Full HTML for

Scripted foilset Status of HPCC in USA and HPCC based on Commodity Technologies

Given by Geoffrey C. Fox at Beijing and Chang Sha China on 28 Dec 97 to 5 Jan 98. Foils prepared 8 January 98
Outside Index Summary of Material

We describe some aspects of HPCC in USA and in particular the new NCSA computational alliance and the proposed petaflop initiative
We discuss role of commodity (Web) technologies in future high performance computing environments
We describe how a network of Web/CORBA/COM servers architecture can naturally support both parallel and distributed computing while
We describe applications to both metacomputing, and parallel computing
We suggest critical importance of CORBA and component based software in HPCC -- Javabeans seem very important
We describe role of collaboration technology in linking computers with people
We describe use of Java as a general coding language for scientific and engineering computation
This approach unifies distributed event driven simulations with classic massively parallel time stepped computations

Table of Contents for full HTML of Status of HPCC in USA and HPCC based on Commodity Technologies

 Denote Foils where Image Critical
Denote Foils where HTML is sufficient
denotes presence of Additional linked information which is lightpurpleed out if missing
1 Status of HPCC in USA and High Performance Computing (HPCC) based on Commodity Technologies December 27,97 -- January 6,98 Beijing and Chang Sha (Hunan Province) China
2 Abstract of Commodity Technologies in HPCC for China
3 Some International HPCC Activities
4 Some Concepts Learnt from HPCC Initiative
5 New Initiatives of Current HPCC
6 Some More Detailed Trends in HPCC
7 National Science Foundation's Partnerships for Advanced Computational Infrastructure
8 Ten More Years!
9 The Alliance National Technology Grid
10 vBNS, Internet-2 and PACI
11 The Alliance is Prototyping the National Technology Grid
12 Launching the Alliance - What We Will Do This Fall
13 Alliance Enabling Technologies Teams - Faculty Leads
14 Alliance Applications Technologies Teams - Faculty Leads and UIUC (NCSA) Anchors
15 Alliance Applications Technologies Teams - Faculty Leads and UIUC (NCSA) Anchors
16 Education, Outreach, and Training FY98 Projects
17 How Did a Supercomputer Center Get Into Enterprise Knowledge Management?
18 NCSA Supplies Alliance With Information Technology Support
19 NCSA Intranet Deployment
20 Alliance Collaboration Spaces
21 NCSA Habanero - Collaborative Support for Scientific Teams
22 The NCSA Biology Workbench - A Web-Based Software Architecture
23 Linking K-12 With Real-Time Science
24 BIMA Distributed Observatory, Digital Library, and Collaboratory
25 Exponential Growth in National User Demand Will Drive Grid Capabilities
26 Replacement of Shared Memory Vector Supercomputers by Microprocessor SMPs
27 NCSA: A Decade of Leadership in Clustered Computing
28 Regional Partners to Partners for Advanced Computational Services
29 Visual Supercomputing Goal: Make Analysis as Powerful as Simulation
30 CAVE Virtual Immersive Environment
31 Building the Visual Supercomputer
32 Peak Supercomputer Performance
33 Overall Remarks on the March to PetaFlops - I
34 Overall Remarks on the March to PetaFlops - II
35 10 Possible PetaFlop Applications
36 Chip Density Projections to year 2013
37 Clock Speed and I/O Speed in megabytes/sec per pin through year 2013
38 II. Major Findings & Recommendations: Findings.
39 II. Major Findings & Recommendations: Findings.
40 II. Major Findings & Recommendations: Recommendations
41 II. Major Findings & Recommendations: Recommendations
42 Supercomputer Architectures in Years 2005-2010 -- I
43 Supercomputer Architectures in Years 2005-2010 -- II
44 Supercomputer Architectures in Years 2005-2010 -- III
45 Performance Per Transistor
46 New "Strawman" PIM Processing Node Macro
47 "Strawman" Chip Floorplan
48 Comparison of Supercomputer Architectures
49 General Philosophy from PetaSoft Meeting
50 Features of the The Layered Software Model
51 Hierarchy of Software Levels
52 Isn't the Web hardware and software too slow to be interesting for HPCC? -Java- II
53 PetaSoft Findings 1) and 2) -- Memory Hierarchy
54 PetaSoft Findings 3) and 4) -- Using Memory Hierarchy
55 PetaSoft Findings 5) and 6) -- Layered Software
56 PetaSoft Recommendations 1) to 3) Memory and Software Hierarchy
57 V. A National program concept: Technology Model
58 Now we follow with A Comparison of JNAC and HPCC
59 Comparison of HPCC and JNAC - I
60 Comparison of HPCC and JNAC - II
61 Comparison of HPCC and JNAC - III
62 The Computing Pyramid
63 Pragmatic Object Web Technology Model
64 A Web-based 3-Tier Computing System
65 High Functionality Software Layer
66 Architecture of HPcc Commidity Technology High Performance Computing System
67 Three Roles of Object Web Technologies in Computing
68 Glossary of Terms VI
69 Computational Grid and the Object Web
70 Object Web Software is the Best
71 Synergy of InterNet and IntraNets
72 Structure(Architecture) of Applications - I
73 Structure(Architecture) of Applications - II
74 Metacomputing with Web Architecture
75 One Strategy for a Object Web-based Metacomputing
76 Web-Server based Metacomputer Capabilities at 3 levels
77 General Object Web based Middle Tier Server Architecture
78 WebFlow Globus and FrontEnd (DARP,SciVis) Architecture II
79 Planned Architecture of DARP User Level Debugging and Rapid Prototyping System
80 Proposed Approach to High Performance Messaging
81 Three Possible Implementations of CFD CSM Linkage
82 Picture of JavaBean and JDK1.1 AWT Event Model
83 Some Capabilities of the Object Web (Server) Architecture for Computing
84 Example of WebFlow = AVS/Khoros using Web
85 WebFlow: Image Processing
86 WebFlow Globus and FrontEnd (DARP,SciVis) Architecture II
87 Web Architecture Supports Interpreted Environments
88 Planned Architecture of DARP User Level Debugging and Rapid Prototyping System
89 Glossary of Terms II
90 Glossary of Terms IV
91 Component Based Programming Environments
92 What are JavaBeans I
93 What are JavaBeans II
94 What is a Module?
95 3 by 3 Diagram of Programming Environments versus System Complexity from PC to HPCC
96 HPCC ComponentWare: Essential Ideas
97 Pure CORBA Architecture for a distributed Information System (There are similar COM and Javabean /RMI Versions)
98 CORBA Software Model
99 A Parallel Computer Viewed as a Single CORBA Object
100 Each Node of a Parallel Computer viewed as a Separate CORBA Object
101 Fig. 13: Each node of a parallel computer instantiated as a Corba object.
102 A Message or Protocol Optimization Bridge
103 Fig 15: A message optimization bridge
104 Integration of DIS with Object Web Based Computing
105 Technology Convergence Roadmap - Overview for Forces Modeling, Integrated Modeling and Testing for DoD
106 DoD M&S Strategy: An Analogy to City Planning
107 How Did We Get Here?
108 Some Terminology
109 Some More Terminology
110 HLA Comprises Three Components: Rules, Runtime Infrastructure, Templates
111 A Federation Must Play by the Rules
112 Each Federate Must Play By the Rules
113 OMT Components
114 Object Class Structure Table (Template)
115 Object Interaction Table (Roughly Methods in CORBA)
116 Architecture Splits Functions Between Simulations and Runtime Infrastructure
117 Tango Collaboration System
118 Tango Screen: Talking Heads and White Board
119 Typical Web Collaboration Architecture
120 The TANGOsim C2 Application
121 Command and Control Screen with Multimedia Message
122 A demo of animated objects controlled by simulation engine of Tango collaborative system - III
123 TANGO Structure of Multidisciplinary Applications
124 Minimal Web based Multidisciplinary Application
125 Comparison of Communication/Linkage Models
126 Java as the Language for Computational Kernels!
127 Some Critical Features of Java as a Programming Language
128 Comparison of Java and Fortran 77/90
129 Java Links the Bottom and Top of Pyramid
130 Isn't the Web hardware and software too slow to be interesting for HPCC? -Java- I
131 Performance of Java is Dreadful!
132 LinPack Java Performance Updated to Sept 30 1996
133 LinPack Java Performance Updated to June 3 1997
134 What Limits Performance of Compiled Java?
135 Isn't the Web hardware and software too slow to be interesting for HPCC? -Java- III
136 Classes of Simulations and their High Performance Needs
137 Java and Parallelism?
138 "Pure" Java Model For Parallelism
139 Mechanisms for Data Parallelism in HPJava

Outside Index Summary of Material

HTML version of Scripted Foils prepared 8 January 98

Foil 1 Status of HPCC in USA and High Performance Computing (HPCC) based on Commodity Technologies December 27,97 -- January 6,98 Beijing and Chang Sha (Hunan Province) China

