1 Workshop 20: Teaching a Hands-on Undergraduate Grid Computing Course SIGCSE 2010 - The 41st ACM Technical Symposium on Computer Science Education Friday.

Slides:



Advertisements
Similar presentations
Barry Wilkinson Department of Computer Science University of North Carolina at Charlotte Mark Holliday Department of Mathematics and Computer Science Western.
Advertisements

Grid Computing, Barry Wilkinson, 2004A3.1 Assignment 3 Simple Job Submission Using GRAM.
Mark Holliday and Barry Wilkinson, 2004A2.1 Assignment 2 “Simple” Grid Services Assignment.
Barry Wilkinson and Mark Holliday, 2004A1.1 Assignments Preliminaries Several computers are configured at WCU for the assignments. Here, terra.cs.wcu.edu.
A3.1 Assignment 3 Simple Job Submission Using GT 4 GRAM.
1 Workshop 20: Teaching a Hands-on Undergraduate Grid Computing Course SIGCSE The 41st ACM Technical Symposium on Computer Science Education Friday.
Dealing with real resources Wednesday Afternoon, 3:00 pm Derek Weitzel OSG Campus Grids University of Nebraska.
Setting up of condor scheduler on computing cluster Raman Sehgal NPD-BARC.
Condor DAGMan Warren Smith. 12/11/2009 TeraGrid Science Gateways Telecon2 Basics Condor provides workflow support with DAGMan Directed Acyclic Graph –Each.
Intermediate Condor: DAGMan Monday, 1:15pm Alain Roy OSG Software Coordinator University of Wisconsin-Madison.
Dr. David Wallom Use of Condor in our Campus Grid and the University September 2004.
GRID workload management system and CMS fall production Massimo Sgaravatto INFN Padova.
7-2.1 Additional Features of WSRF/GT4 Services A brief outline © 2011 B. Wilkinson/Clayton Ferner. Fall 2011 Grid computing course. Modification date:
A2.1 ITCS 4010/5010 Grid Computing, 2005, UNC-Charlotte, B. Wilkinson Assignment 2 “Simple” Grid Services Assignment.
Tutorial on Distributed High Performance Computing 14:30 – 19:00 (2:30 pm – 7:00 pm) Wednesday November 17, 2010 Jornadas Chilenas de Computación 2010.
1 Teaching Grid Computing across North Carolina and Beyond Dr. Clayton Ferner University of North Carolina Wilmington Dr. Barry Wilkinson University of.
1 Workshop 20: Teaching a Hands-on Undergraduate Grid Computing Course SIGCSE The 41st ACM Technical Symposium on Computer Science Education Friday.
1 Using Condor An Introduction ICE 2008.
1 Workshop 20: Teaching a Hands-on Undergraduate Grid Computing Course SIGCSE The 41st ACM Technical Symposium on Computer Science Education Friday.
1 Short Course on Grid Computing Jornadas Chilenas de Computación 2010 INFONOR-CHILE 2010 November 15th - 19th, 2010 Antofagasta, Chile Dr. Barry Wilkinson.
1 Workshop 20: Teaching a Hands-on Undergraduate Grid Computing Course SIGCSE The 41st ACM Technical Symposium on Computer Science Education Friday.
1 Short Course on Grid Computing Jornadas Chilenas de Computación 2010 INFONOR-CHILE 2010 November 15th - 19th, 2010 Antofagasta, Chile Dr. Barry Wilkinson.
Outline.1 Grid Computing Spring 2010 Tuesday/Thursday 11:00 am - 12:15 pm Instructors © 2010 B. Wilkinson/Clayton Ferner. Modification date: Dec 20, 2009.
Intermediate HTCondor: Workflows Monday pm Greg Thain Center For High Throughput Computing University of Wisconsin-Madison.
Introduction to Condor DMD/DFS J.Knudstrup December 2005.
Utilizing Condor and HTC to address archiving online courses at Clemson on a weekly basis Sam Hoover 1 Project Blackbird Computing,
ITCS 6/8010 CUDA Programming, UNC-Charlotte, B. Wilkinson, Jan 22, 2011assignprelim.1 Assignment Preliminaries ITCS 6010/8010 Spring 2011.
CONDOR DAGMan and Pegasus Selim Kalayci Florida International University 07/28/2009 Note: Slides are compiled from various TeraGrid Documentations.
Grid Computing, B. Wilkinson, 20046d.1 Schedulers and Resource Brokers.
1 History and New Developments in State-wide Grid Computing Course NC A & T University Saturday Oct 18, 2008 Dr. Barry Wilkinson University of North Carolina.
ITCS 4/5145 Parallel Programming, UNC-Charlotte, B. Wilkinson, 2012, Jan 18, 2012assignprelim.1 Assignment Preliminaries ITCS 4145/5145 Spring 2012.
High Throughput Computing with Condor at Purdue XSEDE ECSS Monthly Symposium Condor.
April Open Science Grid Campus Condor Pools Mats Rynge – Renaissance Computing Institute University of North Carolina, Chapel Hill.
