Data Analysis on Ranger and Spur

Curriculum Overview

October 12-13, 2009
9:00am-5:00pm
Cornell University

Enrollment Closed. Class limit reached.

The Sun Constellation Cluster, Ranger, is a computational resource available to the National Science Foundation research community, providing users access to a compute cluster with a theoretical peak performance of 579 Teraflops. A description of Ranger can be found at the following URL:

http://www.tacc.utexas.edu/resources/hpcsystems/#ranger

Instructors will present topics covering the architecture and use of Ranger. This class will be of particular interest to computational scientists with large data computation or analysis. Members of the Cornell University Center for Advanced Computing (CAC) will be available during the entire training session to assist users in their code development and porting efforts.

The course will include lectures, labs, and discussions on:

Location

Engineering Library, Carpenter Hall (Blue Room)
Cornell University
Ithaca, NY

Cornell Maps - http://www.cornell.edu/maps/
Carpenter Hall - http://www.library.cornell.edu/node/163

Sugested Accommodations

Class attendees traveling to the training are responsible for arranging and paying for their own travel, daily expenses, and hotel accommodations.

Lodging information - http://www.visitithaca.com/lodging/
Getting to Ithaca - http://www.cornell.edu/visiting/ithaca/visiting.cfm

Agenda

Day 1 9:00am-5:00pm
General Topics9:00Welcome, Overview, and Hardware
9:30HPC Environment
11:00Data Transfer, Movement, and Storage
12:00Lunch
Data Analysis1:00Data and Database Formats
2:00Data Analysis with MATLAB
3:00Data Analysis with Python and R
4:00MapReduce with Hadoop
5:00Adjourn
Day 2 9:00am-5:00pm
Day 2: October 13, 2009
Visualization9:00Scientific Visualization
10:00ParaView
11:00VisIt
12:00Lunch
Code Improvement1:00Optimization
2:00Computational Steering
3:00Scientific Workflows and Provenance
5:00Adjourn