San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org Curriculum Development – 2 for CyberInfrastructure (CI) : Essential Resources & Core Courses MSI C(I)^2 SDSC 27June06 Dr. Kris Stewart San Diego State University/California State University System

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org Partnerships an Asset EOT-PACI partnership provided evaluators (Julie Foertsch and Baine Alexander) U. Wisconsin Leadership through Evaluation, Adaptation and Dissemination (LEAD) center [Susan Millar, Dir.] Having external evaluators helped raised awareness for campus Deans/Chairs of importance of Computational Science It is a costly process, so partner with an evaluator in proposals (as an equal).

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org Where did it begin? 1998/99 Assessment by LEAD Background Workshop in Wisconsin April 1997 to learn about assessment and make it real to the EOT-PACI (NPACI and NCSA Education Teams) NPACI started 01 October 1997 EC/CSE requested assessment for 1998 project

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org Evaluation & Assessment using Outside Wisdom (Foertsch/Alexander) U. Wisconsin LEAD for 1998/99 Ed Center evaluation by Julie Foertsch & Baine Alexander “Integrating High Performance Computing into the Undergraduate curriculum: How PACI and the ECCSE can Succeed” Follow-on Activities (Susan Millar, LEAD) CATS (Classroom Assessment Techniques) FLAG (Field-tested Learning Assessment Guide) SALG (Student Assessment Learning Guide)

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org Grand Challenges for HPC Stewart & Zaslavsky, SC98, HPC=High Performance Computing 1.Faculty system of rewards does not encourage teaching innovations 2.Lack of awareness of HPC technologies already used in research or teaching for different fields 3.Faculty & students unaware of benefits and accomplishments of HPC 4.HPC technologies considered too complex/inaccessible for undergraduate instruction 5.Sequential HPC-related curricula is absent 6.Curricula using very large data sets not widely available 7.Adjust to different learning styles when material is complex 8.Variety of platforms/software leads to fragmented curricula 9.School administration/support staff not ready for HPC 10.Specs of computers and networks below user expectations We had been thinking about this (based on April 97 LEAD Workshop in WI)

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org UCES Paradigm UCES = Undergrad Computational Science & Engineering thanks Tom Marchioro and the “crew”, 1994 My previous exposure to “assessment” How well does numerical approximation match the original problem? Along comes SDSC, NPACI and EOT-PACI

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org 1998/99 Assessment by LEAD Background Workshop in Wisconsin April 1997 to learn about assessment and make it real NPACI starts 01 October 1997 EC/CSE requested assessment for 1998 project Assessment as a Collaboration Preparation SDSU Campus Visit/B.Alexander & J.Foertsch Discussions at SC98- Supercomputing Fall98 Orlando to SDSU faculty gather attitudes

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org Interviews on SDSU Campus LEAD applying info from surveys (Spr 99) Faculty skepticism: Convincing evidence that computer-based tools enhance teaching process? Knowledge of modern computational methods and availability? Incentive from department and insufficient tech support? LEAD Interviews with V.P. Singer, Deans, Chairs Faculty Fellows program identified as a target OUTCOMES: our local infrastructure at SDSU took us more seriously survey instruments refined Online tool (SWB) recog.

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org Assessment not just requirement Rather, found to be vital tool to assist in clarifying student and faculty needs improve prioritization skills validation of focus on human factors to integrate HPC (modeling & visualization) into undergrad curriculum

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org California Education Infrastructure K12 Education (standards based, performance based – now is 1 st year of exit exams before diploma – being challenged) Community Colleges(Freshman/Sophomore) Vocational (and service to local community) University preparation California State University System (24 campuses) University of California (9 campuses, Merced soon) Independents (Stanford U., CalTech, U. Southern California)

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org Involving University Faculty Infrastructure for Change NPACI/SDSU Faculty Fellows Local Support from College Deans and Department Chairs (participation buy-in and faculty recognition) SDSU Academic Advisors (across disciplines) Professional Meeting (SC2001, SIAM, ACM, SIGCSE, your suggestions?)

