University of California, Davis Daniel Zinn 1 University of California, Davis Daniel Zinn 1 Parallel Virtual Machines in Kepler Daniel Zinn Xuan Li Bertram.

Slides:



Advertisements
Similar presentations
Nimrod/K: Towards Massively Parallel Dynamic Grid Workflows David Abramson, Colin Enticott, Monash Ilkay Altinas, UCSD.
Advertisements

A Workflow Engine with Multi-Level Parallelism Supports Qifeng Huang and Yan Huang School of Computer Science Cardiff University
Legacy code support for commercial production Grids G.Terstyanszky, T. Kiss, T. Delaitre, S. Winter School of Informatics, University.
Parallel Virtual Machine Rama Vykunta. Introduction n PVM provides a unified frame work for developing parallel programs with the existing infrastructure.
SALSA HPC Group School of Informatics and Computing Indiana University.
1 OBJECTIVES To generate a web-based system enables to assemble model configurations. to submit these configurations on different.
Jianwu Wang, Daniel Crawl, Ilkay Altintas San Diego Supercomputer Center, University of California, San Diego 9500 Gilman Drive, MC 0505 La Jolla, CA ,
An Automata-based Approach to Testing Properties in Event Traces H. Hallal, S. Boroday, A. Ulrich, A. Petrenko Sophia Antipolis, France, May 2003.
Distributed, parallel web service orchestration using XSLT Peter Kelly Paul Coddington Andrew Wendelborn.
Experiences in Integration of the 'R' System into Kepler Dan Higgins – National Center for Ecological Analysis and Synthesis (NCEAS), UC Santa Barbara.
6th Biennial Ptolemy Miniconference Berkeley, CA May 12, 2005 Distributed Computing in Kepler Ilkay Altintas Lead, Scientific Workflow Automation Technologies.
Ngu, Texas StatePtolemy Miniconference, February 13, 2007 Flexible Scientific Workflows Using Dynamic Embedding Anne H.H. Ngu, Nicholas Haasch Terence.
7th Biennial Ptolemy Miniconference Berkeley, CA February 13, 2007 Scheduling Data-Intensive Workflows Tim H. Wong, Daniel Zinn, Bertram Ludäscher (UC.
1 Application Specific Module for P-GRADE Portal 2.7 Application Specific Module overview Akos Balasko MTA-SZTAKI LPDS
Biology.sdsc.edu CIPRes in Kepler: An integrative workflow package for streamlining phylogenetic data analyses Zhijie Guan 1, Alex Borchers 1, Timothy.
The SAM-Grid Fabric Services Gabriele Garzoglio (for the SAM-Grid team) Computing Division Fermilab.
January, 23, 2006 Ilkay Altintas
DIANE Overview Germán Carrera, Alfredo Solano (CNB/CSIC) EMBRACE COURSE Monday 19th of February to Friday 23th. CNB-CSIC Madrid.
 Cloud computing  Workflow  Workflow lifecycle  Workflow design  Workflow tools : xcp, eucalyptus, open nebula.
1 port BOSS on Wenjing Wu (IHEP-CC)
PVM. PVM - What Is It? F Stands for: Parallel Virtual Machine F A software tool used to create and execute concurrent or parallel applications. F Operates.
A Web 2.0 Portal for Teragrid Fugang Wang Gregor von Laszewski May 2009.
Composing Models of Computation in Kepler/Ptolemy II
A Hybrid Decomposition Scheme for Building Scientific Workflows Wei Lu Indiana University.
Nimrod/K using Opal Services for Virtual Screening David Abramson, Ilkay Altintas, Daniel Crawl, Wilfred Li, Jane Ren, Jianwu Wang, Colin Enticott(presenter)
1 Overview of the Application Hosting Environment Stefan Zasada University College London.
ITPA/IMAGE 7-10 May 2007 Software and Hardware Infrastructure for the ITM B.Guillerminet, on behalf of the ITM & ISIP teams (P Strand, F Imbeaux, G Huysmans,
Contents 1.Introduction, architecture 2.Live demonstration 3.Extensibility.
Accelerating Scientific Exploration Using Workflow Automation Systems Terence Critchlow (LLNL) Ilkay Altintas (SDSC) Scott Klasky(ORNL) Mladen Vouk (NCSU)
Jan Hatje, DESY CSS ITER March 2009: Technology and Interfaces XFEL The European X-Ray Laser Project X-Ray Free-Electron Laser 1 CSS – Control.
11 CORE Architecture Mauro Bruno, Monica Scannapieco, Carlo Vaccari, Giulia Vaste Antonino Virgillito, Diego Zardetto (Istat)