IntroductiontotoHTCHTC 2015 OSG User School, Monday, Lecture1 Greg Thain University of Wisconsin– Madison Center For High Throughput Computing.
An Introduction to High-Throughput Computing Monday morning, 9:15am Alain Roy OSG Software Coordinator University of Wisconsin-Madison.
The Glidein Service Gideon Juve What are glideins? A technique for creating temporary, user- controlled Condor pools using resources from.
Condor Tugba Taskaya-Temizel 6 March What is Condor Technology? Condor is a high-throughput distributed batch computing system that provides facilities.
1 SIGCSE 2008 Technical Symposium on Computer Science Education Friday, March 14, 2008 Grid Computing at the Undergraduate Level: Can We Do It? Jens Mache.
Job Submission Condor, Globus, Java CoG Kit Young Suk Moon.
Grid Computing I CONDOR.
3-2.1 Topics Grid Computing Meta-schedulers –Condor-G –Gridway Distributed Resource Management Application (DRMAA) © 2010 B. Wilkinson/Clayton Ferner.
Part 6: (Local) Condor A: What is Condor? B: Using (Local) Condor C: Laboratory: Condor.
A Geographically-Distributed, Assignment-Structured Undergraduate Grid Computing Course Mark A. Holliday, Barry Wilkinson, Jeffrey House, Samir Daoud,
Condor Project Computer Sciences Department University of Wisconsin-Madison A Scientist’s Introduction.
3-1.1 Schedulers Slides for Grid Computing: Techniques and Applications by Barry Wilkinson, Chapman & Hall/CRC press, © Chapter 3, pp For.
Grid job submission using HTCondor Andrew Lahiff.
Resource Brokering in the PROGRESS Project Juliusz Pukacki Grid Resource Management Workshop, October 2003.
Grid Compute Resources and Job Management. 2 Local Resource Managers (LRM)‏ Compute resources have a local resource manager (LRM) that controls:  Who.
Dealing with real resources Wednesday Afternoon, 3:00 pm Derek Weitzel OSG Campus Grids University of Nebraska.
Grid Infrastructure group (Charlotte): Barry Wilkinson Jeremy Villalobos Nikul Suthar Keyur Sheth Department of Computer Science UNC-Charlotte March 16,
1 SIGCSE 2008 Technical Symposium on Computer Science Education Thursday, March 13, 2008 Towards a Top-Down Approach to Teaching an Undergraduate Grid.
3-1.1 Schedulers © 2011 B. Wilkinson/Clayton Ferner. Fall 2011 Grid computing course. Modification date: Oct 15, 2011.
1 Short Course on Grid Computing Jornadas Chilenas de Computación 2010 INFONOR-CHILE 2010 November 15th - 19th, 2010 Antofagasta, Chile Dr. Barry Wilkinson.
Grid Security: Authentication Most Grids rely on a Public Key Infrastructure system for issuing credentials. Users are issued long term public and private.
July 11-15, 2005Lecture3: Grid Job Management1 Grid Compute Resources and Job Management.
HTCondor and Workflows: An Introduction HTCondor Week 2015 Kent Wenger.
Outline.1 Grid Computing Fall 2011 Tuesday/Thursday 9:30 am - 10:45 pm Instructors © 2011 B. Wilkinson/Clayton Ferner. Modification date: Aug 22, 2011.
ITCS 4/5145 Parallel Programming, UNC-Charlotte, B. Wilkinson, Dec 26, 2012outline.1 ITCS 4145/5145 Parallel Programming Spring 2013 Barry Wilkinson Department.
1 Introduction to Teaching Grid Computing Dr. Clayton Ferner University of North Carolina Wilmington Dr. Barry Wilkinson University of North Carolina Charlotte.
1 Workshop 9: General purpose computing using GPUs: Developing a hands-on undergraduate course on CUDA programming SIGCSE The 42 nd ACM Technical.
Grid Compute Resources and Job Management. 2 How do we access the grid ?  Command line with tools that you'll use  Specialised applications Ex: Write.
Peter F. Couvares Computer Sciences Department University of Wisconsin-Madison Condor DAGMan: Managing Job.
Grid Compute Resources and Job Management. 2 Grid middleware - “glues” all pieces together Offers services that couple users with remote resources through.
Group 2 Aurora Cain Michael Liu Brian Rozier UNC Charlotte ITCS 4146 Spring 2007.
Intermediate HTCondor: More Workflows Monday pm
Condor DAGMan: Managing Job Dependencies with Condor
Intermediate HTCondor: Workflows Monday pm
Grid Compute Resources and Job Management
Grid Computing Course Development team: Barry Wilkinson and Clayton Ferner (Instructors), and Mark Holliday Student assistants: Jeff House and Sam Daoud.
Job Matching, Handling, and Other HTCondor Features
Presentation transcript:

1 Workshop 20: Teaching a Hands-on Undergraduate Grid Computing Course SIGCSE The 41st ACM Technical Symposium on Computer Science Education Friday March 12, 2010 Dr. Barry Wilkinson University of North Carolina Charlotte Dr. Clayton Ferner University of North Carolina Wilmington © 2010 B. Wilkinson/Clayton Ferner SIGCSE 2010 Workshop 20 session4.ppt Modification date: Feb 27, 2010 Session 4: Overview of other assignments 9:15 pm - 9:40 pm

Condor scheduler assignment Assignment installing GT4 core, creating, deploying, and testing a GT4 Web service GridSphere portlet design assignment Session 4: Overview of other assignments 2

Condor Scheduler Assignment Tasks: Issue various Condor commands Submit jobs to Condor Manage jobs in job queue C and Java programs Vanilla and Java universes Job and resource ClassAds for matching jobs to resources Using DAGMAN to create workflows Assignment done on severs coit-grid03.uncc.edu and coit-grid05.uncc.edu 3

Systems used for assignment coit-grid01.uncc.edu – coit-grid05.uncc.edu coit-grid01 Users (:8080) Portal Dual Xeon processors Four quad-core Xeon processors, 64GB memory, 1.2 TB RAID disk coit- grid04 switch coit- grid05 coit- grid02 coit- grid03 CA User access User’s home directory (/nfs-home) shared across system External interfaces Used in Condor assignment as written 4

5 Name OpSys Arch State Activity LoadAv Mem ActvtyTime LINUX X86_64 Unclaimed Idle :25:04 LINUX X86_64 Unclaimed Idle :53:00 LINUX X86_64 Unclaimed Idle :29:17 LINUX X86_64 Unclaimed Idle :32:14 LINUX X86_64 Unclaimed Idle :59:18 LINUX X86_64 Unclaimed Idle :59:19 LINUX X86_64 Unclaimed Idle :59:20 LINUX X86_64 Unclaimed Idle :59:21 LINUX X86_64 Unclaimed Idle :59:22 LINUX X86_64 Unclaimed Idle :59:23 LINUX X86_64 Unclaimed Idle :59:16 LINUX X86_64 Unclaimed Idle :10:04 LINUX X86_64 Unclaimed Idle :12:43 LINUX X86_64 Unclaimed Idle :59:19 LINUX X86_64 Unclaimed Idle :59:20 LINUX X86_64 Unclaimed Idle :59:21 LINUX X86_64 Unclaimed Idle :59:22 LINUX X86_64 Unclaimed Idle :59:23 LINUX X86_64 Unclaimed Idle :59:16 LINUX X86_64 Unclaimed Idle :59:17 Total Owner Claimed Unclaimed Matched Preempting Backfill X86_64/LINUX Total Check Condor pool status: condor_status.

