General Foils 1997 Collection of General Foils January-December 1997 See: http://www.npac.syr.edu/users/gcf/generalfoils97 Geoffrey Fox Syracuse University 111 College Place Syracuse New York 13244-4100 Abstract of General 1997 Foils This is a general collection of foils for a variety of projects Initially Education Activities! Then PetaFlop Remarks JISC Activity in United Kingdom Black Hole Message Passing Analysis Education in a Box -- Initial Steps to Understand Web Based Education Certificate in Computational Science offered at Harbin Institute of Technology (North China) Spring 1996 -- started March 1,1996 -- is a prototype of Syracuse University extension school Although Internet Connections to China is pretty dreadful, we still use Web Technology but package Electronic Course material on a Pentium PC at Harbin running WindowsNT and Java Can use Oracle and Digital Video Server Support as well Can use RealAudio digital audio (I record my lectures as I give them) This is updated in batch mode by Syracuse -- China Connection Students in China Interact with mentors in China and with NPAC via Internet (web and email) Project got good reviews from students and press! Project designed by Fox,Leskiw(NPAC), Xiaoming Li(Harbin and Peking visiting NPAC) Scope of International Certificate in Computational Science Graduate Students -- initially 6 guinea pigs presented with certificates July 96 4 Courses -- each 10 1 hour lectures -- offered Spring 1996 over 20 week period ICPS 700: Introduction to Simulation Track of Computational Science Parallel Architectures, Software, Grand Challenges, Parallel Algorithms (scaled back version of CPS615) ICPS 701: Parallel Programming Laboratory Use of MPI to do real problems on a Cluster of Workstations/PC's ICPS710: World Wide Web Technology Scaled back version of CPS616 ICPS711: Web Programming Laboratory companion course to ICPS700. Initially will use Java PERL4 and PERL5. Next Steps in International Certificates Important Lesson from my trip to China last summer Exciting dynamic country where all learn English But many don't understand difficult material in English! So we have used initial students to translate base material into Chinese We intend broad Distribution via CD-ROM of this combined modern practical computer science includes MPI and HPF base documents in Chinese HTML China wants me to give a one week Web Tutorial (in English) at Peking this May 97 Current Computational Science (CPS) Core Courses SImulation Track -- Graduate CPS615 Overview of Simulation Track Computational Science CPS713 Case Studies in Simulation Applications of Computational Science Masters in Computational Science built around these core courses CPS615,713 Simulation Track -- Undergraduate CPS 311,312,313,314 Core Material and Laboratory with topics similar to CPS615/713 Senior Project CPS 451 10 Possible PetaFlop Applications Nuclear Weopens Stewardship (ASCI) Cryptology and Digital Signal Processing Satellite Data Analysis Climate and Environmental Modeling 3-D Protein Molecule Reconstruction Real-Time Medical Imaging Severe Storm Forecasting Design of Advanced Aircraft DNA Sequence Matching Molecular Simulations for nanotechnology Large Scale Economic Modelling Intelligent Planetary Spacecraft Remarks on Petaflop Technology and National Program See: http://www.npac.syr.edu/users/gcf/petaflopjan97 Geoffrey Fox Syracuse University 111 College Place Syracuse New York 13244-4100 Abstract of PetaFlop Presentation Jan 97 This uses material from Paul Smith and Peter Kogge as well as Fox We describe the "National PetaFlop Study(s)" and what you can expect with or without a specific initiative We discuss traditional, Processor in Memory, Superconducting, Special Purpose architectures as well as future Quantum Computers! We survey possible applications, new needs and opportunities for software as well as the technologies and designs for new machines one can expect in the year 2007! We review findings of studies and structure of a possible initiative What did we do Right and Wrong in HPCC Software ? There are at least 3 issues in HPCC software Fundamental problems in memory bandwidth/latency; data distribution and nifty parallel algorithms Engineering implementation including scaling robust systems Useability -- is the HPCC programming environment attractive and productive We did well on "fundamental issues" but failed on "in principle" straightforward but costly engineering and User Interface areas Petaflop initiative faces much harder (but probably soluble) fundamental issues but is bound to fail unless there is a strong (well enough funded) implementation program Commercial motivation will generate insufficient funds! Remarks on Parallel Computing and Computational Science Education See: http://www.npac.syr.edu/users/gcf/compsci97 Geoffrey Fox Syracuse University 111 College Place Syracuse New York 13244-4100 Abstract of Computational Science Presentation January 1997 We describe role of computational science as capturing academic integration of advanced compuation into general curricula We use a recent