Description

Marconi is the new CINECA class Tier-0 supercomputer, based on the next-generation of the Intel® Xeon Phi™ product family alongside with Intel® Xeon® processor E5-2600 v4 product family. It has been co-designed by Cineca on the Lenovo NeXtScale architecture. Its full deployment will take place in two steps: the first one will start mid-2016 and will be completed within mid-2017, and will have an estimated peak performance of about 20 Pflop/s with storage capacity of 20 petabyte, while maintaining a low energy consumption; the second step will start not late than 2019 and will end within 2020, with an expected final peak performance of 50/60 Pflop/s. This new system will offer the scientific community a technologically advanced and energy-efficient high performance computing system. The present course intends to support the scientific community to efficiently exploit the Marconi system. More precisely, the course aims at providing a full description of the Marconi configuration at Cineca, with special emphasis on main crucial aspects for users and application developers. For instance, details about compilation, debugging and optimization procedures will be provided, together with an overview of the available libraries, tools and applications currently available on the system. Examples of submission jobs will be discussed, together with scheduler (PBS) commands and queue definitions.

Topics:

Overview of MARCONI architecture and software (hardware components, network and partitioning, type of nodes and software stack) Developing applications for MARCONI (compilers, libraries, available debugging and profiling tools) Running and monitoring jobs on MARCONI (modules environment @ CINECA, PBS queueing system, job script examples)

Target audience:

Users and developers on the MARCONI Tier-0 system.

Pre-requisites:

Basic knowledge of Linux/UNIX.

https://events.prace-ri.eu/event/513/