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Geoffrey Fox
Syracuse University
NPAC
111 College Place Syracuse NY 13244 4100
3154432163
HTML version of Scripted Foils prepared 8 January 98

Foil 2 Abstract of Commodity Technologies in HPCC for China

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
We describe some aspects of HPCC in USA and in particular the new NCSA computational alliance and the proposed petaflop initiative
We discuss role of commodity (Web) technologies in future high performance computing environments
We describe how a network of Web/CORBA/COM servers architecture can naturally support both parallel and distributed computing while
We describe applications to both metacomputing, and parallel computing
We suggest critical importance of CORBA and component based software in HPCC -- Javabeans seem very important
We describe role of collaboration technology in linking computers with people
We describe use of Java as a general coding language for scientific and engineering computation
This approach unifies distributed event driven simulations with classic massively parallel time stepped computations
HTML version of Scripted Foils prepared 8 January 98

Foil 3 Some International HPCC Activities

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
There are national HPCC programs in:
  • USA
  • China
  • Europe
  • Japan
The USA activities include
  • NCSA and NPACI NSF Centers
  • DoE ASCI Program
  • DoD Modernization Program involving 4 major centers and 11 application areas (CTA's)
    • Dayton (Wright Patterson Airforce Base), Aberdeen(ARL), CEWES (Vicksburg), NAVO (Stennis space Center in Mississippi)
HTML version of Scripted Foils prepared 8 January 98

Foil 4 Some Concepts Learnt from HPCC Initiative

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
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
HTML version of Scripted Foils prepared 8 January 98

Foil 5 New Initiatives of Current HPCC

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Some of current new developments focus on
  • "Data Intensive" applications which merge data processing (from instruments) and simulation
  • Distributed Computing or Metacomputing
    • Geographically distributed computers
  • Collaboration
    • Linking Computers and People together
  • Multidisciplinary Problems
    • Link several Codes and Disciplines together to solve a large problem with several components
HTML version of Scripted Foils prepared 8 January 98

Foil 6 Some More Detailed Trends in HPCC

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Use of Commodity hardware: PC's offer best performance per dollar (Gigaflop for $30,000)
Use of Commodity software: Windows NT, COM, CORBA, web, Java, VRML ....
Use of Web to produce Seamless (universal) computer interfaces
Java replacing C++ and Fortran for Numerical Computation
Use of databases and collaboration technology to link people, databases and simulation
Integration of parallel and distributed computing
Use of distributed objects (CORBA) to encapsulated remote services
HTML version of Scripted Foils prepared 8 January 98

Foil 7 National Science Foundation's Partnerships for Advanced Computational Infrastructure

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
HTML version of Scripted Foils prepared 8 January 98

Foil 8 Ten More Years!

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
National Science Board Decision-March 28
  • NCSA and NPACI proposals funded
  • Starts October 1
Funded for 10 years
  • Renewable at 5 years
  • Yearly progress reviews
Major Increase in Support Levels
  • NSF
  • State
  • Industry
HTML version of Scripted Foils prepared 8 January 98

Foil 9 The Alliance National Technology Grid

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
HTML version of Scripted Foils prepared 8 January 98

Foil 10 vBNS, Internet-2 and PACI

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
NPACI
NCSA Alliance
Both NCSA Alliance and NPACI
Other sites
64 sites approved, 63 more proposed 7/97
NCSA
vBNS: very high speed Backbone Network Service
HTML version of Scripted Foils prepared 8 January 98

Foil 11 The Alliance is Prototyping the National Technology Grid

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Leading Edge Centers
  • Supernodes of the Grid
Enabling Technology Teams
  • Architects of the Grid
Applications Technologies Teams
  • Specifications for the Grid
Education, Outreach, and Training Teams
  • Content for the Grid
Partners for Advanced Computational Services
  • Support for the Grid
Industrial Partners and Strategic Vendors
  • Technology Transfer for the Grid
HTML version of Scripted Foils prepared 8 January 98

Foil 12 Launching the Alliance - What We Will Do This Fall

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Infrastructure Foundations
  • Build the Alliance Intranet Framework
    • Create the Origin Repository
    • Rough Out Intranet Interfaces
  • Set up the High Performance Networks
    • Link the CAVE Devices
Define the Alliance Work Plan
  • Timelines and Deliverables for First Year
    • Interweaving of ET/AT/EOT/RP and NCSA
    • Identifiy Early Success Goals
  • Define Alliance Software Set for AT/EOT
    • Review Alliance Collaborative Software Plans
    • Requires Strong ET/AT/EOT Brainstorming
HTML version of Scripted Foils prepared 8 January 98

Foil 13 Alliance Enabling Technologies Teams - Faculty Leads

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Parallel Computing (16)
  • Ken Kennedy, Rice U
  • Greg McRae, MIT
Distributed Computing (15)
  • Rick Stevens, Argonne
  • Paul Woodward, U Minnesota
Data and Collaboration (14)
  • Dan Reed, UIUC
  • Roscoe Giles, Boston U
HTML version of Scripted Foils prepared 8 January 98

Foil 14 Alliance Applications Technologies Teams - Faculty Leads and UIUC (NCSA) Anchors

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Cosmology (5)
  • Jeremiah Ostriker, Princeton U
  • Mike Norman, UIUC (NCSA)
Environment Hydrology (11)
  • John Anderson, U Wisconsin
  • Robert Wilhelmson, UIUC (NCSA)
  • V.C. Patel, U Iowa
  • Doug Johnston, UIUC (NCSA)
Chemical Engineering (7)
  • Greg McRae, MIT
  • Richard Braatz, UIUC (NCSA)
HTML version of Scripted Foils prepared 8 January 98

Foil 15 Alliance Applications Technologies Teams - Faculty Leads and UIUC (NCSA) Anchors

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Bioinformatics (9)
  • Sangtae Kim, U Wisconsin
  • Shankar Subramaniam, UIUC (NCSA)
Nanomaterials (11)
  • John Wilkins, OSU
  • David Ceperley, UIUC (NCSA)
  • Robert Dutton, Stanford U
  • Karl Hess, UIUC (NCSA)
Scientific Instruments (8)
  • David Agard, UCSF
  • Clint Potter, UIUC (NCSA)
  • Paul Vanden Bout, NRAO
  • Richard Crutcher, UIUC (NCSA)
HTML version of Scripted Foils prepared 8 January 98

Foil 16 Education, Outreach, and Training FY98 Projects

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Enabling Technologies
Applications Technologies
EOT-PACI Project Activities
National Education Community
Digital Libraries
Collaboration Tech.
Distance Learning
VR/Simulations
Molecular Biology
Cosmology
Scientific Instruments
Chickscope
Biology Workbench
WW2010
Chemistry Visualization
HTML version of Scripted Foils prepared 8 January 98

Foil 17 How Did a Supercomputer Center Get Into Enterprise Knowledge Management?

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Microprocessor Market Convergence
  • NT/Intel Challenging UNIX/RISC
Desktop Software Tool Development
  • From NCSA Telnet to NCSA Mosaic
Large Scientific Databases
  • From Scientific Visualization to Info Viz
Emergence of Distributed Object Architecture
  • Java, ActiveX, CORBA, and the Web
Supercomputer Center to PACI
  • Need to Create Alliance Intranet and Collaboratory
HTML version of Scripted Foils prepared 8 January 98

Foil 18 NCSA Supplies Alliance With Information Technology Support

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Alliance Information Management
  • Production intranet, management databases
Information Analysis
  • Knowledge discovery, machine learning, info.visualization
NT Clusters
  • NCSA Symbio
Technical Collaboration Environments
  • Habanero, ISAAC, EMERGE
Scientific Data Management
  • HDF
HTML version of Scripted Foils prepared 8 January 98

Foil 19 NCSA Intranet Deployment

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Strategy
  • Open standards
  • Commercial technology foundation
  • Strategic alliance with major software vendors
  • Interfaces to experimental and development efforts
Implementation
  • Lotus Notes / Domino
  • Web browser as sole client side user interface
  • Java
  • SQL accessible relational databases
HTML version of Scripted Foils prepared 8 January 98

Foil 20 Alliance Collaboration Spaces

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Asynchronous
  • Participants Separated in Time
  • Federated Document Repositories
  • Threaded Discussions
  • Examples: Usenet News, Lotus Notes
Synchronous
  • Participate in Real Time
  • Share Screen Views and Applications
  • Video and/or Audio Connections
  • Examples: Netmeeting, Habanero, Tango
HTML version of Scripted Foils prepared 8 January 98

Foil 21 NCSA Habanero - Collaborative Support for Scientific Teams

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Image Courtesy of Ken Bishop, U Kansas and NCSA
HTML version of Scripted Foils prepared 8 January 98

Foil 22 The NCSA Biology Workbench - A Web-Based Software Architecture

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
HTML version of Scripted Foils prepared 8 January 98

Foil 23 Linking K-12 With Real-Time Science

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
http://vizlab.beckman.uiuc.edu/chickscope/
HTML version of Scripted Foils prepared 8 January 98

