Sessions 2016

This is the general overview of the school’ sessions, detailed in the program. A brief description of the sessions is given in the table below:

HPC overview
Parallel programmingNext generation
ESMFFuture directions in HPC
11:00-13:00Good practices
Parallel programmingGPUAdjointFuture directions in HPC
Data management
Data management
Parallel programming
Parallel programmingGPUAdjoint
18:00-20:00RegistrationIce breaker
Student's poster sessions
City tourConcert "Corale Novantanove"
20:00-City tourDinner
Color legend:Theoretical lecturePractical sessionSocial event

Short description of sessions

  • HPC overview: Some history, description of main computer architectures and clusters, overview of world’s top computers, required resources to develop a High Performance Computing service in terms of fundings, professional profiles, infrastructure.
  • Good practices: Hints on good programming practices to make codes (e.g. Fortran) more efficient and flexible.
  • Batch: Introduction to queuing (batch) systems on clusters.
  • Cloud: Overview of cloud services and basics for their effective use (setting up environment, virtual machines, etc.). Techniques for effective storage of large datasets.
  • Data management: issues and solutions for efficient access and analysis of large amount of data. Earth System Grid Federation (ESGF), Ophidia.
  • Parallel programming: Parallel programming with shared and distributed resources. OpenMP and Message Passing Interface (MPI). Tips and tricks for parallelization of Earth System Science codes.
  • Next Generation/GPU: New generation computer architectures for parallel calculus. Many Integrated Core (MIC) architectures, Graphical Processing Units (GPU) programming.
  • ESMF: Earth System Modelling Framework, tips and tricks for effective use.
  • Adjoint: adjoint model formulation and development for sensitivity simulations.
  • Future directions in HPC: overview of upcoming architectures and programming tools.