Page 1 PACS GRITS 17 June 2011 Herschel Data Analysis Guerilla Style: Keeping flexibility in a system with long development cycles Bernhard Schulz NASA.

Slides:



Advertisements
Similar presentations
A Workflow Engine with Multi-Level Parallelism Supports Qifeng Huang and Yan Huang School of Computer Science Cardiff University
Advertisements

Copyright © 2008 SAS Institute Inc. All rights reserved. SAS and all other SAS Institute Inc. product or service names are registered trademarks or trademarks.
Stephan Ott (Herschel Science Centre, ESTEC) on behalf of the Herschel Science Ground Segment Consortium
ORBIT DETERMINATION SOFTWARE DESIGN AT ESOC Ruaraidh Mackenzie, Frank Budnik.
Lecturer: Sebastian Coope Ashton Building, Room G.18 COMP 201 web-page: Lecture.
Operating Systems Concepts Professor Rick Han Department of Computer Science University of Colorado at Boulder.
Chapter 6: An Introduction to System Software and Virtual Machines
© 2008 IBM Corporation Behavioral Models for Software Development Andrei Kirshin, Dolev Dotan, Alan Hartman January 2008.
Astronomical GRID Applications at ESAC Science Archives and Computer Engineering Unit Science Operations Department ESA/ESAC.
Upcoming Enhancements to the HST Archive Mark Kyprianou Operations and Engineering Division Data System Branch.
JCE A Java-based Commissioning Environment tool Hiroyuki Sako, JAEA Hiroshi Ikeda, Visible Information Center Inc. SAD Workshop.
FP OntoGrid: Paving the way for Knowledgeable Grid Services and Systems WP8: Use case 1: Quality Analysis for Satellite Missions.
Hunt for Molecules, Paris, 2005-Sep-20 Software Development for ALMA Robert LUCAS IRAM Grenoble France.
Distributed Development: Lessons learned by Herschel GRITS 2011, June 17 Colin Borys.
Imperial College Tracker Slow Control & Monitoring.
SPACE TELESCOPE SCIENCE INSTITUTE Operated for NASA by AURA COS Pipeline Language(s) We plan to develop CALCOS using Python and C Another programming language?
NHSC SPIRE Webinar Q/A Session 20 th Nov 2013 Archive Query and Data Retrieval SPIRE Question and Answer Session Bernhard Schulz NHSC/IPAC.
Viewgraph 1 S. Ott Herschel Data Processing System NHSC Herschel Data Processing Workshop 26th August 2013 Overview of the Herschel Data Processing System.
PACS Hitchhiker’s Guide to Herschel Archive Workshop – Pasadena 6 th - 10 th Oct 2014 SPIRE Photometer Data Reprocessing Bernhard Schulz (NHSC/IPAC) on.
PACS NHSC Data Processing Workshop – Pasadena 26 th - 30 th Aug 2013 SPIRE Day team, goals, schedule, logistics Bernhard Schulz NHSC/IPAC.
PACS NHSC SPIRE Point Source Spectroscopy Webinar 21 March 2012 David Shupe, Bernhard Schulz, Kevin Xu on behalf of the SPIRE ICC Extracting Photometry.
CHAPTER TEN AUTHORING.
PACS Hitchhiker’s Guide to Herschel Archive Workshop – Pasadena 6 th - 10 th Oct 2014 Introduction to HIPE 12.1 (Part 2) Bernhard Schulz (NHSC/IPAC) on.