Foil 24 BIMA Distributed Observatory, Digital Library, and Collaboratory

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
BIMA, Courtesy Richard Crutcher, UIUC
http://bima-server.ncsa.uiuc.edu/imagelib/VRMLHighlights.html
HTML version of Scripted Foils prepared 8 January 98

Foil 25 Exponential Growth in National User Demand Will Drive Grid Capabilities

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Source: Quantum Research Database
HTML version of Scripted Foils prepared 8 January 98

Foil 26 Replacement of Shared Memory Vector Supercomputers by Microprocessor SMPs

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
HTML version of Scripted Foils prepared 8 January 98

Foil 27 NCSA: A Decade of Leadership in Clustered Computing

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Distributed Memory
Distributed Shared Memory
IBM RS6000
6 x 1
HP 735
12 x 1
SGI Power
Challenge Array
(10x16)
SGI Origin-2000
8 x 32
HP/Convex SPP-1200
4 x 16
HP/Convex
SPP-2000
8 x 16
HP PP cluster (8 x 2), classroom (20 x 1)
Proposed NT Cluster
64 x 4
Proposed
SGI Origin-2000
8 x 128
1990
1992
1994
1996
1998
HTML version of Scripted Foils prepared 8 January 98

Foil 28 Regional Partners to Partners for Advanced Computational Services

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
PACS Focus Area
  • Regional Access to Workshops
  • Distributed Training and User Services
  • Mid-Level Computational Resources
  • Specialized Technology Development Sites
PACS Members
  • 4 Mid-Level Centers
    • BU (Origin), OSC (Triton, T3E), Kentucky (HP SPP), Maui (SP-2)
  • 6 Tech Development Sites
    • Minn, Wisc, Rice, ANL, Wash, UVa
  • 3 Outreach Consortia (CIC, SURA, EPSCoR)
HTML version of Scripted Foils prepared 8 January 98

Foil 29 Visual Supercomputing Goal: Make Analysis as Powerful as Simulation

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Colliding Galaxies (Smithsonian IMAX)-Donna Cox, Bob Patterson, NCSA-From "Cosmic Voyage"-Nominated for Academy Award 1997
Virtual Director in CAVE
1000 Hour SDSC Supercomputer Run to Generate Data
Tens of Thousands of Hours of NCSA SGI Time to Render Data
Cross-Country Transfer to IMAX Film of Massive Amounts of Data
HTML version of Scripted Foils prepared 8 January 98

Foil 30 CAVE Virtual Immersive Environment

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
HTML version of Scripted Foils prepared 8 January 98

Foil 31 Building the Visual Supercomputer

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Rotating Turbulent Gas Ball Model of the Sun
Nine Day Run on NCSA Origin (128-processors)
Generated 2 Terabytes of Data, LCSE Visualized in 3 Days
Dave Porter, Paul Woodward, et al., LCSE, Univ of Minnesota, June 1997
HTML version of Scripted Foils prepared 8 January 98

Foil 32 Peak Supercomputer Performance

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
For "Convential" MPP/Distributed Shared Memory Architecture
Now(1996) Peak is 0.1 to 0.2 Teraflops in Production Centers
  • Note both SGI and IBM are changing architectures:
  • IBM Distributed Memory to Distributed Shared Memory
  • SGI Shared Memory to Distributed Shared Memory
In 1999, one will see production 1 Teraflop systems
In 2003, one will see production 10 Teraflop Systems
In 2007, one will see production 50-100 Teraflop Systems
Memory is Roughly 0.25 to 1 Terabyte per 1 Teraflop
If you are lucky/work hard: Realized performance is 30% of Peak
HTML version of Scripted Foils prepared 8 January 98

Foil 33 Overall Remarks on the March to PetaFlops - I

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
I find study interesting not only in its result but also in its methodology of several intense workshops combined with general discussions at national conferences
Exotic technologies such as "DNA Computing" and Quantum Computing do not seem relevant on this timescale
Note clock speeds will NOT improve much in the future but density of chips will continue to improve at roughly the current exponential rate over next 10-20 years
Superconducting technology is currently seriously limited by no appropriate memory technology that matches factor of 100-1000 faster CPU processing
Current project views software as perhaps the hardest problem
HTML version of Scripted Foils prepared 8 January 98

Foil 34 Overall Remarks on the March to PetaFlops - II

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
All proposed designs have VERY deep memory hierarchies which are a challenge to algorithms, compilers and even communication subsystems
Major need for hig-end performance computers comes from government (both civilian and military) applications
  • DoE ASCI (study of aging of nuclear weopens) and Weather/Climate prediction are two examples
Government must develop systems using commercial suppliers but NOT relying on traditionasl industry applications to motivate
So Currently Petaflop initiative is thought of as an applied development project whereas HPCC was mainly a research endeavour
HTML version of Scripted Foils prepared 8 January 98

Foil 35 10 Possible PetaFlop Applications

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
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
HTML version of Scripted Foils prepared 8 January 98

Foil 36 Chip Density Projections to year 2013

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Extrapolated from SIA Projections to year 2007 -- See Chapter 6 of Petaflops Report -- July 94
HTML version of Scripted Foils prepared 8 January 98

Foil 37 Clock Speed and I/O Speed in megabytes/sec per pin through year 2013

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Extrapolated from SIA Projections to year 2007 -- See Chapter 6 of Petaflops Report -- July 94
HTML version of Scripted Foils prepared 8 January 98

Foil 38 II. Major Findings & Recommendations: Findings.

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
PetaFLOPS possible; accelerate goals to 10 years.
Many important application drivers exist.
Memory dominant implementation factor.
Cost, power & efficiency dominate.
Innovation critical, new technology necessary.
Layered SW architecture mandatory.
Opportunities for immediate SW effort.
HTML version of Scripted Foils prepared 8 January 98

Foil 39 II. Major Findings & Recommendations: Findings.

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
New technology means paradigm shift.
  • Superconductivity technology is example.
Memory bandwidth.
Latency.
Software important.
Closer relationship between architecture and programming is needed.
Role of algorithms must improve.
HTML version of Scripted Foils prepared 8 January 98

Foil 40 II. Major Findings & Recommendations: Recommendations

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Conduct point design studies.
  • in hardware and software.
  • of promising architecture
Develop engineering prototypes.
  • Multiple technology track demonstrations
Start SW now, independent of HW.
Develop layered software architecture for scalability and code reuse
Explore algorithms for special purpose & reconfigurable structures.
HTML version of Scripted Foils prepared 8 January 98

Foil 41 II. Major Findings & Recommendations: Recommendations

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Support & accelerate R&D in paradigm shift technologies:
  • Superconductor RSFQ
  • Holographic photo-refractive storage
  • Optical guided and free-space interconnect
  • New semiconductor materials.
Perform detailed applications studies at scale.
Develop petaFLOPS scale latency management.
HTML version of Scripted Foils prepared 8 January 98

Foil 42 Supercomputer Architectures in Years 2005-2010 -- I

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Conventional (Distributed Shared Memory) Silcon
  • Clock Speed 1GHz
  • 4 eight way parallel Complex RISC nodes per chip
  • 4000 Processing chips gives over 100 tera(fl)ops
  • 8000 2 Gigabyte DRAM gives 16 Terabytes Memory
Note Memory per Flop is much less than one to one
Natural scaling says time steps decrease at same rate as spatial intervals and so memory needed goes like (FLOPS in Gigaflops)**.75
  • If One Gigaflop requires One Gigabyte of memory (Or is it one Teraflop that needs one Terabyte?)
HTML version of Scripted Foils prepared 8 January 98

Foil 43 Supercomputer Architectures in Years 2005-2010 -- II

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Superconducting Technology is promising but can it compete with silicon juggernaut?
Should be able to build a simple 200 Ghz Superconducting CPU with modest superconducting caches (around 32 Kilobytes)
Must use same DRAM technology as for silicon CPU ?
So tremendous challenge to build latency tolerant algorithms (as over a factor of 100 difference in CPU and memory speed) but advantage of factor 30-100 less parallelism needed
HTML version of Scripted Foils prepared 8 January 98

Foil 44 Supercomputer Architectures in Years 2005-2010 -- III

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Processor in Memory (PIM) Architecture is follow on to J machine (MIT) Execube (IBM -- Peter Kogge) Mosaic (Seitz)
  • More Interesting in 2007 as processors are be "real" and have nontrivial amount of memory
  • Naturally fetch a complete row (column) of memory at each access - perhaps 1024 bits
One could take in year 2007 each two gigabyte memory chip and alternatively build as a mosaic of
  • One Gigabyte of Memory
  • 1000 250,000 transistor simple CPU's running at 1 Gigaflop each and each with one megabyte of on chip memory
12000 chips (Same amount of Silicon as in first design but perhaps more power) gives:
  • 12 Terabytes of Memory
  • 12 Petaflops performance
  • This design "extrapolates" specialized DSP's , the GRAPE (specialized teraflop N body machine) etc to a "somewhat specialized" system with a general CPU but a special memory poor architecture with particular 2/3D layout
HTML version of Scripted Foils prepared 8 January 98

