Luke Winslow OWI alacarte Overview What is Parallel Computing Example Uses Intro to Condor Demo Parallel Computing Cluster Why would I need this Dealing with Large Datasets Lake Shape Characteristics ID: 575133
Download Presentation The PPT/PDF document "Introduction to High Throughput Computin..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Introduction to High Throughput Computing
Luke Winslow
OWI a-la-carteSlide2
OverviewWhat is Parallel ComputingExample UsesIntro to CondorDemoSlide3
Parallel ComputingSlide4
ClusterSlide5
Why would I need this?Slide6
Dealing with Large DatasetsSlide7
Lake Shape CharacteristicsSlide8
Bootstrapping AnalysesSlide9Slide10
Bootstrapping AnalysesSlide11
Large-scale simulationsSlide12
HTCondorScavenging wasted compute cyclesRuns on basically any normal computerComputer Labs are perfect
A lot of dedicated clusters existSlide13
Known by many namesHTCondorSLURMLSF
PBS SGESlide14
ParallelizabilityFor i=1:1000
var(i) = i;End
For
i=1:1000
var
(i) =
var
(i-1) + i;
EndSlide15
Important Term:Embarrassingly parallelSlide16
Example1D model run on all lakes on a landscape
Models don’t need to “talk”Slide17
Counter example
Modeling a connected
hydrologic network
Stream model must pass info to lake model
Lake model must info to downstream model
Etc
, etcSlide18
parallel package in RInteractive demoCross your fingersSlide19
Further informationSome working exampleshttps://github.com/USGS-OWI/condor-examples
HTCondor Helphttps://research.cs.wisc.edu/htcondor/Feel free to contact us
ldecicco@usgs.gov
lwinslow@usgs.govSlide20
Questions?Slide21Slide22