« Back to Events

Call for Participation: ISPASS 2013 – Tutorial on Modeling Exascale Applications with SST/macro and Eiger

iCal Import
Start:
April 21, 2013 1:30 am
End:
April 21, 2013 5:30 am
Venue:
Austin, Texas

Submitted by Eric Anger
http://casl.gatech.edu/research/eiger-tutorial/
Tutorial on Modeling Exascale Applications with SST/macro and Eiger

Held in conjunction with the ISPASS 2013 Conference
Sunday, April 21 2013 1:30 PM to 5:30 PM.

This tutorial will present attendees with the techniques and methodologies to
leverage the SST/macro simulator and the Eiger performance modeling framework
for modeling large scale applications on upcoming supercomputer hardware.

In high performance computing (HPC), the importance of fast, accurate models
that can be used at large scales is increasing as we move towards the next
frontier of exascale. The SST/macro simulator provides HPC engineers the ability
to explore current and future machine architectures and programming models with
coarse grain on-line simulation that executes real application code to reproduce
communication behavior, a vital part of the hardware/software codesign process.
Instead of using cycle-accurate simulation of instructions executing on
processors, SST/macro relies on analytical models for computation performance.
The Eiger Performance Modeling Framework enables the generation of performance
models by applying statistical techniques from the field of machine learning on
empirically acquired performance data. While macro-scale simulation can provide
a reasonable overview of system wide phenomena, Eiger can leverage data acquired
from micro-scale sources to create models that form elements of large scale
simulations in SST/macro. The Eiger methodology constructs models from
aggregated data from micro-scale sources such as simulators, emulators, and
runtime instrumentation.

This presentation is geared towards domain experts and HPC hardware designers,
as well as students and researchers whose work requires exploration of
programming models, interaction between computation and communication, and
data-driven modeling techniques for large scale systems. These tools are geared
toward ease of use and rapid iteration, allowing area experts to generate
verbose performance models without requiring intricate knowledge of every facet
of the computing environment. This tutorial will require only a basic level of
programming skill.

More details about the tutorial can be found at the tutorial website.