A Web Based Job Submission System for a Physics Computing Cluster David Jones IOP Particle Physics 2004 Birmingham 1.

Slides:



Advertisements
Similar presentations
Building Portals to access Grid Middleware National Technical University of Athens Konstantinos Dolkas, On behalf of Andreas Menychtas.
Advertisements

University of Southampton Electronics and Computer Science M-grid: Using Ubiquitous Web Technologies to create a Computational Grid Robert John Walters.
S. Gadomski, "ATLAS computing in Geneva", journee de reflexion, 14 Sept ATLAS computing in Geneva Szymon Gadomski description of the hardware the.
Caching the MDSPlus Data via Hibernate By Ajith M Jose Comp6703 Project Client: Raju Karia Supervisor: Dr. Henry Gardner (Development of “WebScope”)
A CHAT CLIENT-SERVER MODULE IN JAVA BY MAHTAB M HUSSAIN MAYANK MOHAN ISE 582 FALL 2003 PROJECT.
Programming project #4 1 CS502 Spring 2006 Programming Project #4 Web Server CS-502 Operating Systems Spring 2006.
INTERNET DATABASE. Internet and E-commerce Internet – a worldwide collection of interconnected computer network Internet – a worldwide collection of interconnected.
Multiple Tiers in Action
CS CS 5150 Software Engineering Lecture 13 System Architecture and Design 1.
K.Harrison CERN, 23rd October 2002 HOW TO COMMISSION A NEW CENTRE FOR LHCb PRODUCTION - Overview of LHCb distributed production system - Configuration.
CS-3103 & CS-502, Summer 2006 Programming Project #31 Programming Project #3 Web Server CS-3103 & CS-502 Operating Systems.
High Performance Computing (HPC) at Center for Information Communication and Technology in UTM.
Introduction to client/server architecture
1 Chapter Overview Introduction to Windows XP Professional Printing Setting Up Network Printers Connecting to Network Printers Configuring Network Printers.
Bonrix SMPP Gateway Index Introduction Architecture diagram Set up diagram System & Software Requirements Installation Deployment Operations HTTP.
Data Grid Web Services Chip Watson Jie Chen, Ying Chen, Bryan Hess, Walt Akers.
Web server and web browser It’s a take and give policy in between client and server through HTTP(Hyper Text Transport Protocol) Server takes a request.
Sys Prog & Scripting - HW Univ1 Systems Programming & Scripting Lecture 15: PHP Introduction.
CSCI 6962: Server-side Design and Programming Course Introduction and Overview.
Track 1: Cluster and Grid Computing NBCR Summer Institute Session 2.2: Cluster and Grid Computing: Case studies Condor introduction August 9, 2006 Nadya.
Java Omar Rana University of South Asia. Course Overview JAVA  C/C++ and JAVA Comparison  OOP in JAVA  Exception Handling  Streams  Graphics User.
Test Of Distributed Data Quality Monitoring Of CMS Tracker Dataset H->ZZ->2e2mu with PileUp - 10,000 events ( ~ 50,000 hits for events) The monitoring.
Cluster currently consists of: 1 Dell PowerEdge Ghz Dual, quad core Xeons (8 cores) and 16G of RAM Original GRIDVM - SL4 VM-Ware host 1 Dell PowerEdge.
Chapter 17 - Deploying Java Applications on the Web1 Chapter 17 Deploying Java Applications on the Web.
A Distributed Computing System Based on BOINC September - CHEP 2004 Pedro Andrade António Amorim Jaime Villate.
ISG We build general capability Introduction to Olympus Shawn T. Brown, PhD ISG MISSION 2.0 Lead Director of Public Health Applications Pittsburgh Supercomputing.
The SAMGrid Data Handling System Outline:  What Is SAMGrid?  Use Cases for SAMGrid in Run II Experiments  Current Operational Load  Stress Testing.
Daniel Vanderster University of Victoria National Research Council and the University of Victoria 1 GridX1 Services Project A. Agarwal, A. Berman, A. Charbonneau,
LHCb Applications and GRID Integration Domenico Galli Catania, April 9, st INFN-GRID Workshop.
Introduction to J2EE Architecture Portions by Kunal Mehta.
Introduction to Apache Hadoop Zibo Wang. Introduction  What is Apache Hadoop?  Apache Hadoop is a software framework which provides open source libraries.
