In August 2015, the Cray XC40 supercomputer Hornet at HLRS was upgraded to a new system named "Hazel Hen", featuring 7712 compute nodes, each equipped with two 12-core Intel Haswell processors running at 2.5 GHz. Each node is equipped with 128 GB DDR4 memory and connected to the other nodes through the Cray Aries network. The peak performance amounts to 7.4 PFlops.
In order to help users running efficiently on this new large system, HLRS and Cray offer an optimization workshop for applications running at scale. The intent of this workshop is to tune the performance of such codes by detecting, locating and solving bottlenecks.
By doing so, users can raise the volume as well as quality of their scientific findings while the costs (in terms of core hours) stay constant.
Together with specialists from Cray, the users will enhance the efficiency of their code. In order to do so, users will be required to bring their codes to the workshop ("bring your own code", “BYOC”). The specialists from Cray will assist in profiling application runs, analyzing the profile and optimizing the code based on the analysis. All categories of potential bottlenecks (CPU, memory subsystem, communication and I/O) will be addressed.
In addition to working on the code, there will be lectures on:
9:00 - 9:30 Local registration
9:30 - 17:30 Course
2nd and 3rd day
9:00 - 17:30 Course
9:00 - 16:30 Course
Target audience are users of applications which are already scaling to at least 24000 cores (1000 nodes). In order to attend the workshop, you will have to bring your own code including a dataset that can be used at scale. You will be required to do some work on your application before the workshop (details will follow). In order to do so, you should already have an account on HAZELHEN and your application should already be able to run on the system.Information on access to Hazelhen from Germany and Europe is available through this link.
Stefan Andersson, Michael Neff, and Mandes Schönherr (Cray)
Academic participants (i.e., members of universities or public research institutions) from EU or PRACE member countries: Please apply through the PATC web page. After your registration, you will receive an automated "congratulation"-email about your successful registration. This email implies that you have a guaranteed seat in the course and you should organize your travel.
All other participants (not from academia, or from outside Europe), or if the PATC web page is temporarily unavailable, please apply through this online registration form.
for registration is Mar. 28, 2016 (extended deadline).
Late registrations after the deadline are still possible but maybe with reduced quality of the handouts.
Members of German universities and public research institutes: none.
Members of universities and public research institutes within EU or PRACE-member countries: none.
Members of other universities and public research institutes: 120 EUR.
Others: 400 EUR.
(includes coffee breaks)
see our How to find us page.
HLRS is part of the Gauss Centre for Supercomputing (GCS), which is one of the six PRACE Advanced Training Centres (PATCs) that started in Feb. 2012. The mandate for the PATCs is as follows: "The PRACE Advanced Training Centres will serve as European hubs of advanced, world-class training for researchers working in the computational sciences." (see D3.2.3)
A part of this course is a PATC course, see also the PRACE Training Portal and Events. For participants from public research institutions in PRACE countries, the course fee is sponsored for this part of the course through the PRACE PATC program. For details, see the section about the course fee above.