Master Worker Paradigm Support in Software Component Models Hinde Bouziane, Christian Pérez PARIS Research Team INRIA/IRISA Rennes ANR CIGC LEGO (ANR-05-CICG-11)

Slides:



Advertisements
Similar presentations
Message Passing Vs Distributed Objects
Advertisements

Elton Mathias and Jean Michael Legait 1 Elton Mathias, Jean Michael Legait, Denis Caromel, et al. OASIS Team INRIA -- CNRS - I3S -- Univ. of Nice Sophia-Antipolis,
Architecture Representation
RPC Robert Grimm New York University Remote Procedure Calls.
Software Frame Simulator (SFS) Technion CS Computer Communications Lab (236340) in cooperation with ECI telecom Uri Ferri & Ynon Cohen January 2007.
GridRPC Sources / Credits: IRISA/IFSIC IRISA/INRIA Thierry Priol et. al papers.
 Copyright 2005 Digital Enterprise Research Institute. All rights reserved. Workflow utilization in composition of complex applications based.
XML Technologies and Applications Rajshekhar Sunderraman Department of Computer Science Georgia State University Atlanta, GA 30302
A brief look at CORBA. What is CORBA Common Object Request Broker Architecture developed by OMG Combine benefits of OO and distributed computing Distributed.
Persistent State Service 1 CORBA Component  Component model  Container programming model  Component implementation framework  Component packaging and.
Distributed Object Computing Weilie Yi Dec 4, 2001.
1 Quality Objects: Advanced Middleware for Wide Area Distributed Applications Rick Schantz Quality Objects: Advanced Middleware for Large Scale Wide Area.
CS490T Advanced Tablet Platform Applications Network Programming Evolution.
CSE 636 Data Integration Web Services.
Communication in Distributed Systems –Part 2
Software Engineering Module 1 -Components Teaching unit 3 – Advanced development Ernesto Damiani Free University of Bozen - Bolzano Lesson 2 – Components.
Page 1 Building Reliable Component-based Systems Chapter 4 - Component Models and Technology Chapter 4 Component Models and Technology.
Software Issues Derived from Dr. Fawcett’s Slides Phil Pratt-Szeliga Fall 2009.
.NET Mobile Application Development Remote Procedure Call.
©Ian Sommerville 2004Software Engineering, 7th edition. Chapter 18 Slide 1 Software Reuse.
Grid Computing, B. Wilkinson, 20043b.1 Web Services Part II.
Understanding the CORBA Model. What is CORBA?  The Common Object Request Broker Architecture (CORBA) allows distributed applications to interoperate.
INRIA Sophia-Antipolis, Oasis team INRIA Rhône-Alpes, Vasy team Feria–IRIT/LAAS, SVF team Toulouse GET - ENST Paris, LTCI team FIACRE Models and Tools.
Provisioning/Deployment CDDLM-WG (Configuration Description, Deployment and Lifecycle Management) December 4, 2003 Takashi Kojo, NEC.
1 CCA Meeting, Januray 25th 2007 Supporting the Master-Worker Paradigm in the Common Component Architecture Hinde Lilia Bouziane, Christian Pérez, Thierry.
The Grid Component Model: an Overview “Proposal for a Grid Component Model” DPM02 “Basic Features of the Grid Component Model (assessed)” -- DPM04 CoreGrid.
Hinde Bouziane – CBHPC’08 – October 2008 Marco ALDINUCCI and Marco DANELUTTO UNIPI - University of Pisa (Italy) Hinde Lilia BOUZIANE and Christian.
The Grid Component Model and its Implementation in ProActive CoreGrid Network of Excellence, Institute on Programming Models D.PM02 “Proposal for a Grid.
11 September 2008CIS 340 # 1 Topics To examine the variety of approaches to handle the middle- interaction (continued) 1.RPC-based systems 2.TP monitors.
© 2008 Open Grid Forum Independent Software Vendor (ISV) Remote Computing Primer Steven Newhouse.
RELATIONAL FAULT TOLERANT INTERFACE TO HETEROGENEOUS DISTRIBUTED DATABASES Prof. Osama Abulnaja Afraa Khalifah