review of a program in the United Kingdom to illustrate concepts Clusters of PC's are natural deployment mechanism We point out that this can be applied successfully in both simulation (parallel computing) and information arena (the web) We discuss the role of scalable certificates as another way of recognizing knowledge as opposed traditional minors, masters and other degrees We illustrate these ideas with program at Syracuse Initial Discussion of United Kingdom JISC New Technologies Program Both HPCC Training and Education as well as Clustered Computing Discussion of Syracuse Program in Simulation as well as Information Arenas Undergraduate and Graduate The Role of Scalable Certificates offered outside Traditional University Credit Courses and Degrees Including China Internet Experiment Wrap Up Discussion of United Kingdom JISC New Technologies Program Both HPCC Training and Education as well as Clustered Computing Explicit Curricula in Syracuse Program in Simulation as well as Information Arenas Undergraduate and Graduate Review of Curricula and Training Material Available on the Web We seem to be missing algorithm discussions! Sample HPF and DAGH codes These are for different parts of UNIGRID codes HPF diff subroutine calculates differentials DAGH is sample solver GridHierarchy to end_forall are all DAGH commands setting up the Grid hierarchy and involking solver MaxLev=1 in UNIGRID case f_inverse is C++ function that calls Fortran t2_inverse inverse is original Fortran77 solver invoked by DAGH with grids set up with suitable Ghost regions so one can use original sequential Fortran code Some Conclusions on Programming Models This application does NOT suggest Message Passing is Obsolete! Rather message passing is only "complete" solution and the alternative higher level approaches only express parts of the problem This application suggests a tool-kit approach where you can link together many environments and not be tied to a single system Possible an Interpreted front end and/or use of systems like WebFlow This stresses need for interoperable subsystems rather than ambitious complete solutions that can't be completed The Role of Java in Scientific Computing SIAM Meeting Mar 16,97 Minneapolis Session on OO Methods in Parallel Processing organized by Dennis Gannon http://www.npac.syr.edu/users/gcf/javainhpccmar97 Geoffrey Fox Syracuse University 111 College Place Syracuse New York 13244-4100 Abstract of Java in HPCC Presentation Please go to URL http://www.npac.syr.edu/projects/javaforcse We describe 3 major areas where Java (and other Web Technologies) can have significant impact 1) Java can be used to build user Interfaces and here we describe the Virtual Programming Laboratory VPL 2) Java can support coarse grain integration and metacomputing Here we describe both WebFlow (AVS on the web) and Tango (a Java Collaboratory) Both Computational Steering and "Forces Modelling" seem naturally implemented with this technology 3) Java as a traditional compiled language for computational kernels Here we discuss sequential and parallel issues Integration of Interpreted and Compiled Environments Integration of Interpreted and Compiled Environments Java raises issue of role of Interpreters versus Compilers Success of systems like MATLAB and languages like APL show relevance of interpreters in Scientic Computing PERL, JavaScript, TcL, Visual Basic etc. indicate growing use of Interpreters in other domain Natural as computers get faster! We suggest that integration of Interpreters and compilers is an important research issue and could suggest new models for parallelism We need both Compilers and Interpreters!! Optimizing Interpreters (as in JIT for Java) A library model where interpreted toolkits invoke lovingly parallelized high performance libraries Natural linkage to interpreted data analysis / visualization Structure of NPAC March 97 -- Programs Classic HPCC Research -- Programming Environments and Applications and links of HPCC to Web Technology Basic Information Technology -- Databases, Web, Collaboration, Networking Healthcare/Human Computer Interfaces -- "Center for Really Neat Research" -- work with disabled users Education Technology -- TANGO WebWisdom etc. InfoMall Technology Transfer 4 spin offs from NPAC in last year plus existing CNY companies InfoMall MidHudson supports technology in MidHudson (exIBMers) about 10 very active companies Education Delivery NPAC taught 7 full courses last 2 semesters with only 1 regular faculty! Structure of NPAC March 97 -- People and Infrastructure 19 PhD level Researchers, 5 senior and 4 junior visitors from Poland About 30 students (about half funded, others "volunteers") Local and International Collaboraters CRPC nationally; Peking and Harbin in China Modest Computational facilities focused on high quality information subsystems and networks Large enough computers for rapid prototyping PC and workstation clusters Theme is generalized computational science -- applied computer science and applications of relevance to real world in simulation and information arenas People and