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org Undergraduate Faculty: A Tough Target Group Obstacles: lack of time, tenure and review considerations, lack of awareness about available technologies Undergraduate faculty (SSRL phone survey 1997 – thanks Doug Coe): ¾ have used WWW often or sometimes (1997), but not in the classroom (only 18% ) The gap between those NEVER using computers in the classroom, and those using them OFTEN, is the largest for untenured faculty, increasing towards tenure review Only 12% of surveyed faculty saw themselves as having a use for HPC applications in courses (higher for Sciences and Engineering) 11% of faculty have students working with computer models OFTEN

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org Using computers in the classroom versus number of years as a faculty member (1997 Faculty Survey by SSRL/Doug Coe funded by Academic Affairs)

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org Primarily graduate facultyPrimarily undergraduate faculty College of Arts & LettersCollege of Sciences 1. 1 OFTEN 2. 2 SOMETIMES 3. 3 RARELY 4. 4 NEVER 5. 9 DK/REF (Missing values) Students Using Computers in the Classroom (1997 Faculty Survey)

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org Strategies for Building Faculty Community Reliance on most enthusiastic and technically advanced instructors who are already using computing and modeling in classes The Faculty Fellows program: Stakeholders: –College Deans - Specific support through faculty release time –Faculty - Compensation, and acknowledgement, of the value of the faculty members contribution Benefits –College –Department (Faculty Fellows as discipline-specific spokespersons for EC/CSE and NPACI) –Faculty (as individuals) –Ed Center on Computational Science and Engineering Building a special infrastructure for curriculum transformation: human, institutional, technical – is a requirement for successful introduction of advanced techniques (since they are more demanding on faculty time and efforts)

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org Faculty Fellows during Faculty Fellows representing departments from five colleges and the Library: Geological Sciences, Geography, Linguistics, Library & Info Access, Music, Education Technology, Biology, Computer Engineering, Computer Science, Business Information Systems Bi-weekly meetings at the Ed Center Faculty Fellows as “ambassadors” of computational science Partnership with LEAD for evaluation during

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org Faculty Fellows Fall 01 Synergy among themselves and with their chairs and deans People, Time, Support, Recognition …

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org Lessons Learned from ECCSE Institutional support required for program to be sustainable Individual reform-ready faculty is focus for support Infrastructure: Build a Synergistic Environment (across disciplines) for Faculty Continuous monitoring through interviews, surveys, discussions

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org More Outside Wisdom – JSB John Seely Brown – SDSU

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org Apply JSB Insights to my Curriculum JSB highlights multimedia literacy CS596 3d Game Programming - Student Learning Outcomes student group presentations were great

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org Apply JSB Insights Students have grown up digital (natives); faculty are analog (immigrants) Capitalize on creativity by honoring the vernacular of today’s students (multimedia- literate) Communicate complexity simply (great skill) MIT’s architecture studio – all work in public (development and critique) – in context Learning to learn “in situ” is key

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org CS 575 Supercomputing Crucial topic: Computer Security (Students may not appreciate the importance yet) ACM/IEEE SC 1995 Conference (SC'95) p. 19 HPC Undergraduate Curriculum Development at SDSU Using SDSC ResourcesACM/IEEE SC 1995 Conference (SC'95) Technology – changing frequently and course not taught since Fall 2003 – might be useful as template for you. Text: High Performance Computing, Oreilley Better – who is providing your class resources? Vendors and providers (SDSC, NCSA, PSC, OSU, …) have workshops. Signup and make new friends. Personally, I recommend Mary Thomas’ Grid course, next

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org Grid Curriculum – Mary Thomas Introduction to Grid Computing rohan.sdsu.edu/faculty/mthomas/courses/ spring05/cs696/index.html (all on 1 line) Areas of Research Grid Portal Toolkit Grid-enabled Web Services DOE portal development on the SciDAC Grid Grid Portals Information Repository Middleware for Grid Portals Grid-Enabled Portals for Advanced Computing NSF Extensible Terascale Facility (ETF) or TeraGrid Performance (MS Thesis)

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org Mary Thomas Grid Computing Prerequisites: This course is designed for Computer Science, Computational Science, and Math graduate students with suitable preparation. Students must have graduate standing, have taken courses such as Java (e.g. CS435, CS535) or C++, CS580 (or equivalent). Undergraduates with highly advanced skills who have taken courses such as CS535, CS575, CS558 will be considered. Priorities will be given to students interested in active research careers