Foil 45 Performance Per Transistor

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Performance data from uP vendors
Transistor count excludes on-chip caches
Performance normalized by clock rate
Conclusion: Simplest is best! (250K Transistor CPU)
Millions of Transistors (CPU)
Millions of Transistors (CPU)
Normalized SPECINTS
Normalized SPECFLTS
HTML version of Scripted Foils prepared 8 January 98

Foil 46 New "Strawman" PIM Processing Node Macro

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
HTML version of Scripted Foils prepared 8 January 98

Foil 47 "Strawman" Chip Floorplan

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
HTML version of Scripted Foils prepared 8 January 98

Foil 48 Comparison of Supercomputer Architectures

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Fixing 10-20 Terabytes of Memory, we can get
16000 way parallel natural evolution of today's machines with various architectures from distributed shared memory to clustered heirarchy
  • Peak Performance is 150 Teraflops with memory systems like today but worse with more levels of cache
5000 way parallel Superconducting system with 1 Petaflop performance but terrible imbalance between CPU and memory speeds
12 million way parallel PIM system with 12 petaflop performance and "distributed memory architecture" as off chip access with have serious penalities
There are many hybrid and intermediate choices -- these are extreme examples of "pure" architectures
HTML version of Scripted Foils prepared 8 January 98

Foil 49 General Philosophy from PetaSoft Meeting

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Define a "clean" model for machine architecture
  • Memory hierarchy including caches and geomterical (distributed) effects
Define a low level "Program Execution Model" (PEM) which allows one to describe movement of information and computation in the machine
  • This can be thought of as "MPI"/assembly language of the machine
On top of low level PEM, one can build an hierarchical (layered) software model
  • At the top of this layered software model, one finds objects or Problem Solving Environments (PSE's)
  • At an intermediate level there is Parallel C C++ or Fortran
One can program at each layer of the software and augment it by "escaping" to a lower level to improve performance
  • Directives (HPF assertions) and explicit insertion of lower level code (HPF extrinsics) are possible
HTML version of Scripted Foils prepared 8 January 98

Foil 50 Features of the The Layered Software Model

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
This is not really a simple stack but a set of complex relations between layers with many interfaces and modules
Interfaces are critical ( for composition across layers)
  • Enable control and performance for application scientists
  • Decouple CS system issues and allow exploration and innovation
Enable Next
10000
Users
For First 100
Pioneer Users
Higher Level abstractions
nearer to
application domain
Increasing Machine
detail, control
and management
HTML version of Scripted Foils prepared 8 January 98

Foil 51 Hierarchy of Software Levels

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Numerical Objects in (C++/Fortran/C/Java)
Expose the Coarse Grain Parallelism
Expose All Levels of Memory Hierarchy
a) Pure Script (Interpreted)
c) High Level Language but Optimized Compilation
d) Machine Optimized RunTime
b) Semi- Interpreted
a la Applets
Memory Levels in High
Performance CPU
Nodes of Parallel/ Distributed System
HTML version of Scripted Foils prepared 8 January 98

Foil 52 Isn't the Web hardware and software too slow to be interesting for HPCC? -Java- II

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Applications requires a range of capabilities in any language
High level ("Problem Solving Environment") manipulating"large" objects
  • Semi Interpreted (Java Applet) or Interpreted (Improved JavaScript)
Intermediate level Compiled Code targetted at "sequential" (multi-threaded) architecture
  • Existing Native Compiled Java using Simple types (arrays) for numerically intensive parts
  • Note as no pointers and no overloading of basic operators, Java code should be very efficient
Lower level runtime exploiting parallelism and memory hierarchies
  • "Hints" from higher level languages (in HPF style?) referencing highly functional efficient runtime optimized for high performance architectures
  • Requires extensions to both message passing and data parallel interfaces for whatever language one uses
HTML version of Scripted Foils prepared 8 January 98

Foil 53 PetaSoft Findings 1) and 2) -- Memory Hierarchy

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
1)Deep Memory Hierarchies present New Challenges to High performance Implementation of programs
  • Latency
  • Bandwidth
  • Capacity
2)There are two dimensions of memory hierarchy management
  • Geometric or Global Structure
  • Local (cache) hierachies seen from thread or processor centric view
HTML version of Scripted Foils prepared 8 January 98

Foil 54 PetaSoft Findings 3) and 4) -- Using Memory Hierarchy

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
3)One needs a machine "mode" which supports predictable and controllable memory system leading to communication and computation with same characteristics
  • Allow Compiler optimization
  • Allow Programmer control and optimization
  • For instance high performance would often need full program control of caches
4)One needs a low level software layer which provides direct control of the machine (memory hierarchy etc.) by a user program
  • This for initial users and program tuning
HTML version of Scripted Foils prepared 8 January 98

Foil 55 PetaSoft Findings 5) and 6) -- Layered Software

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
5)One needs a layered (hierarchical) software model which supports an efficient use of multiple levels of abstraction in a single program.
  • Higher levels of Programming model hide extraneous complexity
  • highest layers are application dependent Problem Solving Environments and lower levels are machine dependent
  • Lower levels can be accessed for additional performance
  • e.g. HPF Extrinsics. Gcc ASM, MATLAB Fortran Routines, Native classes in Java
6)One needs a set of software tools which match the layered software (programming model)
  • Debuggers, Performance and load balancing tools
HTML version of Scripted Foils prepared 8 January 98

Foil 56 PetaSoft Recommendations 1) to 3) Memory and Software Hierarchy

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
1)Explore issues in design of petaComputer machine models which will support the controllable hierarchical memory systems in a range of important architectures
  • Research and development in areas of findings 3) and 4)
2)Explore techniques for control of memory hierarchy for petaComputer architectures
  • Use testbeds
3)Explore issues in designing layered software architectures -- particularly efficient mapping and efficient interfaces to lower levels
  • Use context of petaflop applications and machines
  • e.g. HPF is a possible layer while HPF Extrinsics is an interface to a lower (MPI) layer
HTML version of Scripted Foils prepared 8 January 98

Foil 57 V. A National program concept: Technology Model

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
HTML version of Scripted Foils prepared 8 January 98

Foil 58 Now we follow with A Comparison of JNAC and HPCC

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Next Three foils isolate some differences and commonalities in two programs
HTML version of Scripted Foils prepared 8 January 98

Foil 59 Comparison of HPCC and JNAC - I

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Both set a hardware goal (teraflop for HPCC and petaflop for JNAC) to focus activity but in each case systems and applications were main justification
Both couple applications with software and architectures in multidisciplinary teams with multi-agency support
HPCC was dominantly research
  • JNAC is roughly 50-50 research and development
HPCC inevitably developed MPP's and transferred parallel computing to computing mainstream
  • JNAC's challenge is memory hierarchy and will transfer understanding of this to mainstream independent of parallelism
HTML version of Scripted Foils prepared 8 January 98

Foil 60 Comparison of HPCC and JNAC - II

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
HPCC aimed at Grand challenges in Industry Government and Academia
  • JNAC aimed at government (including NSF) mission critical applications
HPCC developed software (PSE's) largely independently in each Grand Challenge
  • JNAC will link software efforts to a few PSE's and a common set of JNAC Interfaces
HPCC tended to develop hardware with rapidly changing architectures which software "chased" rather laboriously
  • JNAC develops software simultaneously with hardware and to a uniform common architecture allowing better re-use of both application and systems software
HTML version of Scripted Foils prepared 8 January 98

Foil 61 Comparison of HPCC and JNAC - III

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
HPCC aimed to transfer technology to Industry for commercialization
  • JNAC relies on industry to build systems designed by laboratory, university and industry consortia
HPCC is Research -->Capitalization-->Product
  • JNAC is mission driven development linked to supporting research with engineering prototypes as capitalization stage
HPCC was a broad program aimed at "all" (large scale) users of computers
  • JNAC is a focused program and aims at "top 100" power users
HTML version of Scripted Foils prepared 8 January 98

Foil 62 The Computing Pyramid

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Bottom of Pyramid has 1000 times dollar value and compute power of best supercomputer (tip of pyramid) but supercomputer has high performance network to support close synchronization needed by classic parallel algorithms
Use of
Web Technologies
is naturally a
HTML version of Scripted Foils prepared 8 January 98

Foil 63 Pragmatic Object Web Technology Model

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
The current incoherent but highly creative Web will merge with distributed object technology in a multi-tier client-server-service architecture with Java based combined Web-ORB's
COM(Microsoft) and CORBA(world) are competing cross platform and language object technologies
  • Javabeans plus RMI is 100% Java distributed object technology
Need to abstract entities (Web Pages, simulations) and services as objects with methods(interfaces)
How do we do this while infrastructure still being designed!
One can anticipate this by building systems in terms of Javabeans e.g. develop Web-based databases with Javabeans using standard JDBC (Java Database Connectivity) interfaces
Design and Use Java Framework for Computing which will become a "CORBA facility"
  • Do not sacrifice significant performance!
HTML version of Scripted Foils prepared 8 January 98