Infrastructure in NPAC 19 PhD level Researchers, 5 senior and 4 junior visitors from Poland About 30 students (about half funded, others "volunteers") Local and International Collaboraters Member of leading NSF and Darpa activities Work with collaboraters in Europe and China State of the Art Practical Computational facilities focused on high quality information subsystems and networks Large enough computers for rapid prototyping PC and workstation clusters Theme is leading edge computer science and its applications of relevance to real world for academia, business and community Nature of NPAC Programs Basic Information Technology -- Databases, Web, Collaboration, Networking Healthcare/Human Computer Interfaces -- "Center for Really Neat Research" -- work with disabled users Education Technology -- TANGO, Video Servers WebWisdom etc. High Performance Computing and Communication Research -- Programming Environments and Applications and links of HPCC to Web Technology InfoMall Technology Transfer 4 spin offs from NPAC in last year plus existing CNY companies InfoMall MidHudson supports technology in MidHudson (exIBMers) about 10 very active companies Education Delivery NPAC taught 7 full courses last 2 semesters and so uses its own technology! Education and Training at NPAC Advances in networking and compute technology with growing pervasive access to the World Wide Web will have profound implications for education and Training New Business models for the Virtual University / Training Center with international clients and a Lifelong learning model Better Curricula should allow easier learning for students! Nobody (yet) knows what will work! Through Syracuse University NPAC is working with several departments and the local K-12 schools Through InfoMall NPAC is supporting 3 companies in this area and we expect this to grow! New Types of Curricula Material Web Delivery of HTML versions of traditional curricula with linked associated audio of lecturers Java and VRML2.0 animated resources to provide new insights to complex problems Digital Video Servers with indices Online Laboratories for science, programming etc. Database search oif full curricula to find topics of interest New Delivery and Assesment Concepts The new technology supports both compute enhanced classroom and distance delivery Distance delivery can be synchronous (teacher and students interacting in real time) or asynchronous (student self paced study with "batch" interactions by electronic mail/bulletin boards etc.) Collaboration technology such as NPAC's TANGO supports these new models Integration of Databases with delivery system allows logging of either pages accessed or even of full multimedia sessions Leads to new approachs to assesment! What is NPAC's Role? NPAC has developed some unique capabilities in WebWisdom Delivery, TANGO collaboration and Video Server areas NPAC has state of art experience in areas such as VRML2 Java and Web-linked Databases NPAC can deliver customized Web Education technology linking components of interest NPAC can deliver training in several areas -- either in class or at a distance Broad courses in Web Technologies or High Performance Computing Collaboration The Web links people to both Information Sources and to other people Currently most progress has been in access and searching of information in HTML(text), Animated Java Applets, and 3D VRML Images AOL Compuserve and Web have great success with bulletin boards and "chat systems" for forums, clubs and other electronic societies New generation of Java Collaboration Systems will greatly increase capabilities of electronic collaboration Supercede current videoconferencing systems and fuse with Web information resources The TANGO Collaboration System Ability to define Interacting Group of people with either pure peer or teacher-student type relationship Integrated digital VideoConferencing among participants Shared Browser (HTML Java VRML Pages) Shared Chat System for text messages Shared Whiteboard for drawing Multimedia Mail Linked Digital Video Server The TANGO Collaboration System Under development is integrated database for multimedia logging of sessions Support for multiple but linked "electronic societies" with different rooms Shared Applications such as Microsoft Word Has well defined Interface for linking other applications in C C++ Java JavaScript This has been tested in several cases JavaScript TangoWisdom links TANGO to WebWisdom educational delivery system Two and Three dimensional C++ Geographical Information Systems Java Interactive Weather Forecast System Java Physics Simulations used in Education (planetary motion) Some TANGO Applications at NPAC Initial application is Decision Support System aimed at emergency planning and command and control Can be extended to Business Enterprise Support Distributed simulations involving machines, people and computers in the loop HealthCare from a rich TeleMedicine environment to Home Healthcare including support for disabled individuals Distance Education -- TANGOWisdom "TANGOlite" or "HomeTANGO" to support low bandwidth links