A Grid-Enabled Engine for Delivering Custom Science- Grade Images on Demand

Slides:



Advertisements
Similar presentations
3 September 2004NVO Coordination Meeting1 Grid-Technologies NVO and the Grid Reagan W. Moore George Kremenek Leesa Brieger Ewa Deelman Roy Williams John.
Advertisements

Aug 3-4, 2003 Victoria1 Montage, Pegasus and ROME G. B. Berriman, J.C. Good, M. Kong, A. Laity IPAC/Caltech J. Jacob, A. Bergou, D. S. Katz JPL R. Williams.
Grid Astronomy with Image Federation Roy Williams Michael Feldmann California Institute of Technology.
Pegasus on the Virtual Grid: A Case Study of Workflow Planning over Captive Resources Yang-Suk Kee, Eun-Kyu Byun, Ewa Deelman, Kran Vahi, Jin-Soo Kim Oracle.
SkewReduce YongChul Kwon Magdalena Balazinska, Bill Howe, Jerome Rolia* University of Washington, *HP Labs Skew-Resistant Parallel Processing of Feature-Extracting.
Science Archives in the 21st Century Best Practices in Ingestion and Data Access at the NASA/IPAC Infrared Science Archive
The ADAMANT Project: Linking Scientific Workflows and Networks “Adaptive Data-Aware Multi-Domain Application Network Topologies” Ilia Baldine, Charles.
Ewa Deelman, Integrating Existing Scientific Workflow Systems: The Kepler/Pegasus Example Nandita Mangal,
Optimal Photometry of Faint Galaxies Kenneth M. Lanzetta Stony Brook University.
AME: An Any-scale many-task computing Engine Zhao Zhang, University of Chicago Daniel S. Katz, CI University of Chicago.
Grist: Grid Data Mining for Astronomy Joseph Jacob, Daniel Katz, Craig Miller, and Harshpreet Walia (JPL) Roy Williams, Matthew Graham, and Michael Aivazis.
20 Spatial Queries for an Astronomer's Bench (mark) María Nieto-Santisteban 1 Tobias Scholl 2 Alexander Szalay 1 Alfons Kemper 2 1. The Johns Hopkins University,
Montage: An Astronomical Image Mosaic Service for the NVO Anastasia C. Laity, IPAC Nate Anagnostou, IPAC Bruce Berriman,
Ewa Deelman, Optimizing for Time and Space in Distributed Scientific Workflows Ewa Deelman University.
Science Applications of Montage: An Astronomical Image Mosaic Engine ESTO G. Bruce Berriman, John Good and Anastasia Laity.
The Montage Image Mosaic Service: Custom Mosaics on Demand ESTO John Good, Bruce Berriman Mihseh Kong and Anastasia Laity IPAC, Caltech
Ewa Deelman, Clouds: An Opportunity for Scientific Applications? Ewa Deelman USC Information Sciences.
Montage: Experiences in Astronomical Image Mosaicking on the TeraGrid ESTO.
Pegasus: Mapping complex applications onto the Grid Ewa Deelman Center for Grid Technologies USC Information Sciences Institute.
An Astronomical Image Mosaic Service for the National Virtual Observatory
October, 2004ADASS XIV O1.31 Hires: Super-resolution for the Spitzer Space Telescope Charles Backus Velu Velusamy Tim Thompson John Arballo Description.
An Astronomical Image Mosaic Service for the National Virtual Observatory / ESTO.
Ewa Deelman Using Grid Technologies to Support Large-Scale Astronomy Applications Ewa Deelman Center for Grid Technologies USC Information.
Astronomical Image Mosaicking ESTO.
CREATING A MULTI-WAVELENGTH GALACTIC PLANE ATLAS WITH AMAZON WEB SERVICES G. Bruce Berriman, John Good IPAC, California Institute of Technolog y Ewa Deelman,
Ewa Deelman, Clouds: An Opportunity for Scientific Applications? Ewa Deelman USC Information Sciences.
Managing Workflows with the Pegasus Workflow Management System
Near Earth Objects Near-Earth Objects (NEOs) are comets and asteroids that have been nudged by the gravitational attraction of nearby planets into orbits.
Ewa Deelman, Pegasus and DAGMan: From Concept to Execution Mapping Scientific Workflows onto the National.
Pegasus A Framework for Workflow Planning on the Grid Ewa Deelman USC Information Sciences Institute Pegasus Acknowledgments: Carl Kesselman, Gaurang Mehta,
Why Build Image Mosaics for Wide Area Surveys? An All-Sky 2MASS Mosaic Constructed on the TeraGrid A. C. Laity, G. B. Berriman, J. C. Good (IPAC, Caltech);
The Grid is a complex, distributed and heterogeneous execution environment. Running applications requires the knowledge of many grid services: users need.
- page 1 July NHSC Mini-workshop PACS NASA Herschel Science Center PACS Photometer AORs How to Prepare an Observation with HSpot: 2 Science Use.