EÖTVÖS UNIVERSITY BUDAPEST Department of Physics of Complex Systems VO Spectroscopy Workshop, ESAC Spectrum Services 2007 László Dobos (ELTE)
PACS Hitchhiker’s Guide to the Herschel Science Archive Pasadena, 6 th -10 th October 2014 Quickstart Guide to HIPE and the HSA David Shupe User Support.
NHSC Workshop August 2013 NASA Herschel Science Center - page 1 PACS David R. Ardila User Support coordinator / NHSC Archive Scientist The Herschel Science.
Planetary Science Archive PSA User Group Meeting #1 PSA UG #1  July 2 - 3, 2013  ESAC PSA Archiving Standards.
Consortium Meeting La Palma October ICC Status Tanya Lim 1 SPIRE Consortium Meeting La Palma, Oct. 1 – ICC Status Tanya Lim RAL.
1 Performance Optimization In QTP Execution Over Video Automation Testing Speaker : Krishnesh Sasiyuthaman Nair Date : 10/05/2012.
Integrated Data Cycle Systems Harvey E. Rhody Chester F. Carlson Center for Imaging Science.
SDD/DFS R. Hook. VLT 2 nd Generation Instrumentation Pipelines, 18 Apr Richard Hook Sampo & ESO Reflex background, status and plans.
Consortium Meeting La Palma October PV-Phase & Calibration Plans Sarah Leeks 1 SPIRE Consortium Meeting La Palma, Oct. 1 – PV Phase and.
1 AutoCAD Electrical 2008 What’s New Name Company AutoCAD Electrical 2008 What’s New AMS CAD Solutions
- page 1 NHSC – DP workshop – Feb – N. Billot PACS PACS Photometer Standard Pipeline Level 0 to Level 1 processing: From raw to calibrated data cubes.
PACS NHSC Data Processing Workshop – Pasadena 10 th - 14 th Sep 2012 SPIRE AOTs, Products and Quick Look Tools Bernhard Schulz NHSC/IPAC on behalf of the.
NHSC Workshop August 2013 NASA Herschel Science Center - page 1 PACS David R. Ardila User Support coordinator / NHSC Archive Scientist Recap of Data Access.
Page 1 PACS HSC SPIRE Webinar 6 th Mar 2013 Introduction and General HIPE changes that impact SPIRE Bernhard Schulz NHSC/IPAC on behalf of the SPIRE ICC.
EVLA Software Bryan Butler. 2007May22EVLA SAGE Meeting2 Requirements and Goals of EVLA Software Maximize scientific throughput of the instrument At a.
NHSC PACS How to Use getObservation And saveObservation To Access and Save Herschel Data in HIPE Yi Mei (NHSC/Caltech) Y. Mei NHSC Data Workshop – August.
NASA Herschel Science Center - page 1 NHSC Cycle 1 Open Time Proposal Planning Workshop 3-4 June 2010 PACS Accepted Programs & Observations to Date B.
Debugging parallel programs. Breakpoint debugging Probably the most widely familiar method of debugging programs is breakpoint debugging. In this method,
WBL - 1 Observation Planning Workshop, Pasadena, CA 3-4 June 2010 Herschel Observer Support Overview William B. Latter NHSC Deputy Director.
6/1/2001 Supplementing Aleph Reports Using The Crystal Reports Web Component Server Presented by Bob Gerrity Head.