for clubs and forums The National/Next Generation Information Infrastructure in the service of Education Training and Computational Science RCI Conference Williamsburg April 17,97 Geoffrey Fox Syracuse University 111 College Place Syracuse New York 13244-4100 Abstract of NII for Education and HPCC We describe issues affecting the depoloyment and impact of the NII/Internet2/NGI on HPCC and Education These two areas are chosen because they are applications I have some insight in Other areas will be equally or more important Probably the biggest issue will be determining new "enterprise models" and working out how to change organizations Java JavaScript VRML2 Multimedia Web (audio/video) Web-linked databases, Web Collaboration are key component technologies Web Software is the Best! By definition, Web Software will be the "best" software ever built because it has the largest market (and so greatest leverage of investment dollars) and most creative business model (harness the world's best minds together with open interfaces) As Web Software is "distributed computing infrastructure", one will need to customize and integrate for each application Note PC software and perhaps even IBM business OS was high quality software but not so open and not a complete model One should build upwards from the "democractic Web" e.g. up from POTS --> ISDN/Cable Modem/ADSL --> ... Not down from ..........<--- ATM This allows you to both deliver your application to the general public (not always required but often desireable) and use the best leveraged software Note Web Software tends to offer highest functionality as opposed to highest performance and HPCC often requires different trade-offs We have the Web Tools in Place - I ! Web Technology is still uncertain and there may be major changes but "enough" capabilities are in place to build very general (~all) applications MicroSoft v Sun / Netscape uncertainities Rapidly evolving Standards and a mechanism to get rapid consensus Fortran 77 -> Fortran90 --> HPF --> Fortran2000 (23 years) VRML Idea (1994) --> VRML1 deployed (95) --> VRML2 deployed (early 97) (2.3 years) VRML2 (3D datastructures and Java(Script) enabled methods) is more sophisticated than Fortran ? Java development time was also fast compared to "traditional" standard adoption times We have the Web Tools in Place - II! Classic Web: HTTP Mime HTML CGI Perl etc. Java and JavaScript Compiled to almost compiled (applet) to fully Interpreted Programming Language VRML2 as a dynamic 3D Datastructure for products and their simulation object Java Database Connectivity (JDBC) and general Web linked databases Dynamic Java Servers and Clients Rich Web Collaboration environment building electronic societies Security -- still needs maturing as very clumsy or non existent at present in many cases Compression/ Quality of Service for Web Multimedia Do need higher bandwidth for real video although current POTS (perhaps ISDN needed) adequate for conferencing Emerging Web Object model including integration of Corba (see JavaBeans and Orblets) What is the Problem -I? 1)Compute Power ? Maybe 2)Network Bandwidth? In some cases 3)Implementing / Discovering new ways of doing Business? Usually the major issue e.g. Can academic Compute Science compete with Industry? Importance of industry implies "we" need a greater focus on "products" Which organizations will still be here 10 years from now ? Consider University education as an example Business Model for Education Many US Universities export education. Is global communication and "Virtual University Technology" good or bad? There will be access to more students You can teach them "in place". Maybe not so many will come and spend money in US New organizations can compete as don't need so much capital investment to become a Virtual as a Real University! One could get changes in teaching versus residential function of Universities Many other Enterprises have similar Challenges and Opportunities! What is the Problem -II? 4)Web Technologies are very rich and are perhaps 10 times as complicated as HPCC and Parallel Programming Much harder to match the drumbeat of web than drumbeat of HPCC I use a research <--> Teach iterative cycle to learn and understand significance of new technologies I now spend a great deal more time teaching than I used to! The amount of important material appears to require this! 5)Rapid evolution of technologies implies that any "product" is bound to be out of date Use (Web) standards to make it easier to incorporate new standards But VRML2 very different from VRML1 JDBC very different from Oracle WoW database technology Opportunities in HPCC - I? Problem with HPCC is not ideas but rather finding enough people to implement robust rich software No more people so take software from elsewhere View parallel computing as a special case of distributed computing with tighter synchronization and lower latency Enhance Web technologies accordingly and use as new HPCC software infrastructure Java/JavaScript front ends for interoperability and visualization is first step This will link to