European Research Network on Foundations, Software Infrastructures and Applications for large scale distributed, GRID and Peer-to-Peer Technologies Scalability.

Slides:



Advertisements
Similar presentations
All rights reserved © 2006, Alcatel Grid Standardization & ETSI (May 2006) B. Berde, Alcatel R & I.
Advertisements

European Research Network on Foundations, Software Infrastructures and Applications for large scale distributed, GRID and Peer-to-Peer Technologies Grid.
European Research Network on Foundations, Software Infrastructures and Applications for large scale distributed, GRID and Peer-to-Peer Technologies Experiences.
Distributed Data Processing
Vrije Universiteit Interdroid: a platform for distributed smartphone applications Henri Bal, Nick Palmer, Roelof Kemp, Thilo Kielmann High Performance.
7 april SP3.1: High-Performance Distributed Computing The KOALA grid scheduler and the Ibis Java-centric grid middleware Dick Epema Catalin Dumitrescu,
Database Architectures and the Web
Distributed Systems 1 Topics  What is a Distributed System?  Why Distributed Systems?  Examples of Distributed Systems  Distributed System Requirements.
Copyright 2009 FUJITSU TECHNOLOGY SOLUTIONS PRIMERGY Servers and Windows Server® 2008 R2 Benefit from an efficient, high performance and flexible platform.
Web-based Distributed Flexible Manufacturing System (FMS) Monitoring and Control Student: Wei Liu Instructor: Dr. Chang Apr. 23, 2003.
Lesson 11-Virtual Private Networks. Overview Define Virtual Private Networks (VPNs). Deploy User VPNs. Deploy Site VPNs. Understand standard VPN techniques.
OCT1 Principles From Chapter One of “Distributed Systems Concepts and Design”
Overview Distributed vs. decentralized Why distributed databases
All Organizations Need to Share and Communicate Information...
An Introduction to Cloud Computing. The challenge Add new services for your users quickly and cost effectively.
Open Cloud Sunil Kumar Balaganchi Thammaiah Internet and Web Systems 2, Spring 2012 Department of Computer Science University of Massachusetts Lowell.
VAP What is a Virtual Application ? A virtual application is an application that has been optimized to run on virtual infrastructure. The application software.
A Brief Overview by Aditya Dutt March 18 th ’ Aditya Inc.
Opensource for Cloud Deployments – Risk – Reward – Reality
Barracuda Load Balancer Server Availability and Scalability.
Cloud computing is the use of computing resources (hardware and software) that are delivered as a service over the Internet. Cloud is the metaphor for.
Hands-On Microsoft Windows Server 2008 Chapter 1 Introduction to Windows Server 2008.
©Ian Sommerville 2006Software Engineering, 8th edition. Chapter 12 Slide 1 Distributed Systems Architectures.
Version 4.0. Objectives Describe how networks impact our daily lives. Describe the role of data networking in the human network. Identify the key components.
Cloud Computing 1. Outline  Introduction  Evolution  Cloud architecture  Map reduce operation  Platform 2.
DISTRIBUTED COMPUTING
1 CS 501 Spring 2003 CS 501: Software Engineering Lecture 16 System Architecture and Design II.
Presented by Xiaoyu Qin Virtualized Access Control & Firewall Virtualization.
GT Components. Globus Toolkit A “toolkit” of services and packages for creating the basic grid computing infrastructure Higher level tools added to this.
INFSO-RI Enabling Grids for E-sciencE SA1: Cookbook (DSA1.7) Ian Bird CERN 18 January 2006.
1 Introduction to Middleware. 2 Outline What is middleware? Purpose and origin Why use it? What Middleware does? Technical details Middleware services.
7-1 Management Information Systems for the Information Age Copyright 2004 The McGraw-Hill Companies, Inc. All rights reserved Chapter 7 IT Infrastructures.
Advanced Computer Networks Topic 2: Characterization of Distributed Systems.
Developing & Managing A Large Linux Farm – The Brookhaven Experience CHEP2004 – Interlaken September 27, 2004 Tomasz Wlodek - BNL.
出處 :2010 2nd International Conference on Signal Processing Systems (ICSPS) 作者 :Zhidong Shen 、 Qiang Tong 演講者 : 碩研資管一甲 吳俊逸.
SONIC-3: Creating Large Scale Installations & Deployments Andrew S. Neumann Principal Engineer, Progress Sonic.
Grid Middleware Tutorial / Grid Technologies IntroSlide 1 /14 Grid Technologies Intro Ivan Degtyarenko ivan.degtyarenko dog csc dot fi CSC – The Finnish.
Chapter 5 McGraw-Hill/Irwin Copyright © 2011 by The McGraw-Hill Companies, Inc. All rights reserved.
NA-MIC National Alliance for Medical Image Computing UCSD: Engineering Core 2 Portal and Grid Infrastructure.
SONIC-3: Creating Large Scale Installations & Deployments Andrew S. Neumann Principal Engineer Progress Sonic.
Company small business cloud solution Client UNIVERSITY OF BEDFORDSHIRE.
Globus and PlanetLab Resource Management Solutions Compared M. Ripeanu, M. Bowman, J. Chase, I. Foster, M. Milenkovic Presented by Dionysis Logothetis.
GRID ANATOMY Advanced Computing Concepts – Dr. Emmanuel Pilli.
3/12/2013Computer Engg, IIT(BHU)1 CLOUD COMPUTING-1.
1 Active Directory Service in Windows 2000 Li Yang SID: November 2000.
David Foster LCG Project 12-March-02 Fabric Automation The Challenge of LHC Scale Fabrics LHC Computing Grid Workshop David Foster 12 th March 2002.
Background Computer System Architectures Computer System Software.
Towards ‘Ubiquitous’ Ubiquitous Computing: an alliance with ‘the Grid’ Oliver Storz, Adrian Friday, and Nigel Davies Computing Department, Lancaster University,
INFSO-RI Enabling Grids for E-sciencE Grid & Cloud Computing Introduction
Enterprise Security Management Franklin Tinsley COSC 481.
A Semi-Automated Digital Preservation System based on Semantic Web Services Jane Hunter Sharmin Choudhury DSTC PTY LTD, Brisbane, Australia Slides by Ananta.
XtreemOS IP project is funded by the European Commission under contract IST-FP Scientific coordinator Christine Morin, INRIA Presented by Ana.
Clouding with Microsoft Azure
Prof. Jong-Moon Chung’s Lecture Notes at Yonsei University
Chapter 1 Characterization of Distributed Systems
DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING CLOUD COMPUTING
Netscape Application Server
An Introduction to Cloud Computing
Principles of Network Applications
CHAPTER 1 INTRODUCTION:
GRID COMPUTING PRESENTED BY : Richa Chaudhary.
Introduction to Cloud Computing
Enterprise Service Bus (ESB) (Chapter 9)
Time Gathering Systems Secure Data Collection for IBM System i Server
CSE 451: Operating Systems Spring 2005 Module 20 Distributed Systems
CSE 451: Operating Systems Winter 2004 Module 19 Distributed Systems
Introduction To Distributed Systems
STATEL an easy way to transfer data
Software Architecture Taxonomy
Presentation transcript:

European Research Network on Foundations, Software Infrastructures and Applications for large scale distributed, GRID and Peer-to-Peer Technologies Scalability in Grids Thilo Kielmann Vrije Universiteit, Amsterdam

European Research Network on Foundations, Software Infrastructures and Applications for large scale distributed, GRID and Peer-to-Peer Technologies Scalability...is a desirable property of a system, a network or a process, which indicates its ability to either handle growing amounts of work in a graceful manner, or to be readily enlarged.

European Research Network on Foundations, Software Infrastructures and Applications for large scale distributed, GRID and Peer-to-Peer Technologies Grids Integrating globally dispersed resources that are not subject to centralized control, to deliver non-trivial qualities of service.

European Research Network on Foundations, Software Infrastructures and Applications for large scale distributed, GRID and Peer-to-Peer Technologies Scalability in Grids Network delays (due to physical distance) bandwidth goes up, but speed of light remains the barrier Number of resources involved –1000s of CPUs working together –But only O(10) machines/clusters at a time Number of data integrated / processed – Frillions of bytes in remote files/DBs

European Research Network on Foundations, Software Infrastructures and Applications for large scale distributed, GRID and Peer-to-Peer Technologies Scalability in Grids (2) The real issues: The application that has worked yesterday won't work today. –Network disconnections (a.k.a. firewalls) –Service non-interoperability (a.k.a. versioning) –And even some hardware failures Scalability problem: O(10) to O(100) independent administrative authorities...

European Research Network on Foundations, Software Infrastructures and Applications for large scale distributed, GRID and Peer-to-Peer Technologies Scalability: challenges (1) From the GridLab testbed: Delphoi (Web) service provides network monitoring data

European Research Network on Foundations, Software Infrastructures and Applications for large scale distributed, GRID and Peer-to-Peer Technologies Scalability: challenges (1) Monitoring networks requires O(N²) measurements –This becomes practically infeasible, even with O(10) sites. Challenge: build systems that can work without ubiquitous monitoring information

European Research Network on Foundations, Software Infrastructures and Applications for large scale distributed, GRID and Peer-to-Peer Technologies Scalability: challenges (2) Service non-interoperability scaling to O(10) middleware/service platforms GAT:

European Research Network on Foundations, Software Infrastructures and Applications for large scale distributed, GRID and Peer-to-Peer Technologies Scalability: challenges (3) - Grid plugtests - Work building systems for O(1000) machines

European Research Network on Foundations, Software Infrastructures and Applications for large scale distributed, GRID and Peer-to-Peer Technologies Scalability: challenges (3) Changes in our Ibis system due to scalability problems: –Spread all-to-all connection setup over the runtime (false positive denial-of-service, TCP socket limits) –Optimize central registry (multi threading, message combining) Not many grid software is built for large-scale use.

European Research Network on Foundations, Software Infrastructures and Applications for large scale distributed, GRID and Peer-to-Peer Technologies Scalability: challenges (4) What can go wrong, will go wronk. (Murphy) Due to the large amount of components (hardware, networks, middleware, applications...) there will always be something that is not working (as expected) The real challenge is to build systems that are –self-* –autonomic –simply working in a reliable manner...

European Research Network on Foundations, Software Infrastructures and Applications for large scale distributed, GRID and Peer-to-Peer Technologies Scalability: challenges (5) single/simple Web services and servers provide critical points of failures XtreemOS is building virtual servers with hand-over based on peer-to-peer technology

European Research Network on Foundations, Software Infrastructures and Applications for large scale distributed, GRID and Peer-to-Peer Technologies Scalability: challenges (6) Security, AAA –Authentication and authorization for users and services –Should be a trivial (straight forward) problem –... if it would not be for scaling to MANY users Manual granting of credentials, non-technical but human (or legal) issues Challenge: automated security mechanisms that are powerful, flexible, and trustworthy

European Research Network on Foundations, Software Infrastructures and Applications for large scale distributed, GRID and Peer-to-Peer Technologies Conclusions Many scalability problems can be circumvented by doing our homework (write solid, well-designed software) The real challenge is to build systems that are –self-* –autonomic –simply working in a reliable manner... as this is the key to addressing the scalability issues!