Zitzlsberger, Michael Klemm, Thomas Steinke, June 2017, KART: Kernel Compilation at Run Time for Improving HPC Application Performance, IXPUG
- Zhengji Zhao, Martjn Marsman, Florian Wende, and Jeongnim Kim, May 2017, Performance of Hybrid MPI/OpenMP VASP on Cray XC40 Based on Intel Knights Landing Many Integrated Core Architecture, CUG 2017
- Matthias Noack, November 2017, The OpenCL Platform in Scientific High Performance Computing, HPCDevCon17
- Florian Wende, Martijn Marsman, Zhengji Zhao, Jeongnim Kim, September 2017, Porting VASP from MPI to MPI+OpenMP [SIMD], Springer & IWOMP 2017
- ZIB - Zuse Institute Berlin, 10/15/2015, Gaining Performance through Vectorization using Fortran by ZIB, ZIB - Zuse Institute Berlin
- ZIB - Zuse Institute Berlin, 11/03/2014, Chpt 12: Concurrent Kernel Offloading, High Performance Parallelism Pearls Volume 1
- ZIB - Zuse Institute Berlin, 07/01/2015, Chpt 19 - OpenCL: There and Back Again, High Performance Parallelism Pearls Volume 2
- ZIB - Zuse Institute Berlin, 05/01/2015, Application Performance on a Cray XC30 Evaluation System with Xeon Phi Coprocessors at HLRN-III, Cray User Group (CUG) 2015
- ZIB - Zuse Institute Berlin, 08/01/2016, Portable SIMD Performance with OpenMP* 4.x Compiler Directives, Euro-Par 2016
On Enhancing 3D-FFT Performance in VASP -- CUG'16, London, UK, 05/2016
Explicit Vectorization in VASP – IXPUG 09/2015
OpenCL: There and Back Again – IXPUG 09/2015
Runtime Kernel Compilation for efficient vectorisation – IXPUG 09/2015
Efficient SIMD-code generation with OpenCL and OpenMP 4.0 – ISC'15 IXPUG BoF, 07/2015