Document Actions

CScADS 10

by Jennifer Harris last modified 2010-07-21 10:31

Organizers



Abstract

Petascale computer systems are now available to the DOE science community; a contract to install a 20 petaflop machine at Lawrence Livermore National Laboratory in 2011 has been announced. Compared to systems of the past, these machines represent a dramatic increase in scale and complexity. Making effective use of systems at this scale will require performance tools that help application developers by pinpointing and explaining application performance bottlenecks. Two performance challenges for applications are to make effective use of processor nodes, and to keep parallelization overhead low enough so that extreme-scale parallelism is practical.

Providing tools that make it possible to address these problems is requiring increasingly sophisticated methods for instrumentation, measurement, analysis and modeling of application performance. Meeting these challenges will require tools with unprecedented capabilities.

Today, research and development in the area of performance tools for parallel systems is fragmented. Individual groups have developed tools and components with unique capabilities. Working together to capitalize these capabilities into shared community infrastructure would reduce duplication of effort, accelerate development of effective tools with increasing sophistication, and simplify the process of developing tools for new systems as they emerge.

The goal of this workshop is to bring together tools researchers to discuss challenges of performance analysis and debugging on emerging petascale systems, review ongoing research, and work together to tackle the challenges of performance analysis on petascale systems together as a community. Attendees at the workshop will be tools developers in the areas of tools infrastructure, debugging, performance instrumentation, measurement, analysis, and visualization. We expect that about a third of the workshop will consist of opportunities to work together in smaller groups to make progress on tackling problems of mutual interest.

Specific aims of the workshop include

  • Identifying performance tool capabilities needed to analyze the spectrum of issues that arise with petascale platforms.
  • Discussing emerging capabilities of research and commercial tools in this space.
  • Identifying common needs, functionality, and opportunities for sharing infrastructure.
  • Discussing design aspects of sharable components.
  • Developing standardized interfaces to facilitate sharing.

Agenda

Sunday, August 1, 2010
Dinner on your own
All Technical Sessions in Summit Room
 
 
Monday, August 2, 2010
7:30 AM Breakfast Buffet - Aerie Private Dining
9:00 AM The Latest in the Deconstruction of Dyninst, Matt Legendre, Bill Williams, Madhavi Krishnan, Drew Bernat, Barton Miller (University of Wisconsin)
 
10:00 AM An Embedded Dyninst, Jeffrey Hollingsworth, Nick Rutar (University of Maryland)
 
10:30 AM Morning Break
11:00 AM Update on the Cray Scalable Tools, Luiz DeRose, Heidi Poxon (Cray Inc.)
 
11:45 AM Configurable Instrumentation Components and Their Use by Scalasca, Markus Geimer (Julich Supercomputing Centre)
 
12:30 PM Lunch -Aerie Private Dining
2:00 PM Emerging Capabilities in HPCToolkit, John Mellor-Crummey, Nathan Tallent, Xu Liu (Rice University)
 
3:00 PM Afternoon Break
3:30 PM Performance Diagnosis for Hybrid Multicore Environments, Karen Karavanic (Portland State University)
 
4:00 PM Performance Analysis of DOE Workloads in HPC and Cloud Environments, David Skinner (Lawrence Berkeley Lab)
 
4:30 PM MPIT: A New Interface for Performance Tools in MPI-3, Martin Schulz (Lawrence Livermore National Lab)
 
6:30 PM Dinner - Aerie Private Dining or Atrium Patio (weather permitting)
 
 
Tuesday, August 3, 2010
7:30 AM Breakfast - Aerie Private Dining
9:00 AM Trace Analysis of PFLOTRAN on the Cray XT5 using Extrae and Paraver, Kevin Huck (Barcelona Supercomputer Center)
 
9:45 AM TAU Potpourri and Working with Open Components, Interfaces, and Environments, Allen Malony (University of Oregon)
 
10:30 AM Morning Break
11:00 AM Performance Prediction of Tree-Based Overlay Networks for Tools and Applications, Phil Roth (Oak Ridge National Lab)
 
11:45 AM Improving Tool Startup and Runtime Performance, Dorian Arnold (University of New Mexico)
 
12:30 PM Lunch
2:00 PM Progress and Status of the perf_events Linux kernel Interface for Performance Monitoring , Stephane Eranian (Google)
 
2:30 PM Thoughts on HPC Performance Analysis and SW Optimization on Teraflop Scale Clusters, Dave Levinthal (Intel)
 
3:00 PM Afternoon Break
3:30 PM Recent Compiler-Based Tools Support in ROSE, Dan Quinlan (LLNL)
 
4:00 PM Status and Internals for the Component Based Tool Framework, Jim Galarowicz (Krell Institute)
 
4:30 PM PAPI-C: What Can Performance Components Do for You? , Dan Terpstra (University of Tennessee)
 
6:30 PM Dinner - Aerie Private Dining or Atrium Patio (weather permitting)
 
 
Wednesday, August 4, 2010
7:30 AM Breakfast - Aerie Private Dining
9:00 AM Working Groups - Planning and Initial Meetings
 
10:30 AM Morning Break
11:00 AM Working Groups
 
12:30 PM Lunch
2:00 PM Working Groups
 
6:30 PM Dinner - Aerie Private Dining or Atrium Patio (weather permitting)
 
 
Thursday, August 5, 2010
7:30 AM Breakfast - Aerie Private Dining
9:00 AM Working Groups
 
10:30 AM Morning Break
11:00 AM Working Group Presentations
 
12:30 PM Lunch - Aerie Private Dining
 
 

Sponsors

This workshop is sponsored by the Center for Scalable Application Development Software, with funding from the Scientific Discovery through Advanced Computing (SciDAC) program.

« April 2018 »
Su Mo Tu We Th Fr Sa
1234567
891011121314
15161718192021
22232425262728
2930
 

Powered by Plone

CScADS Collaborators include:

Rice University ANL UCB UTK WISC