3rd Nov 2000HEPiX/HEPNT CDF-UK MINI-GRID Ian McArthur Oxford University, Physics Department
3rd June 2004 CDF Grid SAM:Metadata and Middleware Components Mòrag Burgon-Lyon University of Glasgow.
COMP3019 Coursework: Introduction to GridSAM Steve Crouch School of Electronics and Computer Science.
Unit – I CLIENT / SERVER ARCHITECTURE. Unit Structure  Evolution of Client/Server Architecture  Client/Server Model  Characteristics of Client/Server.
The PROGRESS Grid Service Provider Maciej Bogdański Portals & Portlets 2003 Edinburgh, July 14th-17th.
Evaluation of Agent Teamwork High Performance Distributed Computing Middleware. Solomon Lane Agent Teamwork Research Assistant October 2006 – March 2007.
1 Welcome to CSC 301 Web Programming Charles Frank.
By: Pramod Jagtap Aniket Bochare. Agenda Introduction to dataset Web service description Service architecture Project plan Intended clients.
Dr. Azeddine Chikh IS444: Modern tools for applications development.
Resource Brokering in the PROGRESS Project Juliusz Pukacki Grid Resource Management Workshop, October 2003.
Chapter 10 Intro to SOAP and WSDL. Objectives By study in the chapter, you will be able to: Describe what is SOAP Exam the rules for creating a SOAP document.
09/02 ID099-1 September 9, 2002Grid Technology Panel Patrick Dreher Technical Panel Discussion: Progress in Developing a Web Services Data Analysis Grid.
1 Web Servers (Chapter 21 – Pages( ) Outline 21.1 Introduction 21.2 HTTP Request Types 21.3 System Architecture.
CSI 3125, Preliminaries, page 1 SERVLET. CSI 3125, Preliminaries, page 2 SERVLET A servlet is a server-side software program, written in Java code, that.
Using RSNA’s Teaching File Software (MIRC): A Hands on Course Mary Wyers, MD.
Timeshared Parallel Machines Need resource management Need resource management Shrink and expand individual jobs to available sets of processors Shrink.
ISG We build general capability Introduction to Olympus Shawn T. Brown, PhD ISG MISSION 2.0 Lead Director of Public Health Applications Pittsburgh Supercomputing.
Tier3 monitoring. Initial issues. Danila Oleynik. Artem Petrosyan. JINR.
Computing Issues for the ATLAS SWT2. What is SWT2? SWT2 is the U.S. ATLAS Southwestern Tier 2 Consortium UTA is lead institution, along with University.
PROGRESS: GEW'2003 Using Resources of Multiple Grids with the Grid Service Provider Michał Kosiedowski.
CIP HPC CIP - HPC HPC = High Performance Computer It’s not a regular computer, it’s bigger, faster, more powerful, and more.
Accounting in DataGrid HLR software demo Andrea Guarise Milano, September 11, 2001.
INRNE's participation in LCG Elena Puncheva Preslav Konstantinov IT Department.
Large-scale accelerator simulations: Synergia on the Grid turn 1 turn 27 turn 19 turn 16 C++ Synergia Field solver (FFT, multigrid) Field solver (FFT,
Grid Execution Management for Legacy Code Architecture Exposing legacy applications as Grid services: the GEMLCA approach Centre.
Next Generation of Apache Hadoop MapReduce Owen
Simulation Production System Science Advisory Committee Meeting UW-Madison March 1 st -2 nd 2007 Juan Carlos Díaz Vélez.
E-commerce Architecture Ayşe Başar Bener. Client Server Architecture E-commerce is based on client/ server architecture –Client processes requesting service.
Claudio Grandi INFN Bologna Virtual Pools for Interactive Analysis and Software Development through an Integrated Cloud Environment Claudio Grandi (INFN.
The Echo Server Problem. Contents  Basic Networking Concepts  The Echo Server Problem.
Introduction to Distributed Platforms
Belle II Physics Analysis Center at TIFR
Introduction to J2EE Architecture
PHP / MySQL Introduction
Introduction to client/server architecture
The Internet of Things (IoT)
Patrick Dreher Research Scientist & Associate Director
Objectives In this lesson you will learn about: Need for servlets
Presentation transcript:

A Web Based Job Submission System for a Physics Computing Cluster David Jones IOP Particle Physics 2004 Birmingham 1

