In order to unify our references to CORSIKA 8, we should all use "CORSIKA 8" (which is the clear result of a doodle poll). If you wish you can always use further attributes like "next-generation", "modern", or whatever you prefer, next to it.
For the list of scientific authors referred to as "CORSIKA 8 Project" for the ICRC2019, see ICRC2019 author list.
Here is the list of abstracts (to be) submitted to ICRC2019 with direct connection to CORSIKA 8:
First results of the CORSIKA 8 air shower simulation framework
Diego Melo, Maximilian Reininghaus, Felix Riehn, Ralf Ulrich, for the CORSIKA 8 project
Draft, 15.07.2019: main.pdf
CORSIKA 8 is a novel C++ framework for Monte Carlo simulations of particle cascades in air and other media. It is the designated successor of the well-known, long-standing Fortran version (CORSIKA 7), prepared to serve the astroparticle physics community over the next decades. Designed as a modular and open framework, the possible domains of applicability of CORSIKA 8 reach beyond its predecessor. In this contribution we give a status report of the project and demonstrate some of the first capabilities of CORSIKA 8, including the ability to define and combine arbitrary physical volumes/media in which showers develop. We present the particle interaction and physics models already implemented and point out differences in direct comparisons to other codes, i.e. to CORSIKA 7 and AIRES. A particular focus is made on the hadron and muon components of air shower cascades. Further plans and next milestones of the project are also presented.
Performance optimization of the air shower simulation code for the Cherenkov Telescope Array
L Arrabito1, K Bernloehr2, J Bregeon1, G Maier3 for the CTA Consortium P Langlois4, D Parello4, G Revy4 , A. Khattabi4 and R. Ulrich5 , names of who wishes to be named, for the CORSIKA 8 Project
1Laboratoire Univers et Particules, Université de Montpellier Place Eugène Bataillon - CC 72, CNRS/IN2P3, F-34095 Montpellier, France 2Max-Planck-Institut fur Kernphysik, P.O. Box 103980, D-69029 Heidelberg, Germany 3Deutsches Elektronen-Synchrotron, Platanenallee 6, 15738 Zeuthen, Germany 4DALI, Université de Perpignan Via Domitia 66860 Perpignan Cedex 9 France, LIRMM, CNRS : UMR 5506 - Université Montpellier 34095 Montpellier Cedex 5 France 5Institut für Kernphysik, Karlsruher Institut für Technologie (KIT), Karlsruhe, Germany
The Cherenkov Telescope Array (CTA), currently under construction, is the next-generation instrument in the field of very high energy gamma-ray astronomy. Construction is now under way with the first Large Size Telescope being commissioned in La Palma since the end of 2018. Scientific operations are expected to start in 2022 for a duration of about 30 years. In order to characterise the instrument response to the Cherenkov light emitted by extensive air showers, detailed Monte Carlo simulations will be regularly performed in parallel to CTA operations. The estimated CPU time associated to these simulations is very high, of the order of 200 millions HS06 hours per year. Reducing the CPU time devoted to simulations would allow either to reduce infrastructure cost or to better cover the large parameters phase space. In this contribution we focus on the main computing step (70% of the whole CTA simulations CPU time) implemented in the CORSIKA program. We present the first optimizations that we obtained applying vectorization techniques (SIMD instructions) on the module responsible for the production and propagation of Cherenkov photons in the atmosphere. Our approach takes care, as automatically as possible, of the hardware portability constraints introduced by the grid computing environment that hosts these simulations. A preliminary study on the potential performance gain obtained through computing precision tuning is also presented. Over the lifetime of CTA, simulations will be mostly based on the new generation CORSIKA 8 framework, we will work on an optimal Cherenkov production code in particular also within this new framework.
Technical foundations of CORSIKA 8: new concepts for scientific computing
Hans Dembinski, Lukas Nellen, Maximilian Reininghaus, Ralf Ulrich for the CORSIKA 8 Project
Draft, 15.07.2019: main.pdf
Draft, 17.07.2019: main.pdf
CORSIKA is the leading simulation code for air showers in the field of astroparticle physics. CORSIKA 8 is a new project aiming to make CORSIKA ready for the next decades of research; a rewrite of CORSIKA in modern C++ with a flexible, efficient, and modular design. CORSIKA 8 makes full use of open development, being a collaborative project with contributors from around the world. The modular design makes modifications and contributions very straightforward and lowers the technical barrier for users to become active developers. CORSIKA 8 is written in C++17, which brings new powerful features useful for scientific high-performance computing. We discuss work on its technical foundations, the geometry and quantity system (a quantity is a number with a dimension). The goal of these systems is to make physical and geometric calculations easy and safe in CORSIKA 8, while maintaining highest computational speed.
Performance optimization of the air shower simulation CORSIKA
** Type: Poster **
Dominik Baack, Wolfgang Rhode for the CORSIKA 8 Project
Draft, 15.07.2019: icrc_baak.pdf
With the steady increase in accuracy, size and complexity of astroparticle physics experiments the need for an extensive amount of high precision Monte Carlo simulations is rapidly growing. Contrary to the increasing demand is the demise of "Moore's law" which leads to situations where the system structure of high-performance computing is fundamentally changing and large amounts of money are invested in new infrastructure. In the field of astroparticle physics CORSIKA 7 is currently the most commonly used simulation program, therefore the presented work is focused on this software. All methods provided are also currently transferred to the new CORSIKA 8 framework which will replace CORSIKA 7 in the near future. Due to various constraints on hardware, geometry or physics no experiment is able to observe the full air shower particle cascade developing in the atmosphere. The removal of the non-visible phase space of the cascade at an early stage of the simulation has immense potential to reduce the expense of calculations without changing the results of the simulation for the experiment. Fast machine learning models allow the identification and removal of particles from those regions to speed up the simulations, in some cases, for example, neutrinos by orders of magnitude. First results are shown to demonstrate this technique. Furthermore, when showers are simulated with the IACT configuration around 75% of the time is spent on the Cherenkov photon creation and propagation. We also show results from parallelizing this part of the simulation on GPUs and CPUs with OpenCL.