wrappers around existing Fortran/C++/HPF/MPI code Users will find it convenient to modify this front end as so this is trojan horse which will lead for existing Fortran users converting to Java Opportunities in HPCC - II? Parallel Java is inevitable and indeed Java will replace Fortran and C++ in general scientific computing We have evaluated what it takes to produce Java interfaces (wrappers) to existing libraries and will deliver this on compiler runtime (see Java-MPI at our Web Site) The (commercial) Web itself will lead to "coarse grain software integration" in AVS like data flow environments This will implement metacomputing with a rather different model than before Link mobile Web modules rather than decomposing problems Web Collaboration technology can revolutionize computational steering Java Servers will link Instruments, People, Computers Opportunities in Education - I HTML Java etc will give higher quality basic Curricula with glossaries, database search, hyperlinking etc. Virtual Laboratories for Science, Programming etc. Use VRML2 for game like science exploratory learning We have JavaScript Perl Java VRML SQL HPF and MPI with Web Interfaces Voice (RealAudio) and video capture of teacher as augmentation of lecture curricula material record when teacher is delivering to capture adrenalin enhancements! Searchable video archives for finding critical few minutes clips to use in class More useful than searching 10,000 movies for "Web BlockBuster" which really makes poor use of digital video Opportunities in Education - II Take Web electronic society technology which link people together and use to join teachers mentors and students Whereas HPCC needs "upscale" collaboration technology, education should be designed so that one can link to peoples' homes and so use democratic technology Meanwhile can look at VRML2 for Immersive virtual learning environments All of these technologies can be used either synchronously (organized sessions) or asynchronously (students learn in their own time and interact with teacher/mentor in batch mode) Log educational sessions (including multimedia material) in databases and use for assesment Opportunities in Education - Applications Initial applications include corporate training and University continuing education Students more motivated than at lower levels K-12-Undergraduate? This is outreach to people wanting to learn new skills and so critical as all organizations redefine themselves Also use in Chinese Universities and HBCU's in the U.S. where we can help faculty upgrade their curricula NPAC will have all these capabilities in a deployable system by end of 97. And the Future is ........ The Use of Java in Scientific Computing and PetaFlop Systems! See: http://www.npac.syr.edu/users/gcf/petaflopapril97 Talks in Germany April 21-23 1997 Geoffrey Fox Syracuse University 111 College Place Syracuse New York 13244-4100 Java and PetaFlop Systems Abstract This talk was presented at the "Kolloquium uber Parallelverarbeitung in technisch-naturwissenschaftlichen" at Aachen April 21, 1997 and (without PetaFlop comments) at the Pallas Software company (Bruhl Germany) April 23, 1997 The visit was sponsored by GMD Bonn SCAI (Ulrich Trottenberg) We discussed the expected PetaFlop architectures with their challenges and then the new software approaches suggested by the Web Please go to URL http://www.npac.syr.edu/projects/javaforcse We describe 3 major areas where Java (and other Web Technologies) can have significant impact 1) Java can be used to build user Interfaces and here we describe the Virtual Programming Laboratory VPL 2) Java can support coarse grain integration and metacomputing Here we describe both WebFlow (AVS on the web) and Tango (a Java Collaboratory) Both Computational Steering and "Forces Modelling" seem naturally implemented with this technology 3) Java as a traditional compiled language for computational kernels Integration of Interpreted and Compiled Environments The latter discussion included details on parallel Java What is Relevance of VRML(2) for Scientific Computing? VRML naturally gives 3D visualization with usual Web advantage of running on PC's and Workstations Its universality implies can use in industry to specify products so can design, manufacture and market from the same(related) specification Should impact PDES/STEP and such industry product specification standards VRML will need extension to handle this but it is a good start and allows user defined types VRML and Parallel Computing? Breakup up a VRML image with INLINE construct to allow parallel rendering of complex images (i.e. parallel VRML viewer) Invocation of parallel simulation programs as a VRML2 Script e.g. VRML2 cars invoke parallel crash codes when they hit each other! The Use of Web Technologies (at NPAC) in HPCC Collaboration and Education http://www.npac.syr.edu/users/gcf/decmay97 Talk at Digital Equipment Corp May 8 1997 Geoffrey Fox Syracuse University 111 College Place Syracuse New York 13244-4100 Abstract of DEC Presentation We described the overall structure and programs of NPAC We