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org Mary Thomas – Students must know Students must know: Java (optionally Python or Smalltalk) HTML and other Web Technologies Students should be familiar with: XML, CSS, XSLT computational science, parallel programming Web Services, SOAP, HTTP and other transport protocols basics of client-server programing some database experience would be useful

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org COURSE OUTLINE (TENTATIVE): Introduction and Overview of Grid Computing: Trends, Challenges, Technologies, and Applications (1 week) Web Services and related technologies: XML, XML-Schemas, WSDL, WSRF, etc (2 week) Distributed Object Technology for Grid computing(1 week) Overview of Grid Middleware (3 weeks) Distributed Object Technology for Grid computing (OGSA, WS-RF) (1 week) Grid Middleware: GridPort Toolkit, javaCoG, Globus, GSI, FTP, etc. (1 week) Developing Grid Services (2 weeks) Introduction to Grid Computing Portals (1 week) Overview of Grid Portal Frameworks (JetSpeed, OGSA) (2 weeks) Deploying Grid Portals (1 week)

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org Readings for Week 1 Ian Foster Paper What is the Grid? A Three-point Checklist Chapters 1 and 4 in text: “The Grid 2” by Foster and Kesselman Investigate grid related Web sites (see course web pages) Hint: use the Google “What is X” command

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org Why do We Have Grid Computing? The term was coined ca 1996 by Ian Foster and Carl Kesselman Used to describe software that was needed by the rapidly growing, highly advanced community of high-performance Computing (HPC) Resources that scale with technologies: Supercomputers (MFlops in96, but now using TFlops) Big and not portable Large data sets (GB in 96, but now peta-bytes) Need fast networks to move data around to resources Need security: NSF (and other gov agencies) spend money to build infrastructure, so it is hard to get access

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org A ‘small’ grid (UTGrid) Research campus Main campus TACC PWR4 TACC Vis NOC Ext nets GAATN ACES Switch CMS TACC Storage Switch TACC Cluster TACC Cluster ICES Cluster PGE Cluster Switch PGE

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org Portals: typical Hibernate Applications Remote Services

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org What is Grid Computing? Is it a new, unique idea or the next generation of distibuted or meta-computing? See: Ian Foster Paper “What is the Grid? A Three-point Checklist”

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org Ian Foster’s 3 point checklist A Grid is a system that is able to coordinate “resources that are not subject to centralized control” Use “standard, open, general-purpose protocols and interfaces” “to deliver nontrivial qualities of service.” What does this mean? We will try to understand this in this course.

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org Defining Grid Computing There are several competing definitions for “The Grid” and Grid computing These definitions tend to focus on: Implementation of Distributed computing A common set of interfaces, tools and APIs Some stress the inter-institutional aspect of grids and Virtual Organizations “The Virtualization of Resources” abstraction of resources

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org What is Grid and Grid Computing? Grid computing promises a standard, ‘complete’ set of distributed computing capabilities There is a lot of hype around grid computing Traditional users need to get work done now! Some CS researchers see it as a fad But there is real-world value! In e-science and e-business

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org What is Grid and Grid Computing? Grid computing must provide basic functions 1.resource discovery and information collection & publishing 2.data management on and between resources 3.process management on and between resources 4.common security mechanism underlying the above 5.process and session recording/accounting Current grid computing tools such as Globus provide most of the above at some level The current capabilities are incomplete New web service based-standard will help current tools become interoperable.

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org Mary Thomas - Extensive Web Site The previous 15 slides were taken directly from Mary Thomas’ web page, as a “teaser”. The 15 week semester is completely available for reading (and using) from Mary Thomas’ Web Site www-rohan.sdsu.edu/faculty/mthomas/courses/spring05/cs696/index.html Or Google ( mary thomas site:rohan.sdsu.edu )

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org 3d Game Programming Upper Division Course for Coders Using the Torque Game Engine from Torque is an object oriented scripting language with an extensive library of game capabilities Text: Ken Finney, Thomson Pub

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org 3d Game Programming Spr 06 The first offering of this course was Spr06 and the course will be reoffered Spr07. The class web page will be updated for the next offering during Fall 07. Please check back on this evolving curricular development.

San Diego State University / Ed Center on Computational Science & Engineering MSI C(I)^2 at SDSC 26-30June06 educationgrid.org More Information? Kris Stewart Professor, Computer Science, SDSU Director, ECCSE, NPACI/CSU This work supported by NSF