PACS NHSC Data Processing Workshop – Pasadena 10 th - 14 th Sep 2012 The SPIRE Destriper Bernhard Schulz NHSC/IPAC on behalf of the SPIRE ICC 1.
Page 1 PACS NHSC Webinar: New SPIRE Features in HIPE 10 6 th March 2013, Pasadena What’s New in SPIRE Photometer Pipeline in HIPE 10 Kevin Xu NHSC/IPAC.
March 2004 At A Glance autoProducts is an automated flight dynamics product generation system. It provides a mission flight operations team with the capability.
PACS ICC in Operations SVR-3 MPE, 8/9 Nov 2007 O.H. Bauer.
All rights reserved © Altec ExoMars 2018 Rover Operations Control Centre Science instruments data pipeline G. Martucci.
PACS Hitchhiker’s Guide to Herschel Archive Workshop – Pasadena 6 th - 10 th Oct 2014 The PACS Spectrometer: Overview and Products Roberta Paladini NHSC/IPAC.
PACS NHSC Data Processing Workshop – Pasadena 10 th - 14 th Sep 2012 HIPE Introduction and Setup for SPIRE David Shupe, NHSC/IPAC on behalf of the SPIRE.
Data Processing Workshop NHSC, Pasadena, CA 02 – 04 February 2011 Stephan Ott VG #1 PACS Overview and new developments in.
Page 1 NHSC PACS Web Tutorial PACS 301 nhsc.ipac.caltech.edu/helpdesk NHSC/PACS Web Tutorials Running the PACS Spectrometer pipeline for CHOP/NOD Mode.
NASA Herschel Science Center - page 1 PACS NHSC Cycle 1 Open Time Proposal Planning Workshop 3-4 June 2010 ESA AO and NASA Funding Call Phil Appleton and.
SPDF Science Advisory Group - September 29-30, 2005 Page 12/24/2016 9:09:48 PM Services of the Space Physics Data Facility (SPDF) / Sun-Earth Connection.
HIFI Tutorial 1: Running the HIFI Pipelines and Adaptations A.P.Marston ESAC 27 June 2013.
PACS page 1 NHSC SPIRE Data Processing Webinars 8 th Feb 2012 PACS page 1 Overview of SPIRE Photometer Pipeline C. Kevin Xu (NHSC/IPAC)
PACS NHSC Data Processing Workshop – Pasadena 26 th - 30 th Aug 2013 Preparing HIPE for SPIRE work David Shupe, NHSC/IPAC on behalf of the SPIRE ICC.
1 G4UIRoot Isidro González ALICE ROOT /10/2002.
ASI Review Padova ICC Contribution 1 WP3000 : Padova ICC Contribution Staff Alberto Franceschini (AF) [Professore Ordinario] Giulia Rodighiero (GR) [Ricercatore.
NHSC HIFI DP workshop Caltech, 8-9 March page 1 Pat Morris, NHSC DP Webinar, 9 March 2012 HIFI Spectral Maps in HIPE.
Slide 1 Chapter 8 Architectural Design. Slide 2 Topics covered l System structuring l Control models l Modular decomposition l Domain-specific architectures.
Chapter Goals Describe the application development process and the role of methodologies, models, and tools Compare and contrast programming language generations.
Development Environment
CST 1101 Problem Solving Using Computers
DELLSOFT Technologies Pvt. Ltd.
NHSC/PACS Web Tutorials
Design and Manufacturing in a Distributed Computer Environment
Existing Perl/Oracle Pipeline
Overview and new developments in Herschel Data Processing
CIS16 Application Development – Programming with Visual Basic
Presentation transcript:

page 1 PACS GRITS 17 June 2011 Herschel Data Analysis Guerilla Style: Keeping flexibility in a system with long development cycles Bernhard Schulz NASA Herschel Science Center

page 2 PACS Bernhard Schulz, NHSC GRITS 17 June 2011 HCSS The Herschel Common Science System (HCSS) software accompanied the development of the Herschel mission through all phases of development. Large multinational development effort in Java with >5*10 5 lines of code and >200 contributors Functions: –scientific proposal submission –observation scheduling, –spacecraft and instrument commanding –data downlink –data processing –instrument calibration –data archival

page 3 PACS Bernhard Schulz, NHSC GRITS 17 June 2011 Data Processing Jython is used as script language to invoke Java modules Many libraries: –Numerics Package with up to 5 dim arrays, fitting etc. –Plotting Package –Object Oriented Database Access Layer –Toolboxes: Mapping, Spectroscopy, FITS IO, etc… –Flight dynamics library (coordinates etc.) Other cool features –Task Framework –Data Products have FITS file equivalent –Calibration Source Database –Debugger HIPE: Interactive programming environment similar to Eclipse

page 4 PACS Bernhard Schulz, NHSC GRITS 17 June 2011 System Development The Herschel Science Centre (ESA), the Instrument Control Centres (HIFI, PACS and SPIRE) and the NHSC jointly manage and contribute to the Herschel Data Processing System All releases can be downloaded via – download.shtml Latest HIPE developer releases are available via – sclaimer.htmlhttp://herschel.esac.esa.int/CIB_di sclaimer.html Many intermediate developer builds on a daily basis Major new version so far every 3 months Now changing to 6 months cycle

page 5 PACS Bernhard Schulz, NHSC GRITS 17 June 2011 Libraries/Documentation

page 6 PACS Bernhard Schulz, NHSC GRITS 17 June 2011 Early Issues In the beginning the learning curve was really steep Hard to keep track of changes Highly nested data structures produced very complex Jython expressions to access data in memory Object oriented database system relatively complex Need for instrument builders to use a system that is in the process of development No time to make things “user friendly” print obs.refs["level1"].product.refs[2].product["signal"]["PSWE8"].data[0]

page 7 PACS Bernhard Schulz, NHSC GRITS 17 June 2011 The three data analysis choices Pipeline processing –Easy and straight forward –No flexibility Editing and running a script –Sophisticated and long learning curve –Full flexibility Interactive analysis with GUIs –Relatively easy to learn –Limited flexibility Straight pipeline results may be good enough for science analysis at a future date, but not yet. For astronomers with limited resources to learn the system, the GUI IA seems to be the optimal choice

page 8 PACS Bernhard Schulz, NHSC GRITS 17 June 2011 The Dilemma Need for rapid results during instrument performance verification and calibration Pipelines only available as Jython scripts Need to simplify the system for “normal” astronomers Need to reduce potential for human error during data processing Turnaround cycles slow and decision making slow with many contributing parties Ideological divide between “Command line fundamentalists” and “GUI Geeks”

page 9 PACS Bernhard Schulz, NHSC GRITS 17 June 2011 Guerilla Tactics If you are impatient: Do it yourself! At some point one realizes: –All the building blocks are there!! Take a weekend or two and have fun! ….provided you have such to spare… Result: SPIA (SPIRE Photometer Interactive Analysis)

page 10 PACS Bernhard Schulz, NHSC GRITS 17 June 2011 Tasks in SPIA

page 11 PACS Bernhard Schulz, NHSC GRITS 17 June 2011 File System SPIA Workflow Local Pool Level 0 Products Level 0.5 Products Level 1 Products Observation Context Level 2 Products Auxiliary Calibrati on FITS File HIPE Session Product Store Herschel Science Archive Internet Retrieve Data Load Calib Load Obs Save Results Observation Context Calibration Context

page 12 PACS Bernhard Schulz, NHSC GRITS 17 June 2011 Task Framework Clear structure requiring definition of input and output parameters with type, default values, tooltips, help Integrated into HIPE (registering) Common look default GUI Records history Tasks can be executed as subroutines in other tasks or scripts for batch processing Disadvantage: static GUI that can not change based on user input

page 13 PACS Bernhard Schulz, NHSC GRITS 17 June 2011

page 14 PACS Bernhard Schulz, NHSC GRITS 17 June 2011

page 15 PACS Bernhard Schulz, NHSC GRITS 17 June 2011 Availability Distribution from NHSC Wiki –HIPE Plugin –User’s Manual –Video Tutorials –Publication ADASS 2010 proceedings

page 16 PACS Bernhard Schulz, NHSC GRITS 17 June 2011 Conclusive Remarks HCSS has become a system with huge potential Parallel development of software system and hardware is a successful concept Needs of three communities must be balanced –Developers, Instrument scientists, Astronomers You can not start early enough in a mission with planning for the software environment. Flexibility in a system is very useful –Helps to make up for omissions in the plan –Better: if a clear vision for limits of scope and need dates of functionalities are established early. A user friendly environment that hides most complexities must be part of the plan