illustrate how these fit together in the application of Web Technologies to HPCC and education We describe guiding principles of HPCC -- what we have learnt -- and how this suggests that field needs to build more robust functional modular systems We suggest Java Servers are a natural unifying approach to Software Integration enabling collaborative design, computational steering etc. We present our 4 layer model of HPCC Education benefits from HPCC simulations and can use much of the same integration software Some Concepts Learnt from HPCC Initiative Ideas from HPCC research Good! Not enough people/funding in field to implement robust production systems Must re-use as much software (including infrastructure software) as possible Similarly must build HPCC software in a modular fashion with small enough modules that smallish groups can build effectively Different modules are likely to use different base technologies (Fortran v Java v C++ etc.) and so interoperability essential! No silver bullet on the horizon - maybe pessimistic but implies better HPCC environments implies better implementations of existing ideas. Need to support both production use of MPP's and "rapid prototyping" in development of new applications - latter is not well supported by current HPCC software systems even though need parallel support for prototyping of new 3D simulations Some Relevant Technical Trends PC and workstation clusters are of growing important and this typically distributed memory people's technology is contrasted with distributed shared memory tightly coupled MPP's. Computational science moving to multidisciplinary (multi-component) applications Corresponding growing use of databases (for data-intensive applications) Interoperability between disparate heterogeneous platforms, support of multidisciplinary applications, and metacomputing are three related important areas "full metacomputing" (decompose general problem on general networked resources) may not be relevant The Web is delivering a new operating environment (WebWindows) and a rich distributed computing software infrastructure with especially excellent support for software integration There is a need for a new scalable technical operating system (NT v UNIX v WebWindows) HPCC Software issues We can distinguish Decomposition and Integration Decomposition is performed by an HPF or other Parallelizing compiler; or by a user writing a Fortran + Message Passing code "by hand" MPI integrates decomposed parts together with high bandwidth latency constraints Systems such as AVS integrate larger modules together and much of "software engineering" (modular style of programming) involved with this Web is a powerful integration model suitable for large coarse modules with modest latency and sometimes modest bandwidth requirements naturally "integrate" data(bases), people, computation together Collaboration, computational steering, multidisciplinary science are all integration and not decomposition problems! Use of Java Servers in HPCC We suggest a hierarchical computing model with high functionality modest performance integration/coordination layer controlling classic low level HPCC high performance layer At Lowest layer hardest problem is decomposition At Top (computing) layer, flexible Integration is problem as decomposition natural User Interface gives third level Services at the 3 Levels User View: Interoperable Web Interface as in VPL (Virtual Programming Laboratory) Shared Visualization applets for simulation and performance data client collaboration services -- audio/video conferencing, chat board Integration Level: Network of (Java) Servers linking different clients (aka collaboration), Web-linked databases Metacomputing and Interoperable module linkage High Performance Level: Classic HPCC technologies where low latency and high bandwidth needed! Some General Lessons from BBH Grand Challenge Collaboration does use Fortran 90 (due to many tensors) but does not really use HPCC technologies Mainly because collaboration in rapid prototyping mode and runs "modest" (several hours on R10000 chip) and so no compelling motivation HPCC does not support rapid prototyping in this problem as NO approach easy to use However HPCC will be essential for production runs and we have shown that HPCC technologies will work! Expects to use DAGH as needs HPF2 compilers If either DAGH or HPF used, one will need additional support (specialized translators, MPI extrinsics) Some Conclusions on Programming Models This application does NOT suggest that any system (either HPF DAGH MPI) is either Obsolete or the solution! Message passing is only "complete" solution (but unacceptable for rapid prototyping) while the alternative higher level approaches express parts of the problem which better programming environments which could support rapid prototyping This application suggests a tool-kit approach where you can link together many environments and not be tied to a single system Possible an Interpreted front end and/or use of systems like WebFlow This stresses need for interoperable subsystems rather than