IOP Particle Physics 2004 Birmingham David Jones Introduction & Motivation MAP (Monte-Carlo Array Processor) Simplify the process of submitting a computing Job to MAP Provide more feedback to the user Provide insight into how to turn MAP into a Grid resource

IOP Particle Physics 2004 Birmingham David Jones What is MAP? Monte Carlo Array Processor ‘Commodity’ Personal computers mounted in rack – –MAP nodes –MAP nodes Wired together with a private network Work in parallel running same simulation with different parameters Good for Physics and Engineering problem solving One of the World’s most powerful computers Concept developed & refined at Liverpool Software written and maintained in house; More to it than just this slide

IOP Particle Physics 2004 Birmingham David Jones MAP Philosophy A scientist sets up a job on their machine Once the job is working –Job files and libraries are tarred into one file –A second file is used;- Don’t want 1000 identical jobs running Introduces differences between nodes By, for example, random number initialisation A good alternative to batch systems

IOP Particle Physics 2004 Birmingham David Jones Pictures MAP and MAP2 Lower left: a bank of MAP 2 nodes. Upper left: the innards of a MAP2 node Right: MAP1 with Private network wiring visible MAP 2 Each Node: CPU 3000 GHz Xeon Pentium 4 1GB RAM 100TB of direct attached storage Slide mounted Nodes

IOP Particle Physics 2004 Birmingham David Jones A Web Interface Existing Job Submission Tool is MAPXQS –An X-Windows based Program DUCK (Devolved User Control Konsole) Submit a MAP job a web browser My contribution to the project Forms middle part of three tier system –Client - User’s Web browser –DUCK - Middleware (A Web application) –MAP - The computing resource Web browser MAP System Query daemon Queue daemon DUCK Client Web container 3 rd Tier Middleware

IOP Particle Physics 2004 Birmingham David Jones User Interfaces Contrast between the X-Windows MAP interface (top right) & the Web interface (Left)

IOP Particle Physics 2004 Birmingham David Jones Milestones so far Successful submission of jobs using DUCK (Atlas Monte-Carlo Production) –200 Million events for MSSM SUSY studies –500 Million events for Higgs studies –Generated events used for Physics studies –E.g. Search for a light Higgs in ttH channel Successfully deployed DUCK on Linux server

IOP Particle Physics 2004 Birmingham David Jones Queue Daemon Daemon – a program that runs constantly New web queue daemon developed Acts as the communication between DUCK and MAP Socket-based communication of Job parameters;- MAP setup parameters Application-specific parameters

IOP Particle Physics 2004 Birmingham David Jones Application Parameters Parameters that describe the application i.e. –Physics Parameters –Engineering Parameters Database system allows jobs with an arbitrary number of application parameters to be specified Name, description, default value/ value triplet Values are typically random number seeds etc.

IOP Particle Physics 2004 Birmingham David Jones Query Daemon Allows users to see what jobs are waiting to run on MAP over the web Information updated with periodicity of minutes from the MAP Query daemon Avoids traffic saturation of the MAP software

IOP Particle Physics 2004 Birmingham David Jones Inside DUCK - Java Core MAP software written in C for performance reasons DUCK written in Java Takes advantage of Java’s –Database bindings –Networking packages –Web delivery mechanism (servlets) Servlet engine Tomcat (Apache Project product)

IOP Particle Physics 2004 Birmingham David Jones Inside DUCK - Servlets DUCK is made up of multiple Servlets Servlet are –small programs –that produce web pages as output –as response to a request from a web browser Client (web browser) doesn’t store data between pages –MySQL database provides a persistency mechanism

IOP Particle Physics 2004 Birmingham David Jones Inside DUCK – MAP Interface Connection between DUCK and MAP is a simple text protocol over a Network Socket Currently two listeners –Queue –Query Queue data is sent in the form –[variablename]variable Messages sent back at the end of communication in 3 groups –Errors – things that stop a MAP job running –Warnings – things that are possibly a problem –Infos – Information about the job JCB [variablename]variable

IOP Particle Physics 2004 Birmingham David Jones Concluding DUCK is; –Middleware for MAP –deployed on Linux (developed on a win32 platform) –In action; Users are submitting jobs using the interface (Atlas) There are features to be added to the interface if time permits Can already see ways to ‘Grid-ify’ the middleware