Grid Computing Research Lab SUNY Binghamton 1 XCAT-C++: A High Performance Distributed CCA Framework Madhu Govindaraju.
1 Ilkay ALTINTAS - July 24th, 2007 Ilkay ALTINTAS Director, Scientific Workflow Automation Technologies Laboratory San Diego Supercomputer Center, UCSD.
Shannon Hastings Multiscale Computing Laboratory Department of Biomedical Informatics.
Resource Brokering in the PROGRESS Project Juliusz Pukacki Grid Resource Management Workshop, October 2003.
The Grid computing Presented by:- Mohamad Shalaby.
David Adams ATLAS ADA, ARDA and PPDG David Adams BNL June 28, 2004 PPDG Collaboration Meeting Williams Bay, Wisconsin.
Tool Integration with Data and Computation Grid GWE - “Grid Wizard Enterprise”
Wrapping Scientific Applications As Web Services Using The Opal Toolkit Wrapping Scientific Applications As Web Services Using The Opal Toolkit Sriram.
ServiceSs, a new programming model for the Cloud Daniele Lezzi, Rosa M. Badia, Jorge Ejarque, Raul Sirvent, Enric Tejedor Grid Computing and Clusters Group.
Convert generic gUSE Portal into a science gateway Akos Balasko 02/07/
Self-assembling Agent System Presentation 1 Donald Lee.
Moby Web Services Iván Párraga García MSc on Bioinformatics for Health Sciences May 2006.
Kepler includes contributors from GEON, SEEK, SDM Center and Ptolemy II, supported by NSF ITRs (SEEK), EAR (GEON), DOE DE-FC02-01ER25486.
11 CORE Architecture Mauro Bruno, Monica Scannapieco, Carlo Vaccari, Giulia Vaste Antonino Virgillito, Diego Zardetto (Istat)
Experiment Management System CSE 423 Aaron Kloc Jordan Harstad Robert Sorensen Robert Trevino Nicolas Tjioe Status Report Presentation Industry Mentor:
University of California, Davis Daniel Zinn 1 University of California, Davis Daniel Zinn 1 Daniel Zinn Bertram Ludäscher University of California at Davis.
Distributed Computing With Triana A Short Course Matthew Shields, Ian Taylor & Ian Wang.
Core Java Introduction Byju Veedu Ness Technologies httpdownload.oracle.com/javase/tutorial/getStarted/intro/definition.html.
A Demonstration of Collaborative Web Services and Peer-to-Peer Grids Minjun Wang Department of Electrical Engineering and Computer Science Syracuse University,
August 2003 At A Glance The IRC is a platform independent, extensible, and adaptive framework that provides robust, interactive, and distributed control.
Development of e-Science Application Portal on GAP WeiLong Ueng Academia Sinica Grid Computing
Convert generic gUSE Portal into a science gateway Akos Balasko.
Tool Integration with Data and Computation Grid “Grid Wizard 2”
ECHO A System Monitoring and Management Tool Yitao Duan and Dawey Huang.
Satisfying Requirements BPF for DRA shall address: –DAQ Environment (Eclipse RCP): Gumtree ISEE workbench integration; –Design Composing and Configurability,
Lesson 1 1 LESSON 1 l Background information l Introduction to Java Introduction and a Taste of Java.
2.1 Silberschatz, Galvin and Gagne ©2009 Operating System Concepts – 8 th Edition System Programs (p73) System programs provide a convenient environment.
Application Specific Module Tutorial Zoltán Farkas, Ákos Balaskó 03/27/
Copyright 2007, Information Builders. Slide 1 iWay Web Services and WebFOCUS Consumption Michael Florkowski Information Builders.
ACCESSING DATA IN THE NIS USING THE KEPLER WORKFLOW SYSTEM Corinna Gries.
4000 Imaje 4020 – Software Imaje 4020 – Content ■ Content of Chapter Software: 1. Flash Up 2. Netcenter 3. FTP 4. Active X 5. XCL commands 6. Exercise.
© Geodise Project, University of Southampton, Workflow Support for Advanced Grid-Enabled Computing Fenglian Xu *, M.
SHIWA Desktop Cardiff University David Rogers, Ian Harvey, Ian Taylor, Andrew Jones.
Convert generic gUSE Portal into a science gateway Akos Balasko.
Scientific workflow in Kepler – hands on tutorial
Module 01 ETICS Overview ETICS Online Tutorials
Outline Chapter 2 (cont) OS Design OS structure
GGF10 Workflow Workshop Summary
Presentation transcript:

University of California, Davis Daniel Zinn 1 University of California, Davis Daniel Zinn 1 Parallel Virtual Machines in Kepler Daniel Zinn Xuan Li Bertram Ludaescher Thursday, April 16, 2009, Berkeley, California Eighth Biennial Ptolemy Mini-conference

University of California, Davis Daniel Zinn 2 Outline Motivation & Related Work Lightweight Parallel PN Engine Kepler PPN Director Demo Future Directions & Conclusion

University of California, Davis Daniel Zinn 3 Motivation Kepler used to automate and build scientific apps maybe compute intensive maybe data intensive long-running Speed-up executions to save scientists time … by leveraging distributed resources

University of California, Davis Daniel Zinn 4 Distribution Efforts in Kepler / PTII Remote execution of a complete workflow Hydrant (Tristan King) Web service for remote execution (Jianwu Wang) Parameter sweeps with Nimrod/K (Colin Enticott, David Abramson, Ilkay Altintas) Distribution within actors “Plumping Workflows” with ad-hoc ssh-control (Nortbert Podhorszki) Globus actors in Kepler: GlobusJob, GlobusProxy, GridFTP, GridJob. GLite actors available through ITER Webservice executions by actors Distribution of few or all actors Distributed SDF Director (Daniel Cuadrado) Pegasus Director (Daniel Cuadrado and Yang Zhao) Master-Slave Distributed Execution (Chad Berkley and Lucas Gilbert) with DistributedCompositeActor PPN Director (Daniel Zinn and Xuan Li) Thanks to Jianwu for help with overview

University of California, Davis Daniel Zinn 5 Outline Motivation & Related Work Parallel PN Engine Kepler PPN Director Demo Future Directions & Conclusion

University of California, Davis Daniel Zinn 6 Lightweight Parallel PN Engine (LPPN) Motivation PN as inherently parallel MoC Build simple, efficient distributed PN-engine Design Requirements KISS Avoid centralization as much as possible Provide Actor and Port abstractions Allow actors being written in different languages “Experimentation Platform” for scheduling, data routing, … Design Priniciples One actor = one process Communication between actors Central component only for setup, termination detection, …

University of California, Davis Daniel Zinn 7 LPPN – Technology Choices C++ for core libraries Actor, Port, Token as C++ classes Parallel Virtual Machine (PVM) for parallelization Thin layer on top of machine clusters (pool of hosts) Message passing Implemented simple RPC on top of this SWIG for adding higher-languages above core Perl/Python interfaces for writing actors Perl interfaces for composing and starting workflow Java interface for composing, starting, monitoring workflows

University of California, Davis Daniel Zinn 8 LPPN – C++ Core Library: Actor

University of California, Davis Daniel Zinn 9 LPPN – C++ Core Library: Ports Ports are parameterized by the data type sent through the port. Data types: string int, double, … Custom structs BLOB Transfer via PVM messages

University of California, Davis Daniel Zinn 10 LPPN – C++ Core Library: Tokens BLOB_Token to encapsulate BLOB as files Construct from files LinkTo(path) File data is known to the workflow system Data is sent via rsync/scp by the LPPN system Each actor has private workspace in file system Generic Token Polymorphic Tokens for COMAD workflows Open, Close, BLOB, …

