HLRN Parallel Programming Workshop Fall 2017

Speedup Your Code on Intel® Processors at HLRN

Recent Intel® Scalable processors (Skylake) and next-generation Intel® CPUs feature large core counts, wide SIMD units (AVX-512) and high-bandwidth memory (e.g. on Intel® Xeon Phi 7200 series - Knights Landing). Existing codes need to be modernized to take full advantage of the architectural features of these CPUs. Newly developed code should be designed to do so from the beginning.

We invite code developers to a workshop on advanced code optimization and programming techniques on today's and tomorrow's processors. We want to help you to make progress with optimizing your codes.

DATE: Wednesday, October 18, 2017 at 1:00 PM - Friday, October 20, 2017 at 1:00 PM

LOCATION: Berlin, Zuse Institute Berlin (ZIB), Seminar Room (entrance rotunda / Rundbau)

REGISTRATION & TRAVEL INFORMATION: Please use the Registration Page which provides information about the location of the Zuse Institute in Berlin.

Prerequisites and Additional Information

  • Do not forget to bring your code to this workshop.
  • Please bring your own laptop for the Hackathon sessions.
    • The laptop needs to have WLAN support.
    • For attendees without a HLRN account we provide temporary accounts. Please have a tool ready for private/public key generation.
    • An ssh client should be installed on your laptop for login into the HLRN system. For Windows we recommend MobaXterm.
    • To use the GUIs of various tools your laptop OS should have a support for X11 compatible client connections. For Windows, MobaXterm comes with an X11 server.

Additional Material


DAY 1: Wed, Oct 18, 2017
12:30 - 13:00 Registration
13:00 - 13:15 Opening (Thomas Steinke, ZIB)
13:15 - 13:45 Intel CPU Architectures for HPC (Chris Dahnken, Intel)
13:45 - 14:15 Cray Aries Network (Thomas Steinke, ZIB)
14:15 - 14:45 Break
14:45 - 15:15 Access to the HLRN Complex at ZIB (Wolfgang Baumann, ZIB)
15:15 - 15:45 Using the Intel Tools on HLRN (Christian Tuma, ZIB)
16:00 - 18:00 Hackathon - Optimize your code, ask the experts.
18:00 Wrap-up Day 1
DAY 2: Thur, Oct 19, 2017
08:45 - 09:00 Registration
09:00 - 09:45 OpenMP Tasking (Michael Klemm, Intel)
09:45 - 10:20 Intel VTune Amplifier (Michael Klemm, Intel)
10:20 - 10:35 Break
10:35 - 11:05 General SIMD Programming (Florian Wende, ZIB)
11:05 - 11:35 OpenMP SIMD: Details of OpenMP 4.x (Florian Wende, ZIB)
11:35 - 12:05 Using Intel Advisor (Chris Dahnken, Intel)
12:05 - 13:00 Educational Example (Matthias Läuter, ZIB)
13:00 - 14:00 Lunch
14:00 - 15:00 Hackathon (cont.)
15:00 - 15:30 Break
15:30 - 18:00 Hackathon (cont.)
18:00 Wrap-up Day 2
19:00 Social Event: Dinner at the " Luise"
DAY 3: Fri, Oct 20, 2017
09:00 - 09:45 MPI 3.0 DDT, Topologies, and Neighborhood Collectives (Matthias Noack, ZIB)
09:45 - 10:30 MPI 3.0 One-sided Communication (Florian Wende, ZIB)
10:30 - 11:00 Break
11:00 - 13:00 Hackathon (open end)
13:00 Wrap-up Day 3

Start Optimization Work Now

The Hackathon provides the unique opportunity to start or continue modernizing and optimizing your code with the assistance of experts from ZIB and Intel. We highly recommend bringing your own codes, kernels or proxy benchmarks to the workshop. For the Hackathon, the HLRN resources at ZIB, i.e. a Cray TDS with 80 nodes with Intel® Xeon Phi™ 7200 series (KNL) CPUs and the Cray XC40/XC30 system with Intel® Xeon Haswell/Ivybridge nodes are open for code evaluation, optimization and benchmarking. Attendees with a valid HLRN account can transfer their codes to the HLRN system in Berlin in advance.

Workshop Presenters and Hackathon Support

Presentations and the support in the Hackathon sessions are given by the IPCC@ZIB team, HLRN HPC consultants from ZIB and experts from Intel.

Created by ThomasSteinke - 28 Sept 2017

Last modification: ThomasSteinke - 20 Oct 2017 12:02 / 2 years, 5 months, 2 weeks ago. (Version: 27)

Topic attachments
I Attachment History Action Size Date Who Comment
Unix shell scriptsh advisor_wrapper.sh r2 r1 manage 1.9 K 2017-10-17 - 14:05 ChristianTuma  
Unix shell scriptsh vtune_wrapper.sh r2 r1 manage 2.2 K 2017-10-17 - 14:06 ChristianTuma  

Norddeutscher Verbund für Hoch- und Höchstleistungsrechnen
Back to top of page