Parts | Days | Selection | Search | Updates | Downloads | Help
T: Fachverband Teilchenphysik
T 15: GRID Computing
T 15.3: Talk
Monday, March 21, 2022, 16:45–17:00, T-H28
Modelling Large Scale Distributed Computing Systems for Identification of Efficient Architectures — •Maximilian Horzela, Manuel Giffels, Artur Gottmann, Günter Quast, and Achim Streit — Karlsruhe Institute of Technology
One approach to overcome the expected gap between available and required resources at future colliders like the High Luminosity Large Hadron Collider (HL-LHC) is to increase the efficiency of existing workflows, for example via caching of frequently used data. However, in complex environments with distributed computing systems and many users like the Worldwide LHC Computing Grid (WLCG), finding a solution is not trivial. Due to the complexity and size of such systems, it is not feasible to deploy several experimental test-beds at large scales.
Simulation of such systems has proved a scalable and versatile approach to identify efficient and practical computing architectures for High Energy Physics (HEP). A prominent example is the Monarc Simulation Framework, which lead to the present structure of the WLCG. However, the discontinuation of the old software and the demand for the ability to simulate recent scenarios including caching solutions requires a revision of suitable software.
In the context of this talk, a modern example for such applications based on the WRENCH and SimGrid simulation packages will be presented. Furthermore, first results obtained by the simulation will be shown.