University of California, Davis Daniel Zinn 11 LPPN – Actor Communication (Ports) During Setup-Time Connect OutPorts to InPorts (1:1 mapping) During Runtime Port implementation sends data from port to port BlobToken-Ports handle file movement Block on read Block on write when buffer full (still fixed-size) Tokenbuffers in PVM, file-buffers in actor directory Gathering of statistics Actor status, token counts, …

University of California, Davis Daniel Zinn 12 Command-line Actor and Actors in Perl Example: ConvertResize Actor Automatically create input and output ports Read from input ports, call command, write to output ports

University of California, Davis Daniel Zinn 13 University of California, Davis Daniel Zinn 13

University of California, Davis Daniel Zinn 14 Workflow Setup Start actors Connect all ports Unleash actors

University of California, Davis Daniel Zinn 15 Workflow-Script (sneak preview) Simple DSL for defining workflows Create Composite Actors Specify partial parameterizations Connect ports easily Check sanity (all ports connected, are types ok,…) Give deployment directives (hostname, co-locations) Run workflow

University of California, Davis Daniel Zinn 16 Workflow-Script Example

University of California, Davis Daniel Zinn 17 Outline Motivation & Related Work Parallel PN Engine Kepler PPN Director Demo Future Directions & Conclusion

University of California, Davis Daniel Zinn 18 Kepler PPN Director Idea: Use Kepler as sophisticated GUI Create, run and monitor LPPN workflows Marrying LPPN and Kepler – The PPN Director Drag’n’drop workflow creation (1:1 mapping for actors) Parameter support Hints for deployment from user Monitor token sending and receiving Monitor actor status …

University of California, Davis Daniel Zinn 19 PPN Director – Architecture Overview Kepler LPPN Local Machine

University of California, Davis Daniel Zinn 20 PPN Director – Design Decisions Proxy-Actors in Kepler represent Actors in LPPN Repository of available LPPN Actors in XML file (next slide) Actor-name Parameters and default values Ports Generic PPN-Actor is configured using this information Monitor actor state Send data from Kepler Actors to LPPN actors and vice versa PPN Director Start Actors with parameters, deployment info Connect Actors according to Kepler workflow Unleash and stop workflow execution

University of California, Davis Daniel Zinn 21 LPPN Actor Repository

University of California, Davis Daniel Zinn 22 Monitoring Support PPN Actors periodically probe LPPN actors for info Number of tokens sent and received Current actor state: Working Block on receive Block on write Sending BLOB tokens Displayed on actor while workflow is running …

University of California, Davis Daniel Zinn 23 Communication with Regular PN Actors? Sending data from regular Kepler Actors to LPPN and vice versa

University of California, Davis Daniel Zinn 24 Communication with Regular PN Actors

University of California, Davis Daniel Zinn 25 Communication with Regular PN Actors! Sending data from regular Kepler Actors to LPPN and vice versa

University of California, Davis Daniel Zinn 26 Outline Motivation & Related Work Lightweight Parallel PN Engine Kepler PPN Director Demo Future Directions & Conclusion

University of California, Davis Daniel Zinn 27 University of California, Davis Daniel Zinn 27

University of California, Davis Daniel Zinn 28 University of California, Davis Daniel Zinn 28

University of California, Davis Daniel Zinn 29 University of California, Davis Daniel Zinn 29

University of California, Davis Daniel Zinn 30 Outline Motivation & Related Work Lightweight Parallel PN Engine Kepler PPN Director Demo Future Directions & Conclusion

University of California, Davis Daniel Zinn 31 Future Directions Adding Black-box (Java) actors as actors in LPPN Detailed measurements when actors need time for what Automatic movement of actors for CPU congestions (deploying spring/mass model) Automatic data parallelism (actor cloning and scatter+gather) Overhaul of LPPN, maybe in Java, RMI, JNI Better resource management

University of California, Davis Daniel Zinn 32 Conclusions LPPN – a simple, fast and extensible PN engine Kepler successfully used as front-end for LPPN Kepler for staging & monitoring Interoperability between Kepler and LPPN

University of California, Davis Daniel Zinn 33 University of California, Davis Daniel Zinn 33