Rendering Adaptive Resolution Data Models Daniel Bolan Abstract For the past several years, a model for large datasets has been developed and extended.
Large-Scale Science Through Workflow Management Ewa Deelman Center for Grid Technologies USC Information Sciences Institute.
Running Scientific Workflow Applications on the Amazon EC2 Cloud Bruce Berriman NASA Exoplanet Science Institute, IPAC Gideon Juve, Ewa Deelman, Karan.
Combining the strengths of UMIST and The Victoria University of Manchester Utility Driven Adaptive Workflow Execution Kevin Lee School of Computer Science,
 The workflow description modified to output a VDS DAX.  The workflow description toolkit developed allows any concrete workflow description to be migrated.
Pegasus-a framework for planning for execution in grids Ewa Deelman USC Information Sciences Institute.
Pegasus: Planning for Execution in Grids Ewa Deelman Information Sciences Institute University of Southern California.
Pegasus: Mapping Scientific Workflows onto the Grid Ewa Deelman Center for Grid Technologies USC Information Sciences Institute.
Condor Week 2005Optimizing Workflows on the Grid1 Optimizing workflow execution on the Grid Gaurang Mehta - Based on “Optimizing.
Ten Years of Software Sustainability at The Infrared Processing and Analysis Center G. Bruce Berriman and John Good NASA Exoplanet Science Institute, Infrared.
Combining the strengths of UMIST and The Victoria University of Manchester Adaptive Workflow Processing and Execution in Pegasus Kevin Lee School of Computer.
Pegasus: Running Large-Scale Scientific Workflows on the TeraGrid Ewa Deelman USC Information Sciences Institute
Pegasus: Mapping complex applications onto the Grid Ewa Deelman Center for Grid Technologies USC Information Sciences Institute.
Hyperatlas Coregistered Federated Imagery Roy Williams Bruce Berriman George Djorgovski John Good Reagan Moore Caltech CACR Caltech IPAC Caltech Astronomy.
GriPhyN Virtual Data System Grid Execution of Virtual Data Workflows Mike Wilde Argonne National Laboratory Mathematics and Computer Science Division.
Virtual Observatories, Astronomical Telescopes and Instrumentation (August 2002)1 An Architecture for Access To A Compute Intensive Image Mosaic Service.
Pegasus-a framework for planning for execution in grids Karan Vahi USC Information Sciences Institute May 5 th, 2004.
1 GONG Magnetogram Data Products. 2 Sky-coordinate images Single-site or merged? –Single-site requires users to select. –Merge has lower noise, may require.
Planning Ewa Deelman USC Information Sciences Institute GriPhyN NSF Project Review January 2003 Chicago.
Pegasus: Planning for Execution in Grids Ewa Deelman, Carl Kesselman, Gaurang Mehta, Gurmeet Singh, Karan Vahi Information Sciences Institute University.
Bayesian Evolutionary Analysis by Sampling Trees (BEAST) LEE KIM-SUNG Environmental Health Institute National Environment Agency.
Young Star Disk Evolution in the Planet Formation Epoch Gregory Mace Mentor: Lisa Prato Northern Arizona University Physics and Astronomy.
1 USC Information Sciences InstituteYolanda Gil AAAI-08 Tutorial July 13, 2008 Part IV Workflow Mapping and Execution in Pegasus (Thanks.
Ewa Deelman, Managing Scientific Workflows on OSG with Pegasus Ewa Deelman USC Information Sciences.
Resource Allocation and Scheduling for Workflows Gurmeet Singh, Carl Kesselman, Ewa Deelman.
1 Performance Impact of Resource Provisioning on Workflows Gurmeet Singh, Carl Kesselman and Ewa Deelman Information Science Institute University of Southern.
EGEE-III INFSO-RI Enabling Grids for E-sciencE EGEE and gLite are registered trademarks Constraints on primordial non-Gaussianity.
Pegasus WMS Extends DAGMan to the grid world
Cloudy Skies: Astronomy and Utility Computing
(for the Algorithm Development Group of the US Planck Team)
Montage: An On-Demand Image Mosaic Service for the NVO
Pegasus and Condor Gaurang Mehta, Ewa Deelman, Carl Kesselman, Karan Vahi Center For Grid Technologies USC/ISI.
Clouds: An Opportunity for Scientific Applications?
University of Southern California
Pegasus Workflows on XSEDE
High Throughput Computing for Astronomers
Frieda meets Pegasus-WMS
Presentation transcript:

A Grid-Enabled Engine for Delivering Custom Science- Grade Images on Demand

Les Montagiers Nate Anagnostou IPAC Nate Anagnostou IPAC Bruce Berriman IPAC Bruce Berriman IPAC Attila Bergou JPL Attila Bergou JPL Ewa Deelman ISI Ewa Deelman ISI John Good IPAC John Good IPAC Joseph Jacob JPL Joseph Jacob JPL Daniel Katz JPL Daniel Katz JPL Carl Kesselman ISI Anastasia Laity IPAC Thomas Prince Caltech Gurmeet Singh ISI Mei-Hui Su ISI Roy Williams CACR

What is Montage? Delivers custom, science grade image mosaics Delivers custom, science grade image mosaics User specifies projection, coordinates, spatial sampling, mosaic size, image rotation User specifies projection, coordinates, spatial sampling, mosaic size, image rotation Preserve astrometry & photometric accuracy Preserve astrometry & photometric accuracy Modular “toolbox” design Modular “toolbox” design Loosely-coupled Engines for Image Reprojection, Background Rectification, Image Co-addition Loosely-coupled Engines for Image Reprojection, Background Rectification, Image Co-addition Control testing and maintenance costs Control testing and maintenance costs Flexibility; e.g custom background algorithm; use as a reprojection and co-registration engine Flexibility; e.g custom background algorithm; use as a reprojection and co-registration engine Implemented in ANSI C for portability Implemented in ANSI C for portability Enabling technology for multi-wavelength image federation Enabling technology for multi-wavelength image federation Public service will be deployed on the Teragrid Public service will be deployed on the Teragrid Order mosaics through web portal Order mosaics through web portal

Serial Processing of Images - Version 1.7  Available for download via a clickwrap license issued by Caltech at u User’s Guide User’s Guide Emphasizes accuracy in photometry and astrometry Emphasizes accuracy in photometry and astrometry Images processed serially Images processed serially Reprojection performed on surface of sphere Reprojection performed on surface of sphere BUT generality at expense of speed BUT generality at expense of speed AND mosaic size limited to available memory AND mosaic size limited to available memory Performance Mosaic of 54 2MASS images, 1 deg x 1 deg Pentium GHz, 1 GB RAM Reprojection 5500 s Background Modeling 55 s Rectification 28 s Co-addition 11 s

1 2 3 mProject 1mProject 2mProject 3 mDiff 1 2 mDiff 2 3 D 12 D 23 mFitplane D 12 mFitplane D 23 mBgModel ax + by + c = 0dx + ey + f = 0 a 1 x + b 1 y + c 1 = 0 a 2 x + b 2 y + c 2 = 0 a 3 x + b 3 y + c 3 = 0 mBackground 1mBackground 2mBackground mAdd Final Mosaic Parallel Processing in Montage 1 2 3

Montage on the Grid (Version 2.0) Pegasus Montage Workflow Specification (abstract) Grid Information Systems Information about available resources, data location Grid Condor DAGMan Maps an abstract workflow to an executable form Executes the workflow

Teragrid Performance Job # Jobs Avg Run Time (s) mAdd194 mBackground mBgModel1180 mConcatFit19 mDiff mFitplane mProject Data Transfer In Data Transfer Out deg x 2 deg 2MASS mosaic of M16 Workflow Run Time: 107 min (1515 jobs) Exposes highest degree of parallelism Overhead in scheduling lots of small jobs => Reduce overheads by aggregating modes

Custom Reprojection Algorithms Transform directly from input pixel to output pixels Transform directly from input pixel to output pixels Approach developed by Spitzer for tangent plane projections Approach developed by Spitzer for tangent plane projections Augment with “distorted” gnomonic projections Augment with “distorted” gnomonic projections Pixel locations distorted by small distance relative to image projection plane Pixel locations distorted by small distance relative to image projection plane Performance improvement in reprojection by x 30 Performance improvement in reprojection by x 30 AND Co-addition no longer limited by memory - output images read into memory one line at a time, co-added and written to disk => 30% performance degradation acceptable

Montage As A Reprojection Engine Application of general reprojection engine ZEA to CAR Supports a science service required by Herschel Serve spatial subsets of the images Dust emission, galactic emission and extinction along line of sight E/PO products - fold-out icosahedrons 100 µm sky; aggregation of COBE and IRAS maps. (Schlegel, Finkbeiner and Davis, 1998)

Generation of Large Scale Mosaics  2MASS 3-color mosaic of galactic plane 44 x 8 degrees; 36.5 GB per band; CAR projection 44 x 8 degrees; 36.5 GB per band; CAR projection 158,400 x 28,800 pixels; covers 0.8% of the sky 158,400 x 28,800 pixels; covers 0.8% of the sky NVO compliant-service NVO compliant-service 4 hours wall clock time on cluster of 4 x 1.4-GHz Linux boxes 4 hours wall clock time on cluster of 4 x 1.4-GHz Linux boxes Pilot project to estimate resources for all-sky mosaic Pilot project to estimate resources for all-sky mosaic

Spitzer IRAC Image Mosaics SWIRE mosaics SWIRE mosaics In mission planning, Montage used to build sky simulations in mission planning In mission planning, Montage used to build sky simulations in mission planning Fast background rectification and co- addition of in-flight images Fast background rectification and co- addition of in-flight images Part of a 2.5 GB IRAC image near the Tadpole Nebula