Back to Results
First PageMeta Content
Market research / Variance reduction / Sample size determination / Stratified sampling / AMPL / Simple random sample / Profiling / Cluster sampling / Statistical power / Statistics / Sampling / Data analysis


Scalable Methods for Monitoring and Detecting Behavioral Equivalence Classes in Scientific Codes ∗ Todd Gamblin Rob Fowler
Add to Reading List

Document Date: 2008-03-21 16:17:19


Open Document

File Size: 408,80 KB

Share Result on Facebook

Company

IBM / Lawrence Livermore National Laboratory / Report Guide Group / Intel / Amdahl / /

/

Facility

Renaissance Computing Institute University of North Carolina / AMPL library / SciDAC Performance Engineering Research Institute / Profiling Library / Chapel Hill Abstract / University of Oregon’s Tuning / /

IndustryTerm

gas dynamics problem / task-level tracing tools / scientific applications / torus network / gas dynamics benchmark / online monitoring / large-scale computing systems / data collection tools / to parallel applications / large systems / large-scale scientific applications / wellknown scientific applications / well known algorithm / sPPM algorithm / concurrent systems / large parallel applications / test applications / petascale systems / larger systems / /

OperatingSystem

Linux / Microsoft Windows / /

Organization

University of Oregon / SciDAC Performance Engineering Research Institute / University of North Carolina / Department of Energy / /

Person

Parallel File / Lu / Daniel A. Reed / Todd Gamblin Rob Fowler Daniel / /

/

Position

Prime Minister / General / engineer / current performance leader / analyst / /

Product

BlueGene/L system / Pentax K-x Digital Camera / Woodcrest / Latte W10 Portable Audio Device / IBMs xlC / BlueGene/L / /

ProgrammingLanguage

Fortran 77 / AMPL / C++ / /

ProvinceOrState

Oregon / Manitoba / North Carolina / /

Technology

RAM / 3-D / Clustering algorithms / Linux / sPPM algorithm / 1 Introduction Processor / shared memory / well known algorithm / 1 / 000 processors / gigabit / simulation / 212 / 992 processors / gigabit ethernet / /

SocialTag