Prospects of DDP application to the data processing software: discussion

• Bartosz Sobol (Jagiellonian University)
• Dirk Hutter (Frankfurt Institute for Advanced Studies (FIAS))
• Grzegorz Korcyl (Jagellonian University Krakow(JUK))
• Jan de Cuveland (Frankfurt Institute for Advanced Studies (FIAS))
• Pierre-Alain Loizeau (GSI Helmholtzzentrum für Schwerionenforschung GmbH(GSI))
• Sergei Zharko (GSI Helmholtzzentrum für Schwerionenforschung GmbH(GSI))
• Sergey Gorbunov (GSI Helmholtzzentrum für Schwerionenforschung GmbH(GSI))
• Volker Friese (GSI Helmholtzzentrum für Schwerionenforschung GmbH(GSI))
• Florian Uhlig (GSI Helmholtzzentrum für Schwerionenforschung GmbH(GSI))

The area of discussion includes the following data processing steps: - unpacking - local timeslice reconstruction - global timeslice reconstruction - event building - local and global event reconstruction - event selection Discussion points (extendable): - pros and cons of the DDP (experience from other experiments) - possible topologies of DDP - messaging patterns between the OP steps - parallelization model (CPU, GPU) Update: DDS -> DDP

AoB