6 Submitting a job Create a job description file Submit job with condor_submit hostname_test1 Output: condor_submit condor_test1 Submitting job(s). Logging submit event(s). 1 job(s) submitted to cluster 73. Contents of hostname_test1 : # comment Universe = vanilla Executable = /bin/hostname Output = hostname.out Error = hostname.error Log = hostname.log Should_transfer_files = YES When_to_transfer_output = ON_EXIT Queue

Can query status of Condor queue with: condor_q Get output of form: Queue -- Submitter: coit-grid02.uncc.edu : : ID OWNER SUBMITTED RUN_TIME ST PRI SIZE CMD abw 5/23 17: :00:00 I uptime 16 jobs; 1 idle, 0 running, 15 held Status: H (hold), R (running), I (idle, waiting for machine), C (Completed), U (unexpanded, never being run) or X (removed). Monitoring 7

8 Managing job  Set up notifications Different Condor universes  Write and submit C program to compute pi by Monte Carlo method  Rewrite in Java and submit to Java universe ClassAd Job/resource matching  Examine resource ClassAds of computers to find out Java version, memory and performance  Add job ClassAd to job description to make job run only on computer with specific version number, memory and performance Other tasks

Condor’s ClassAd Matchmaking Mechanism Fig 3.14 Include in submit description file Held in assessible file on system 9

10 Condor’s workflow scheduler DAGMAN (Directed Acyclic Graph Manager ) Example # diamond.dag Job A a.sub Job B b.sub Job C c.sub Job D d.sub Parent A Child B C Parent B C Child D Job A Job CJob B Job D Start with: condor_submit_dag diamond.dag

11 Some assignment tasks with DAGMAN Write program to compute π -- specifying different random sequences for each of three instances and then construct a DAG as shown below Job CJob B Job A Job D Compute π Compute average

This assignment is based upon a GT4 service described in the book Globus Toolkit 4 Programming Java Services by Borja Sotomayor and Lisa Childers, All the code comes from this book. The book is based upon the on-line tutorial available at Assignment -- Installing GT4 core, creating, deploying, and testing a GT4 Web service 12

Assignment Goals Show how stateful WSRF Web services can be created and deployed in Globus 4.0. Assignment is to be done on one’s own computer (or a lab computer). Requires one to install Globus 4.0 core and associated software – all available for download. 13

Testing installation Start Container Command: globus-start-container -nosec -nosec specifies “no security”, which simplifies assignment (no need for user certificates) Will display list of deployed services. 14

Standard list of deployed services in core Fig 7.7

Purpose of Service To store an integer called value, acted upon by three methods to: Add a to value Subtract a from value Get value Methods given. Further methods will be implemented. Service is stateful (value retained between accesses). public interface Math { public void add(int a); public void subtract(int a); public int getValueRP(); } 16

Math Web Service Resource Resource properties “value” (integer) “ last operation performed” (string) Client Resource Properties Code provided actually two resource properties 17

Steps in Assignment Preliminary set-up - Install GT4 and associated software Download provided files:  WSDL service interface file  Math web service code (in Java)  WSDD Deployment Descriptor file  Client code (Java) to exercise service Build Math service Deploy Math service into GT 4 container Write and compile the client Start container and execute client. Add functionality to service 18

GT4 container Mathservice After Service Deployed

To learn how to deploy portlets within the Gridsphere Grid portal but on your own computer. Fully understand the underlying software structure of JSR 168 grid portals. Design portlets as front-ends to Grid services and applications. GridSphere portlet design assignment 20

Portlets Fig

Fig 8.15 JSR 168 portlet 22

Step 1: Download and Install Tomcat Step 2: Download and Install Gridsphere

Creating your own portlet Goal Create and deploy a portlet that will accept one number and say whether it is odd or even (“oddeven” portlet) The code for this portlet is given. You simply have to deploy it. Later you will to ceate your own portlets. 24

11a.25 Final portlet displayed in GridSphere

26 Make a portlet that allows a user creates a portlet that can add, subtract, multiply and divide two numbers Developing your own portlet Example

Portlet acting as a front-end to a Web service 27

11.28 More Information on Gridsphere See the links on the Gridsphere portal page