Foil 64 A Web-based 3-Tier Computing System

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Middle Tier
Basic Web Server
Custom Web Server
TP Server
Business Transaction Management
You Write Software
at Client and Server
Old and New Useful Backend Software
HTML version of Scripted Foils prepared 8 January 98

Foil 65 High Functionality Software Layer

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
This is "middleware" which is implemented in simplest form as a network of Java Servers
  • Web based distributed or Metacomputing based on linking mobile Web modules
  • Collaboration as in Habanero or Tango
  • Support services including databases, object brokers, instruments, MPP's etc.
Access
Resources
Store
Multimedia Information
Collaboration Server
File Systems
and/or Database
Object Broker
Database
Simulation (Network-enabled
servers such as NEOS, Netsolve)
Sequential
or Parallel
Computer
HTML version of Scripted Foils prepared 8 January 98

Foil 66 Architecture of HPcc Commidity Technology High Performance Computing System

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
HTML version of Scripted Foils prepared 8 January 98

Foil 67 Three Roles of Object Web Technologies in Computing

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
1)One can "just" use Object Web technologies as a software infrastructure for building parallel, distributed or sequential computing environments which can have a very different architecture from the Web
  • e.g. write software in Java
  • e.g. adopt interfaces/standards such as JDBC, VRML, Java3D
2)Harness the power of the Web as a computer -- use up the idle cycles on the WebTV's in every home -- typically a Web Client based system
  • We will not discuss this here
3)One can view the Object Web as a distributed information system with modest performance and build a metacomputing system with the Web architecture
  • we "only" need to enhance the Object Web to get high performance
  • several ways of doing this -- common feature is that they all inherit rich web capabilities -- typically Web or Object Server based
  • By inheriting Object Web architecture, naturally track evolving and improving commodity technology base
HTML version of Scripted Foils prepared 8 January 98

Foil 68 Glossary of Terms VI

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
HPCC (High Performance Computing and Communication)
  • Originally a formal federal initiative but even after this ended in 1996, this term is used to describe the field devoted to solving large-scale problems with powerful computers and networks.
Computational Grid
  • A recent term used by the HPCC community to describe large scale distributed computing which draws on analogies with electrical power grids as enabling a revolution
HPcc (High Performance commodity computing)
  • NPAC project to develop a commodity computing based high performance computing software environment. Note that we have dropped "communications" referred to in the classic HPCC acronym. This is not because it is unimportant but rather because a commodity approach to high performance networking is already being adopted. We focus on high level services such as programming, data access and visualization that we abstract to the rather wishy-washy "computing" in the HPcc acronym.
HTML version of Scripted Foils prepared 8 January 98

Foil 69 Computational Grid and the Object Web

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Larry Smarr and NCSA Collaboration have stressed analogy of deployment of computer/communication technology with impact that electrical and transportation grids had
  • I.e. they enabled revolutions in society
  • What is relative role of High-end and Commodity end in Grid?
The transportation system was built using lessons from and feed up/down from Sports cars, Cadillacs, Model T's, Ford Escorts etc.
Computational Grid will be shaped by and shape all 5 classes of applications on previous foil
  • Everybody would perhaps agree on general statement but perhaps disagree on importance of it
A highish end computational grid will in some sense (to be disagreed on) be influenced by and influence the "Object Web" which is here defined as "mass-market"/business IntraNet (low to low) use of Internet/distributed Information Systems
HTML version of Scripted Foils prepared 8 January 98

Foil 70 Object Web Software is the Best

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
By definition, Object Web software is and will even more so, be the "best" software ever built because it has the largest market and greatest leverage of investment dollars
  • Further most creative business model - harness the world's best minds together with open interfaces
  • Note previously PC software and IBM business OS were high quality software but not so open and not a complete model
On should build upwards from the "democratic Web"
  • e.g. up from POTS --> ISDN / Cable Modem --> ...
  • Not down from <---- ATM
This allows you to both deliver your application to the general public (when required) and leverage best software
HTML version of Scripted Foils prepared 8 January 98

Foil 71 Synergy of InterNet and IntraNets

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
HTML version of Scripted Foils prepared 8 January 98

Foil 72 Structure(Architecture) of Applications - I

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Applications are metaproblems with a mix of module and data parallelism
Modules are decomposed into parts (data parallelism) and composed hierarchically into full applications.They can be the
  • "10,000" separate programs (e.g. structures,CFD ..) used in design of aircraft
  • the various filters used in Khoros based image processing system
  • the ocean-atmosphere components in integrated climate simulation
  • The data-base or file system access of a data-intensive application
  • the objects in a distributed Forces Modeling Event Driven Simulation
HTML version of Scripted Foils prepared 8 January 98

Foil 73 Structure(Architecture) of Applications - II

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Modules are "natural" message-parallel components of problem and tend to have less stringent latency and bandwidth requirements than those needed to link data-parallel components
  • modules are what HPF needs task parallelism for
  • Often modules are naturally distributed whereas parts of data parallel decomposition may need to be kept on tightly coupled MPP
Assume that primary goal of metacomputing system is to add to existing parallel computing environments, a higher level supporting module parallelism
  • Now if one takes a large CFD problem and divides into a few components, those "coarse grain data-parallel components" can be supported by computational grid technology
  • so no clean division but coarse grain modules are general goal!
HTML version of Scripted Foils prepared 8 January 98

Foil 74 Metacomputing with Web Architecture

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
It is natural to base on either a network of Web Clients or Web Servers
  • Not clear if distinction (in capability) between web servers and clients will remain as clients are adding functionality and any PC can run a server as well as a client!
Web Client Models Include SuperWeb (Javelin) from UCSB and are well illustrated by the January 1997 hotwired article "Suck your Mips".
Greater functionality but less power and pervasiveness is a pure Web Server model as proposed by NPAC
  • Can either use in controlled (IntraNets or run a server on every node of your MPP) or uncontrolled (all the world wide web hosts) fashion
  • Uncontrolled mode has interesting economic implications and is "controversial" for security and network performance area
Note total compute power in all Web "clients" is about 100 times that in all Central Supercomputers
HTML version of Scripted Foils prepared 8 January 98

Foil 75 One Strategy for a Object Web-based Metacomputing

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Object Web Software provides a high functionality but modest performance distributed computing (Metacomputing) environment based on either Web (soon to be CORBA IIOP and HTTP/Java Socket) Servers or Clients
Here we will explore an architecture using servers for control as higher functionality than clients although currently less broadly deployed
  • Some "parallel" computing problems (e.g. embarrassingly parallel ones) can "immediately" use Web as a parallel engine with no performance enhancements
Object Web Only addresses Integration of already decomposed parts!
  • Inherit and extend decomposition (data-parallel) technology from HPCC (parallel compiler and library technology)
  • Compose / Integrate Modules with Web Technology
  • Inherit all the pervasive (not high performance) services from the Web
HTML version of Scripted Foils prepared 8 January 98

Foil 76 Web-Server based Metacomputer Capabilities at 3 levels

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
1:User View: Interoperable Web Interface accessing services through Java Compute Services Framework
2:Network of Java Servers provide distributed services with databases, compute engines, collaboratories, object brokers, instruments
  • All control at this level
  • Some coarse grain computing
Back end "Number Crunchers" linked either by communication at level 2 (slowish but easy) or at level 3 (high performance but more work)
Compute processes linked either to servers or together by MPI if parallel
Java Servers
HTML version of Scripted Foils prepared 8 January 98

Foil 77 General Object Web based Middle Tier Server Architecture

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
We have a set of Services hosted by Object Web Servers which form the middleware and accessed by clients
Groups of clients (electronic societies) are linked by Java server based collaboration systems such as TANGO or Habanero
Access
Resources
Store
Multimedia Information
Collaboration Server
File Systems
and/or Database
Object Broker
Database
Simulation
e.g. NEOS
Netsolve
Computer
Person2
Shared
WhiteBoard
Shared Client Appl
Person1
General User
HTML version of Scripted Foils prepared 8 January 98

Foil 78 WebFlow Globus and FrontEnd (DARP,SciVis) Architecture II

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
HTML version of Scripted Foils prepared 8 January 98

Foil 79 Planned Architecture of DARP User Level Debugging and Rapid Prototyping System

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
HTML version of Scripted Foils prepared 8 January 98

Foil 80 Proposed Approach to High Performance Messaging

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
As a first step, implement multi-module systems with each module linked via Java Servers
  • Modest performance (non - parallel) linkage with easy database, collaborative system etc. access
Where necessary "escape" down to classic HPCC technologies for data transport keeping control at server level
  • e.g. following foil illustrates this for a typical two module coupled CFD(fluids) and CSM(structures) problem
  • e.g. JDBC would access a parallel database in same way as sequential database at control level. However backend would implement "parallel query option" transparently to user.
This seems very convenient in JDK 1.1 "event model" which is mechanism used by Javabeans to communicate
  • this is illustrated in foil after next!
HTML version of Scripted Foils prepared 8 January 98