1 MAIN TABLE OF CONTENTS Definition: SOFTWARE AGENT HOW MANY TYPES OF AGENT? DEFINITION OF MOBILE AGENT: SOFTWARE AGENTS PROPERTIES, WORKING OF MOBILE.
Programming Parallel and Distributed Systems for Large Scale Numerical Simulation Application Christian Perez INRIA researcher IRISA Rennes, France.
XML Web Services Architecture Siddharth Ruchandani CS 6362 – SW Architecture & Design Summer /11/05.
Shannon Hastings Multiscale Computing Laboratory Department of Biomedical Informatics.
OpenCCM MdC Philippe Merle LIFL - INRIA (soon)
Distributed Programming CSCI 201L Jeffrey Miller, Ph.D. HTTP :// WWW - SCF. USC. EDU /~ CSCI 201 USC CSCI 201L.
WebService. Outline Overview of Web Services SOAP (messaging) WSDL (service description) UDDI (registry)
ServiceSs, a new programming model for the Cloud Daniele Lezzi, Rosa M. Badia, Jorge Ejarque, Raul Sirvent, Enric Tejedor Grid Computing and Clusters Group.
07/09/04 Johan Muskens ( TU/e Computer Science, System Architecture and Networking.
Grid programming with components: an advanced COMPonent platform for an effective invisible grid © 2006 GridCOMP Grids Programming with components. An.
Technology Layer. Technology Layer Metamodel Technology Layer Concepts.
The EDGeS project receives Community research funding 1 Porting Applications to the EDGeS Infrastructure A comparison of the available methods, APIs, and.
Kemal Baykal Rasim Ismayilov
Distributed Components for Integrating Large- Scale High Performance Computing Applications Nanbor Wang, Roopa Pundaleeka and Johan Carlsson
ProActive components and legacy code Matthieu MOREL.
Integrated application of eODL 1 J.Fischer Integrated application of eODL Workshop Integrated application of eODL Geneva, 13 September 2003 Harald Böhme,
STCM: A spatio-temporal Component Model Hinde Lilia Bouziane Christian Perez PARIS Project-Team INRIA Rennes/IRISA Toulouse, France, December 7 th 2007.
ANR CIGC LEGO (ANR-CICG-05-11) Bordeaux, 2006, December 11 th Automatic Application Deployment on Grids Landry Breuil, Boris Daix, Sébastien Lacour, Christian.
1 ProActive GCM – CCA Interoperability Maciej Malawski, Ludovic Henrio, Matthieu Morel, Francoise Baude, Denis Caromel, Marian Bubak Institute of Computer.
Advanced Component Models ULCM & HLCM Julien Bigot, Hinde Bouziane, Christian Perez COOP Project Lyon, 9-10 mars 2010.
August 20, 2002 Applying RT-Policies in CORBA Component Model Nanbor Wang Department of Computer Science Washington University in St. Louis
1 Use of SDD in Grid Deployment Based on GGF CDDLM Jun Tatemura CDDLM WG member NEC Laboratories America Sept 13, 2005.
GridRPC Sources / Credits: IRISA/IFSIC IRISA/INRIA Thierry Priol et. al papers.
Toward a Distributed and Parallel High Performance Computing Environment Johan Carlsson and Nanbor Wang Tech-X Corporation Boulder,
CCA Distributed Framework Interoperability. Goals Assume you have two (or more) framework instances. –Assume it contains a network of component instances.
Towards a High Performance Extensible Grid Architecture Klaus Krauter Muthucumaru Maheswaran {krauter,
GridBench: A Tool for Benchmarking Grids
WEB SERVICES.
Web Ontology Language for Service (OWL-S)
Web services, WSDL, SOAP and UDDI
CSE 451: Operating Systems Autumn 2003 Lecture 16 RPC
CSSSPEC6 SOFTWARE DEVELOPMENT WITH QUALITY ASSURANCE
Tools for Composing and Deploying Grid Middleware Web Services
Component--based development
UML profiles.
CSE 451: Operating Systems Winter 2003 Lecture 16 RPC
Distributed System using Web Services
The Grid Component Model and its Implementation in ProActive
CSE 451: Operating Systems Messaging and Remote Procedure Call (RPC)
Presentation transcript:

Master Worker Paradigm Support in Software Component Models Hinde Bouziane, Christian Pérez PARIS Research Team INRIA/IRISA Rennes ANR CIGC LEGO (ANR-05-CICG-11) Bordeaux, 2006, December 11 th

Master-worker applications Understanding very high energy cosmic rays AUGER Project (LRI/LAL-IN2P3) XtremWeb Neuron Simulation with MCELL CRPC Rice University NetSolve

Problem overview  Simultaneous independent computations (~ForAll loop)  Dedicated API/environments  BOINC, XTremWEB, DIET, NetSolve, Nimrod/G,... worker Master worker Workers collection Requests transport Scheduling Fault tolerance

Limits with current component models  Different infrastructures  Multi-core processors, SMP, clusters, grids, etc.  Resources dependant properties  Number of workers  Request transport and scheduling policy  At the burden of the programmer  Complex  No transparence  Objectives  Transparency  Re-use existing MW environments W W worker master request delivery policy W W worker

Features of the model  Resources infrastructure independence  Transparency  No dealing with the number of workers  No dealing with request delivery concerns  An initial number of workers depending on the current resources infrastructure  Introduction of a request delivery policy depending on the current resources infrastructure

Abstract master and worker composition (1/2): a collection  Collection definition Exposed provided port worker W1 Wn W2 Wi Type1Type3 Type2 Type1 T1x Type2 T2y Type3 T3z  Collection at execution Instantiation

Abstract master and worker composition (2/2): abstract assembly Composing components with collections ≈ Component composition m Type1Type3 Type2 X

Integration of a request delivery policy (1/2): transformation  Abstract ADL ->concrete ADL Round-Robin Deployment environment Transformation process Round-Robin

Integration of a request delivery policy (2/2): patterns w w w M Round-Robin / Random LA M w w w MA Simple component based pattern Hierarchical scheduling pattern DIET pattern M Random w w Round-Robin w w

Sum up deployment environment Transformation Round-Robin Programmer/designer view Components Collections Abstract assembly During deployment phase #workers + Pattern selection Set of patterns

Extending the CORBA Component Model

Case of study: CCM  Master and worker component definition  Interface Description Language (IDL3)  Collection definition in two steps  External view: IDL3 extension  Internal content and bindings: Collection Description Language (CDL) in XML format interface Compute {..} ; component master { uses Compute m_port; }; component worker { provides Compute w_port; }; collection coll { provides Compute c_port; }; <internPort elemType=“worker“ port=“w_port"/> c_port worker masterworker w_port m_port

Experimenting MW in CCM  Synthetic MW application  Benchmark  Several flavor of the master  Sequential or parallel operation invocations  Latency & bandwidth oriented request  Regular or irregular request bench  Work generated by one request  Several transport request policy component  None (direct connection from master to workers)  Round-robin proxy  Random proxy  DIET

MW CCM & DIET  Client & Server-side DIET Component  Wrapper  Conversion between IDL & DIET data representation  Home-made CCM compiler  DIET support  Generate adequate code for client & server side components  Control with special comments in IDL3 LA Client Component Server component w w MA M w w

Discussion  MW support for Component Model  Map to CCM, CCA & Fractal  Easy to port application  All the work is to componentize the applications  Two applications have been ported  Experiments are in progress  What to measure?  Which API?  Data dypes  Bunch of calls (~ForAll loops?)  DIET deployed as CCM executable  CSD supports executable  DIET hierarchy described in CAD files!  Current ADAGE planner is not aware of DIET specificity  Deployment control through associate element of the ADAGE control parameters