ambitious complete solutions that can't be completed A note on Numerical Algorithms One probably needs multigrid elliptic solvers which are compatible with hyperbolic evolution using Adaptive Mesh Refinement One might to solve elliptic equations at each time step for subsidary variables However neither Grand Challenge nor (significant) part of numerical analysis community seemed to have experience in this and there appears no "off the shelf" solution. Remarks on Interpreted HPJava -- Data Parallel Java We have proposed an approach which uses client native classes for "compiler runtime" or linkage to server and follows an HPF style with an interpreted front-end like Matlab or APL or "host" programming model as in *LISP on CM-2 e.g. A = HParray.matmul(B,C) Technically Generalizes HPF Interpreter we prototyped in 1993 Interpreters and objects are great as long as "coarse-grain" and do not represent necessarily bad performance i.e. access arrays not array-elements This leads to Java wrappers invoked by HPF-style Java(Script) interpreter which interfaces to native HPF/C++/Java server implementations. e.g. access HPF array Ahpf elements from Java with wrapper object A HParray A = new HParrayConstructor("Ahpf"); A.grabelement(1,100) Integrated Compiled and Interpreted Environments I Note you can "pre-load" server and interpreted front end with all necessary information about server program by preprocessing latter with HPF (Java C++) front end VRML2 scripting nodes do this If HPF at server side, it is natural to interrupt running program at "extrinsic" interface. Erol Akarsu at NPAC has implement visualization and debugging commands at this interrupt However can extend to allow any type of interpreted execution with user request language Java or JavaScript or pure visual interface (as in WebFlow) Integrated Compiled and Interpreted Environments II User Interpreted Commands invoking preloaded Java/Javascript objects Server invokes extrinsic processor (true interpreter or compiler invoked dynamically) Running Program interruptable at extrinsic interfaces And the Future is ........ The Use of Java and the Web in Scientific Computing See: http://www.npac.syr.edu/users/gcf/chinamay97 http://www.npac.syr.edu/projects/javaforcse Talks in Hefei and Beijing China May 27,30 1997 Geoffrey Fox Syracuse University 111 College Place Syracuse New York 13244-4100 Java and Future HPCC Abstract The talk starts with an overview of NPAC and a description of our CPSWT CDROM with Chinese translation of Computational Science courseware We then describe 3 major areas where Java (and other Web Technologies) can have significant impact 1) Java can be used to build user Interfaces and here we describe the Virtual Programming Laboratory VPL 2) Java can support coarse grain integration and metacomputing Here we describe both WebFlow (AVS on the web) and Tango (a Java Collaboratory) Both Computational Steering and "Forces Modelling" seem naturally implemented with this technology 3) Java as a traditional compiled language for computational kernels The latter discussion included details on parallel Java We show how this allows one to build integrated Interpreted and Compiled Environments What is CPSWT CD-ROM? Chinese translation of material in "International Certificate in Computational Science" ICPSE http://www.npac.syr.edu/projects/icpse_spring96 offered over the Internet spring 1996 from Syracuse to Harbin Institute of Technology Translated by Xiaoming Li (Harbin,Peking,Syracuse) and Faculty and Students at Harbin CPSWT is HPCC Computational Science and Web Technologies The Simulation and Information tracks of Computational Science at Syracuse We will add to NPAC Home Page, a link to be able order copies of this free CD-ROM Contents of CPSWT CD-ROM Course I: High Performance Simulation Introduction to CPS MPI, Fortran90, HPF (definitions, tutorial) Applications to Iterative Methods for Partial Differential Equations Multi Threaded Systems Course II: Information (Web) Technologies Introduction to Web Technologies and their Applications PERL JavaScript Java VRML NPAC Presentation at ARL on PET Support Effort June 5 and 6, 1997 David Bernholdt, Gang Cheng, Geoffrey Fox, Wojtek Furmanski Syracuse University 111 College Place Syracuse New York 13244-4100 ARL Statement of Work -- Preamble I This covers a planning period which will be used to prepare background information and get initial start on some activities which will go into "production mode" as soon as possible. This plan is motivated by guidelines for PET activities emerging from recent meeting at Houston which include: Need for clear early successes linked to an overall long term vision. Particular attention to visible PET wide activities. FMS use of HPCC is currently modest and straightforward "porting" approach for existing codes is not likely to exhibit early successes (One exception is noted in next bullet). ARL Statement of Work -- Preamble II Thus our major initial PET strategy is based on "new ways of doing things" using web technologies (especially Java and VRML 2.0) with linkage to Corba and