Foil 81 Three Possible Implementations of CFD CSM Linkage

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
1)Simple Server Approach 2)Classic HPCC Approach
3)Hybrid Approach with control at server and
data transfer at
HPCC level
HTML version of Scripted Foils prepared 8 January 98

Foil 82 Picture of JavaBean and JDK1.1 AWT Event Model

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
4)Invoke High Performance Message Transfer between Observers and Sources specified in Message Event
Server Tier
Data Source
Data Sink (Observers)
5)Actual Data Transfer
High Performance Tier
2)Prepare
Message Event in Source Control
1)Register Observers with Listener
HTML version of Scripted Foils prepared 8 January 98

Foil 83 Some Capabilities of the Object Web (Server) Architecture for Computing

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Here are some examples of using our approach where large scale industry investment in Web technology appears to add significant value to metacomputing systems built with Web architecture
  • We illustrate some of them in following foils
  • many also apply to parallel computing environments
Multidisciplinary and Computational Steering Applications
  • link people data computation with collaboratory server as part of system
Visual and Interpreted Programming Environments
  • MATLAB/AVS/ Khoros like Systems (coarse grain Software Integration)
  • Software Component based approaches
Technologies to get High Performance CORBA
Integration with Forces Modeling (Distributed Event driven Simulation)
Integration with Networked enabled servers such as NEOS and Netsolve
  • These are naturally implemented using CORBA as are
  • Software repositories (RIB from National High Performance Software Exchange)
HTML version of Scripted Foils prepared 8 January 98

Foil 84 Example of WebFlow = AVS/Khoros using Web

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Simulation
Basic Display
Image Filter
is another
module
Output Display after Filter
Runs as a
parallel
module
using
Java Server
host
HTML version of Scripted Foils prepared 8 January 98

Foil 85 WebFlow: Image Processing

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Bunch of Filters and Displays
defined in
Java Graph editor and
running on grid of Java Servers
Original Image
HTML version of Scripted Foils prepared 8 January 98

Foil 86 WebFlow Globus and FrontEnd (DARP,SciVis) Architecture II

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
HTML version of Scripted Foils prepared 8 January 98

Foil 87 Web Architecture Supports Interpreted Environments

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Note Java also integrates compiled and interpreted approaches and so leads to more convenient programming environments
  • Develop and Test with Interpreters
  • Production Run with high performance using compilers and optimized libraries
JavaScript is a fully interpreted language but not really Java
Applets are half-way between traditional compiled and interpreted approaches
Web "systems" can behave like Interpreters with interactive commands at client (gives Web version of MATLAB)
Web Client
including
Java Applets
Web Server
Java/Fortran/C++
Application Backend
HTML version of Scripted Foils prepared 8 January 98

Foil 88 Planned Architecture of DARP User Level Debugging and Rapid Prototyping System

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
HTML version of Scripted Foils prepared 8 January 98

Foil 89 Glossary of Terms II

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
CORBA (Common Object Request Broker Architecture)
  • An approach to cross-platform cross-language distributed object developed by a broad industry group, the OMG. CORBA specifies basic services (such as naming, trading, persistence) the protocol IIOP used by communicating ORBS, and is developing higher level facilities which are object architectures for specialized domains such as banking.
COM (Common Object Model)
  • Microsoft's windows object model, which is being extended to distributed systems and multi-tiered architectures. ActiveX controls are an important class of COM object, which implement the component model of software. The distributed version of COM used to be called DCOM.
ComponentWare
  • An approach to software engineering with software modules developed as objects with particular design frameworks (rules for naming and module architecture) and with visual editors both to interface to properties of each module and also to link modules together.
HTML version of Scripted Foils prepared 8 January 98

Foil 90 Glossary of Terms IV

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Javabean
  • Part of the Java 1.1 enhancements defining design frameworks (particular naming conventions) and inter Javabean communication mechanisms for Java components with standard (Bean box) or customized visual interfaces (property editors). Enterprise Javabeans are Javabeans enhanced for server side operation with capabilities such as multi user support. Javabeans are Java's component technology and in this sense are more analogous to ActiveX than either COM or CORBA. However Javabeans augmented with RMI can be used to build a "pure Java" distributed object model.
RMI (Remote Method Invocation)
  • A somewhat controversial part of Java 1.1 in the enterprise framework which specifies the remote access to Java objects with a generalization of the UNIX RPC (Remote Procedure Call).
HTML version of Scripted Foils prepared 8 January 98

Foil 91 Component Based Programming Environments

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Visual Basic/C++/J++ and ActiveX or Beanboxes with Javabeans give visual approach to software objects
  • specify properties and linkage
Enterprise Javabeans and COM are extending this to distributed computing
Using Web technologies for grid and building modules out of (whatever Javabeans/COM evolves to) allows one to deliver to user HPCC programming environments with comparable friendliness to those in PC world
HTML version of Scripted Foils prepared 8 January 98

Foil 92 What are JavaBeans I

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
They are Java's implementation of "component-based" visual programming
This modern software engineering technique produces a new approach to libraries which become a "software component infrastructure(SCI)"
There is a visual interface to discovery of and setting of values of and information about parameters used in a particular software component
JavaBeans uses the event model of JDK1.1 to communicate between components
  • This is exactly same as new AWT event model and shows power of this new model
HTML version of Scripted Foils prepared 8 January 98

Foil 93 What are JavaBeans II

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
The visual interface allows inspection of and implementation of both individual beans and their linkage . This visual construction of linkage allows one to form nontrivial programs with multiple communicating components
Apart from the event mechanism which is a communication/linkage mechanism, ComponentWare (and JavaBeans in particular) "just" give a set of universal rules (needed for interoperability) for rather uncontroversial (albeit good) object-oriented and visual programming practices
  • Hiding of properties which can only be accessed by methods (which must have special names)
  • Display of these properties (as given by methods)
HTML version of Scripted Foils prepared 8 January 98

Foil 94 What is a Module?

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
In general it is any process, but it is convenient (in the pure form of our web approach) to view each module as a Javabean (or equivalent component)
The Javabean can wrap existing Fortran, Perl or C C++ code by either using native methods or by invoking the code as a separate process
Modules as Javabeans allow them to be stored as objects and inspected visually
  • They can be part of a visual component based programming environment
Wrapping existing code as Javabeans is a good way of renovating "legacy code" so can be used more easily in future!
  • Can document for later use using standard property methods in Javabeans
HTML version of Scripted Foils prepared 8 January 98

Foil 95 3 by 3 Diagram of Programming Environments versus System Complexity from PC to HPCC

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
HTML version of Scripted Foils prepared 8 January 98

Foil 96 HPCC ComponentWare: Essential Ideas

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Large gains in HPCC user productivity will be attained if we can integrate the ideas and technologies of modern (PC) visual programming with classical HPCC approaches
Use of important emerging Web and CORBA technology allows HPCC object (C++.,Java) and visual (CODE, Hence, WebFlow, AVS, Khoros) systems to be enhanced to become parallel component-based visual programming systems.
CORBA does not incorporate HPCC but as it specifies services and not implementation,
  • One can build a high performance HP-CORBA environment where we separate a high functionality modest performance control layer from an optional high performance method invocation layer.
HP-CORBA can be built on Nexus and Globus and it will allow HPCC users access to any CORBA service with an option for high performance when necessary.
The NPAC WebFlow technology can be combined with emerging JavaBean technology to produce a prototype HPcomponent system.
Note industry is ahead with sequential ComponentWare but is only now moving with activeX to distributed systems. HPCC already has visual distributed environments. So HPCC need not be behind if it generalizes modules to Javabeans
HTML version of Scripted Foils prepared 8 January 98

Foil 97 Pure CORBA Architecture for a distributed Information System (There are similar COM and Javabean /RMI Versions)

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
HTML version of Scripted Foils prepared 8 January 98

Foil 98 CORBA Software Model

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
WorkFlow
ORB
System Management
HPcc ?
..............
Trader
Security
..........
Naming
Persistence
Oil & Gas
DMSO Modeling and Simulation
Imagery
Banking
Manufacturing
......
......
Services
Horizontal Facilities
Vertical
Facilities
Standard Interfaces
i.e. Frameworks
HTML version of Scripted Foils prepared 8 January 98

Foil 99 A Parallel Computer Viewed as a Single CORBA Object

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
This is classic host-node computing model
Host is logically distinct but can be on same machine as a "node"
HTML version of Scripted Foils prepared 8 January 98

Foil 100 Each Node of a Parallel Computer viewed as a Separate CORBA Object

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
HTML version of Scripted Foils prepared 8 January 98

Foil 101 Fig. 13: Each node of a parallel computer instantiated as a Corba object.

