HPC2N
High Performance Computing Center North
As electricity is becoming a costlier and costlier resource, new HPC
technologies are developed quickly to improve Supercomputers energy
efficiency (Flops/Watt ratio). It is now becoming very clear that hardware
is evolving rapidly towards many-core. Good examples are Nvidia GPUs, ARM
processors, Intel KNL, etc.
These new systems put pressure on scientists and developers who will
necessarily need to cope with more demanding needs in terms of scalability,
memory bandwidth per core, vectorization, etc.
Allinea's tools are designed to help address these challenges - and ensure
that applications reach the highest level of performance and scalability on
current and future machines.
During this technical session, an Allinea expert will introduce the tools
available to you and will train you to use them in order to resolve
challenges you face every day.
Preliminary schedule:
09:00 -09:30: Welcome
09:30 -09:45: Introduction to Allinea tools and latest changes
09:45 -10:15: Application analysis with Allinea Performance Reports
10:15 -11:00: Introduction to profiling with Allinea MAP
11:00 -12:00: Debug straightforward crashes with Allinea DDT
12:00 -13:00: Lunch break
13:00 -13:45: Automated debugging: a memory bug example
13:45 -14:15: Maximizing science efficiency with Allinea Performance Reports
14:15 -16:15: Bring your own code!
16:15 -16:30: Questions and wrap-up