datbases. These have natural applicability to distributed virtual environments and "human/machine in the loop" distributed simulations. One simple early use of HPCC is to host parallel versions of large scale embedded simulations such as a weather simulation which is part of some FMS models. ARL reports to NPAC that there is interest in large scale parallel databases. There is some synergy between IMT and FMS which should be explored. Training in advanced Web and database technologies at ARL could be useful. PET-wide software repositories are interesting (but postponed until next year) ARL Statement of Work -- Deliverables I Exchange of information describing NPAC and ARL activities One or more visits by Fox and NPAC staff to ARL involving NPAC presentations to ARL and vice versa. This first one! Distributed Computing and Web Technology for FMS and IMT Examination of the requirements of FMS and IMT and how new and emerging software technologies can be used to help meet them. Deliverables White paper describing the results of the study (17 August) Analysis and report on network traffic from simulation data to be provided by Ginny To (17 August) Coordinate with Rome Labs to allow web-based Command and Control demonstration to be implemented on ARLĘcomputer systems in Year 2 ARL Statement of Work -- Deliverables II Parallel Databases for FMS, IMT, and other ARLĘActivities Work is Examination of the role of parallel database Deliverable is White paper describing the results of the study (17 August) Training has 3 topics Elementary Web-Linked Databases (delivery after 31 May) NPAC has the bulk of these materials prepared and can offer this course earlier than the August time frame if desired Additional Training on Parallel Databases and Web-Linked Databases (materials by 1 August, delivery following) Initial development of course materials for "Web Technology for Virtual Environments, FMS, and IMT " to be completed and offered in Year 2 with 3 days, 2 instructors, workshop format Comments on NPAC and FMS Fox led JPL-Caltech group that developed first large scale parallel Battle Management simulation for US AirForce (General Paul was technical monitor) SIM 87,88,89 was hypercube implementation of tracking and response to SDI scenario This work also involved the early Time Warp activities at JPL We are currently finishing up web based command and control system (for Rome Laboratory) which includes simulated and real participants implemented as simple event driven simulation Java engine. Recently NPAC has been leader in use of Java and VRML for (scientific) computing Web Technology to Integrate PCRC Modules We suggest a hierarchical computing model with high functionality modest performance integration/coordination layer controlling classic low level HPCC high performance layer At Lowest layer hardest problem is decomposition At Top (computing) layer, flexible Integration is problem as decomposition natural User Interface gives third level Web Technology to Integrate HPCC Modules We show a hierarchical computing model with high functionality modest performance integration/coordination layer controlling classic low level HPCC high performance layer At Lowest layer hardest problem is decomposition At Middleware (computing) layer, flexible Integration is problem as decomposition natural Clients and User Interfaces gives third level linked by HTTP, Java Applets and as linked sessions through TANGO NPAC HPCC Related Activity Areas NSF Black Hole Grand Challenge CRPC -- Center for Research in Parallel Computation HPF, Education, Applications Some NCSA PACI activity in education and computing Arpa HPF/HPC++/"HPJava" Runtime (finishing) and performance estimation PetaFlops Initiative DoD Modernization Program -- CEWES(Vicksburg) ASC(Dayton) and ARL(Aberdeen) Web based Training (Collaboration, Video Servers etc.) (Parallel) Databases (Oracle, Illustra, JDBC) Support of Distributed Simulation (FMS) and Integrated Modelling and Testing(IMT) Through InfoMall, Monte Carlo based Financial modelling National High Performance Software Exchange HPCC Education and Training Computational Science Courses / Curricula in English and Chinese Both on CDROM's -- 1200 distributed in China include HPF and MPI definitions translated TangoWisdom linkage of Collaboration and Dissemination technology Java Applets as universal interface to HPCC simulations (High profile/risk) project to use electronic delivery of courses to Historically Black Colleges The Web allows democratization of education! webwisdom.org is virtual university and webwisdom.com will deliver technology including database backend, video servers, Tangowisdom HPCC Facilities at NPAC We have modest systems linked by quality commercial ATM and FDDI networks We are a research and development group and not a computer center! Small (12 node) SP2 and (8 processor) Power Challenge Windows NT Cluster for education and computing 8 SUN UltraServer2/model 1200 with OC3 ATM connection used in computing and database applications Other SUN fileservers and clients main client thrust is Windows machines