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
The "Host" is logically a separate Corba object but could of course be instantiated on the same computer as one or more of the nodes. Using the protocol bridge of fig. 15, one could address objects using Corba with local parallel computing nodes invoking MPI and remote accesses using Corba where its functionality (access to very many services) is valuable.
From HPcc as High Performance Commodity Components
HTML version of Scripted Foils prepared 8 January 98

Foil 102 A Message or Protocol Optimization Bridge

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
HTML version of Scripted Foils prepared 8 January 98

Foil 103 Fig 15: A message optimization bridge

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
This allows MPI (or equivalently Nexus, Globus or PVM) and commodity technologies to coexist with a seamless user interface.
From HPcc as High Performance Commodity Components
HTML version of Scripted Foils prepared 8 January 98

Foil 104 Integration of DIS with Object Web Based Computing

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
DoD modeling community is currently evolving towards the HLA(High level Architecture) framework with the RTI (Run Time Infrastructure) based communication bus.
The goal of HLA/RTI is to enhance interoperability across more diverse simulators than in the DIS realm, ranging from real-time to time-stepped to event-driven paradigms.
HLA defines a set of rules governing how simulators (federates) interact with each others. Federates describe their objects via Object Model Template (OMT) and agree on a common Federation Object Model (FOM).
The overall HLA/RTI model is strongly influenced by the CORBA architecture and in fact the current prototype development is indeed CORBA based.
We suggest that next step is to combine CORBA2 (Initial HLA/RTI is CORBA1) with NPS prototype ideas to give a fully object and Web integrated event driven simulation environment.
Java3D is natural visualization environment in this scenario
HTML version of Scripted Foils prepared 8 January 98

Foil 105 Technology Convergence Roadmap - Overview for Forces Modeling, Integrated Modeling and Testing for DoD

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
HTML version of Scripted Foils prepared 8 January 98

Foil 106 DoD M&S Strategy: An Analogy to City Planning

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
HTML version of Scripted Foils prepared 8 January 98

Foil 107 How Did We Get Here?

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
88 89 90 91 92 93 94 95 96
Technical
Management
Limited scope simulations, little interoperability prior to 1988
DSB: Computer Applications
to Training & Wargaming
DIS Standards begun
ALSP- linking of Service wargames
DEPSECDEF Memo
EXCIMS and DMSO established
SIMNET
HLA Baseline approved
HLA begun
Service M&S Offices established
DoDD 5000.59
HTML version of Scripted Foils prepared 8 January 98

Foil 108 Some Terminology

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Federation: a set of simulations, a common federation object model, and supporting RTI, that are used together to form a larger model or simulation
Federate: a member of a federation; one simulation
  • Could represent one platform, like a cockpit simulator
  • Could represent an aggregate, like an entire national simulation of air traffic flow
A federate could be large or small grain -- for initial activity of integrating existing pre HLA simulations, a federate is typically large grain size
However HLA is a "complete" model and one could build simulations where a federate is finr grain object and federation is simulation of these interacting objects
Federation Execution: a session of a federation executing together
HTML version of Scripted Foils prepared 8 January 98

Foil 109 Some More Terminology

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Object: An entity in the domain being simulated by a federation that
  • Is of interest to more than one federate (if internal to an existing non HLA compliant simulation, then this is an object but not an HLA object)
  • Is handled by the Runtime Infrastructure (RTI)
Interaction: a non-persistent, time-tagged event generated by one federate and received by others (through RTI)
Attribute: A named datum (defined in Federation Object Model) associated with each instance of a class of objects
Parameter: A named datum (defined in Federation Object Model) associated with each instance of a class of interactions
HTML version of Scripted Foils prepared 8 January 98

Foil 110 HLA Comprises Three Components: Rules, Runtime Infrastructure, Templates

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
HLA Rules: A set of rules which must be followed to achieve proper interaction of federates during a federation execution. These describe the responsibilities of federates and of the runtime infrastructure in HLA federations
  • HLA Rules correspond roughly to a CORBA vertical facility
Interface Specification: Definition of the interface services between the runtime infrastructure and the federates subject to the HLA
  • In CORBA, this would correspond to services
Object Model Templates: The prescribed common method for recording the information contained in the required HLA Object Model for each federation and federate
  • In CORBA, this would correspond to the IDL specification of objects but in HLA, objects are specified as tables of properties which are linked by interactions whereas in CORBA, one uses methods of objects.
HTML version of Scripted Foils prepared 8 January 98

Foil 111 A Federation Must Play by the Rules

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
1. Federations shall have an HLA Federation Object Model (FOM), documented in accordance with the HLA Object Model Template (OMT)
A FOM is like a particular facility in CORBA
2. In a federation, all representation of objects in the FOM shall be in the federates, not in the runtime infrastructure (RTI)
3. During a federation execution, all exchange of FOM data among federates shall occur via the RTI
4. During a federation execution, federates shall interact with the runtime infrastructure (RTI) in accordance with the HLA interface specification
5. During a federation execution, an attribute of an instance of an object shall be owned by only one federate at any given time
HTML version of Scripted Foils prepared 8 January 98

Foil 112 Each Federate Must Play By the Rules

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
6. Federates shall have an HLA Simulation Object Model (SOM), documented in accordance with the HLA Object Model Template (OMT)
7. Federates shall be able to update and/or reflect any attributes of objects in their SOM and send and/or receive SOM object interactions externally, as specified in their SOM
8. Federates shall be able to transfer and/or accept ownership of attributes dynamically during a federation execution, as specified in their SOM
9. Federates shall be able to vary the conditions (e.g., thresholds) under which they provide updates of attributes of objects, as specified in their SOM
10. Federates shall be able to manage local time in a way which will allow them to coordinate data exchange with other members of a federation
HTML version of Scripted Foils prepared 8 January 98

Foil 113 OMT Components

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Object Class Structure Table
Object Interaction Table
  • Equivalent to methods in CORBA
Attribute/Parameter Table
  • Enumerated Datatype Table
  • Complex Datatype Table
  • Equivalent to properties in CORBA
FOM/SOM Lexicon
  • Remember FOM is the federation object model and SOM the internal simulation object model for a given member federate of the federation
HTML version of Scripted Foils prepared 8 January 98

Foil 114 Object Class Structure Table (Template)

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
General Case
Example
P=Publish and S=Subscribe
HTML version of Scripted Foils prepared 8 January 98

Foil 115 Object Interaction Table (Roughly Methods in CORBA)

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
General Case
Example
HTML version of Scripted Foils prepared 8 January 98

Foil 116 Architecture Splits Functions Between Simulations and Runtime Infrastructure

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Live
Participants
Interface
Runtime Infrastructure
Data Collector/
Passive Viewer
Federation Management Declaration Management
Object Management Ownership Management
Time Management Data Distribution Management
RTI is a bit like IIOP with critical addition of time
management services
HTML version of Scripted Foils prepared 8 January 98

Foil 117 Tango Collaboration System

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
An Applet based system using LiveConnect and plugin with Netscape3 and Signed Applets with Netscape4
Supports general shared event model of collaboration where it can share applications in Java, JavaScript, C, VRML, C++ (Open Inventor)
  • Event sharing coordinated by Java Server
Has conventional general tools
  • Audio/Video Conferencing, Chat rooms, Whiteboard
Developed for command and control
Most extensively used in education -- especially for course between Syracuse and Jackson State
  • Using JavaScript "guided tour" WebWisdom linking to 18,000 foils
HTML version of Scripted Foils prepared 8 January 98

Foil 118 Tango Screen: Talking Heads and White Board

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
From Tango - A Java/WWW-Based Internet Collaborative Software System part of NPAC Overview May 1997
HTML version of Scripted Foils prepared 8 January 98

Foil 119 Typical Web Collaboration Architecture

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
TANGOsim
Basic
Replicated Applications
1)Virtual Users 2)Customized Views
HTML version of Scripted Foils prepared 8 January 98

Foil 120 The TANGOsim C2 Application

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
TANGO Java
Collaboratory
Server
HTTP
Server
MultiMedia Mail
C2 Commander
Chat
VTC
Event Driven
Simulation
Engine
C2 Radar Officer
3D GIS
Scripting
Language
C2 Weather Officer
Message Routing
SW/Data Distrib.
Other
Collaborators
MultiMedia Mail
Chat
Simulation
Engine Controller
All Clients
Typical Clients
HTML version of Scripted Foils prepared 8 January 98

Foil 121 Command and Control Screen with Multimedia Message

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Feb 97 Demonstration of Tango
HTML version of Scripted Foils prepared 8 January 98

Foil 122 A demo of animated objects controlled by simulation engine of Tango collaborative system - III

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
From Tango Project for CEWES Collaborative Tool Meeting
HTML version of Scripted Foils prepared 8 January 98

Foil 123 TANGO Structure of Multidisciplinary Applications

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
TANGO links people and shared applications such as chat board, audio video conferencing, visualizations, shared white board, common AUTOCAD design and related tools
CFD
TANGO Server
Database
Object Broker
MPP
Structures
MPP
Engineer
+ core
services
Visualization e.g.CAVE
Shared AutoCAD
Engineer
+ core
services
HTML version of Scripted Foils prepared 8 January 98

Foil 124 Minimal Web based Multidisciplinary Application

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
This combines TANGO for collaboration with WebFlow to link server side applications
If necessary WebFlow would support high performance inter-module communication as in structures-CFD Linkage example but it would always implement control and this allows TANGO integration with server side computation
  • So we suggest change in usual HPCC integration model to move control to server level so that coupled applications can be directly integrated with services such as TANGO, databases ....
WebFlow communication model is a dynamic dataflow
Of course other server side compute models are possible and in general need (web-linked) data bases, file systems, object brokers etc.,
HTML version of Scripted Foils prepared 8 January 98

Foil 125 Comparison of Communication/Linkage Models

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
WebFlow supports dataflow model where user must supply routines to process input of data that drives module and output of data for other modules
TANGO supports shared state and user supplies routines that read or write either
  • Total state of application or
  • Change in state of application
Can be done for applications like AUTOCAD as vendor supplies necessary API
CFD
Structures
HTML version of Scripted Foils prepared 8 January 98

Foil 126 Java as the Language for Computational Kernels!

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Java for User Interfaces and MetaComputing is natural from its design!
Java for your favourite Conjugate Gradient routine (etc.) is less obvious .....
HTML version of Scripted Foils prepared 8 January 98

Foil 127 Some Critical Features of Java as a Programming Language

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Java likely to be a dominant language as will be learnt and used by a broad group of users
  • We have taught 3 full courses and several tutorials
  • Popular as widely applicable (growing number of API's etc.) and one gets good graphics outpiut easily.
  • Further can use Web to exchange results of your program with peers
  • Expect to be very effective in middle and high school programming
  • Kids will come to University and jobs knowing and expecting to use Java
    • They will not accept Fortran as unfamiliar and less attractive
    • They may accept C++ as a later more complicated language
    • The bottom up revolution!
Java may replace C++ as major system building language
  • Perhaps greater functionality (e.g. pointers) of C++ critical although "WebWindows" favors Java
  • but this is not topic today!
HTML version of Scripted Foils prepared 8 January 98

Foil 128 Comparison of Java and Fortran 77/90

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Clearly Java can easily replace Fortran as a Scientific Computing Language as can be compiled as efficiently and has much better software engineering (object) and graphics (web) capabilities
  • Fortran90 is object oriented but very small user base and not clear if will replace Fortran77
  • Note Fortran90 discussion started in 1978 (after Fortran77 agreed) and took fourteen years and even now Cray's Fortran77 compiler is (on C90 for numerical relativity) much better than their Fortran90 compiler.
  • Originally Fortran90 (as Fortran8X) was designed precisely for Cray architecture systems!
  • This illustrates that informal standards activities (as in the Web and HPF) are most appropriate for rapidly changing technologies
Java can unify classic science and engineering computations with more qualitative macroscopic "distributed simulation and modelling" arena which is critical in military and to some extent industry
HTML version of Scripted Foils prepared 8 January 98

Foil 129 Java Links the Bottom and Top of Pyramid

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Key question is performance of Java
Note Web Software can be run on High Performance IntraNets such as Iway so hardware need NOT be a problem!
HTML version of Scripted Foils prepared 8 January 98

Foil 130 Isn't the Web hardware and software too slow to be interesting for HPCC? -Java- I

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Java is currently semi-interpreted and (as in Linpack online benchmark) is about 50 times slower than good C or Fortran
  • http://www.netlib.org/benchmark/linpackjava/
Java --> (javac)--> Downloadable Universal Bytecodes --> (Java Interpreter)
--> Native Machine Code
  • Just in Time Compilers speed this up by factor of 10
However Language can be efficiently compiled with "native compilers"
Java ----> (native compiler)
---> Native (for Particular Machine) Code
Lots of Interesting Compiler issues for both compiled and scripted Java
HTML version of Scripted Foils prepared 8 January 98

Foil 131 Performance of Java is Dreadful!

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
My SGI INDY gets .54 Megaflops for Java 100 by 100 Linpack
It has 200 Mhz R4400 and current Netlib benchmark for this chip is 32 mflops for optimized Fortran
For better resolution see JPEG Version
HTML version of Scripted Foils prepared 8 January 98

Foil 132 LinPack Java Performance Updated to Sept 30 1996

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
see http://www.netlib.org/benchmark/linpackjava/
Note Just in Time Compilers are giving a factor of 10 from June 96 Measurements!
HTML version of Scripted Foils prepared 8 January 98

Foil 133 LinPack Java Performance Updated to June 3 1997

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
see http://www.netlib.org/benchmark/linpackjava/
HTML version of Scripted Foils prepared 8 January 98

Foil 134 What Limits Performance of Compiled Java?

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Syracuse and Las Vegas Workshops saw no serious problem to High Performance Java on sequential or Shared Memory Machines
Some restrictions are needed in programming model
  • Cynic says write in Fortran with Curly Brackets but more seriously restrictions requires programming discipline but still exploit key advantages of Java
For instance, Avoid Complicated Exception handlers in areas compilers need to optimize!
Should be able to get comparable performance on compiled Java C and Fortran starting with either Java Language or JavaVM bytecodes
The Interpreted (Applet) JavaVM mode would always be slower than compiled Java/C/Fortran -- perhaps by a factor of two with best technology
HTML version of Scripted Foils prepared 8 January 98

Foil 135 Isn't the Web hardware and software too slow to be interesting for HPCC? -Java- III

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
One can use "native classes" which is just a predownloaded library of optimized runtime routines which can be high performance compiled Java, C, C++, Fortran, HPF etc. modules invoked by interpreted or compiled Java
  • This does NOT violate Web Philosophy in our opinion!
Use Native Classes selectively for
  • Compiler Runtime, Matrix Primitives, Image Processing and other engineering/science libraries,
  • PDE primitives such as mesh generators,
  • optimization as needed in resource management or applications
HTML version of Scripted Foils prepared 8 January 98

Foil 136 Classes of Simulations and their High Performance Needs

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
1)Classic solution of large scale PDE or Particle dynamics problem
  • Data parallelism over grid points or particles
2)Modest Grain size Functional Parallelism as seen in overlap of communication and computation in a node process of a parallel implementation.
  • More generally overlap of I/O -- disk,visualization -- and computation
3)Object parallelism seen in Distributed Simulation where "world" modelled (typically by event driven simulation) as set of interacting macroscopic (larger than grid points) objects
  • Objects are weopens, military units etc. in SIMNET/DSI (Forces Modelling)
4)MetaProblems consisting of several large grain functionally distinct components such as
  • Structural Analysis, Airflow, Manufacturing Process, Pricing, Controls etc. in MDO approach to manufacturing and design
  • more generally are components of a Problem Solving Environment
Java: 1) Not Supported, 2) is Thread mechanism, 3) is Java Objects or Applets, 4) is JavaBeans or equivalent
Fortran: 1)is supported in HPF, 2--4) are not supported
HTML version of Scripted Foils prepared 8 January 98

Foil 137 Java and Parallelism?

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
The Web integration of Java gives it excellent "network" classes and support for message passing.
Thus "Java plus message passing" form of parallel computing is actually somewhat easier than in Fortran or C.
Coarse grain parallelism very natural in Java
"Data Parallel" languages features are NOT in Java and have to be added (as a translator) of HPJava to Java+Messaging just as HPF translates to Fortran plus message passing
Java has built in "threads" and a given Java Program can run multiple threads at a time
  • In Web use, allows one to process Image in one thread, HTML page in another etc.
Can be used to do more general parallel computing but only on shared memory computers
  • JavaVM does not support distributed memory systems
HTML version of Scripted Foils prepared 8 January 98

Foil 138 "Pure" Java Model For Parallelism

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Combine threads on a shared memory machine with message passing between distinct distributed memories
"Distributed" or "Virtual" Shared memory does support the JavaVM as hardware gives illusion of shared memory to JavaVM
Message Passing
Message Passing
HTML version of Scripted Foils prepared 8 January 98

Foil 139 Mechanisms for Data Parallelism in HPJava

From General NPAC Foils-B starting June97(PowerPoint) Beijing and Chang Sha China -- 28 Dec 97 to 5 Jan 98. *
Full HTML Index
Java Wrappers (native classes or Server socket connections) around existing data parallel Fortran or C++
Native Java and MPI
  • Need Java MPI Interface as part of framework
Data Parallel Extensions of Java
  • Natural extensions of HPF and HPC++
  • Note one of the best C++ mechanisms (standard template library) doesn't work in Java
Java threads for data parallelism on SMP's
© Northeast Parallel Architectures Center, Syracuse University, npac@npac.syr.edu

If you have any comments about this server, send e-mail to webmaster@npac.syr.edu.

Page produced by wwwfoil on Fri Jan 9 1998