Welcome to the Java Grande Forum Station
at Sun Microsystems Booth
SC 98 - Orlando, Florida

Looking for Technical Reports?

Meet members of the JGF at Sun Microsystems' Booth

Ronald F Boisvert - Denis Caromel - Bryan Carpenter
Geoffrey Fox - Dennis Gannon - Vladimir Getov
David Henty - Tomasz Haupt - Piyush Mehrotra
Cleve Moler - Jose Moreira - Srilekha Mudumbai
Frank Pietryka - Mark Snir - George K. Thiruvathukal
Anne E Trefethen - Julien Vayssiere
Gregor von Laszewski -   Martin Westhead
Guansong Zhang

...and other participants!

Demonstrations and Descriptions:



contact person:

Bryan Carpenter

NPAC, Syracuse University



 mpiJava is an object-oriented Java interface to the standard Message Passing Interface (MPI). The interface was developed as part of the HPJava project, but mpiJava itself does not assume any special extensions to the Java language - it should be portable to any platform that provides compatible Java-development and native MPI environments.

The current release of mpiJava provides the full functionality of MPI 1.1. It is implemented as a set of JNI wrappers to native MPI packages. Platforms currently supported include Solaris using MPICH or SunHPC-MPI, and Windows NT using WMPI 1.1. We are actively developing the release and intend to add new features such as object serialization and support for additional platforms. The Java API is defined in the document "A Draft Java Binding for MPI".


 contact person:

Vladimir Getov

School of Computer Science,
University of Westminster
London, UK



Towards Portable Message Passing in Java: Binding MPI. We present a way of successfully tackling the difficulties of binding MPI to Java with a view to ensuring portability. We have created a tool for automatically binding existing native C libraries to Java, and have applied the Java--to--C Interface generating tool (JCI) to bind MPI to Java. The approach of automatic binding by JCI ensures both portability across different platforms and full compatibility with the MPI specification. To evaluate the resulting combination we have run a Java version of the NAS parallel IS benchmark on a distributed--memory IBM SP2 machine.

 contact person:

Srilekha Mudumbai

Lawrence Berkeley National Laboratory



 Akenti is a security model and architecture that is intended to provide scalable security services in highly distributed network environments. The project goals are:
  • to achieve the same level of expressiveness of access control that would be accomplished through a local human controller in the decision loop
  • to accurately reflect the existing policy: authority, delegation, and responsibility present in these environments.

The approach makes use of:

  • digitally signed certificates capable of carrying:
    • user identity authentication
    • resource usage requirements ("use-conditions")
    • user attribute authorizations ("attribute certificates")
    • delegated authorization
  • authorization decisions split among on-line and off-line entities



contact person:

Tomasz Haupt

NPAC, Syracuse University



 Our research addresses needs for high level programming environments and tools to support distance computing on heterogeneous distributed
commodity platforms and high-speed networks, spanning across labs and facilities. More specifically, we are developing WebFlow - a scalable, high level, commodity standards based HPDC system that integrates:
  • High-level front-ends for visual programming, steering, run-time data analysis and visualization, and collaboration built on top of the Web and OO commodity standards (Tier 1).
  • Distributed object-based, scalable, and reusable Web server and Object broker Middleware (Tier 2)
  • High Performance Backend implemented using the metacomputing toolkit of GLOBUS (Tier 3)


contact person:

Denis Caromel

Univ. de Nice Sophia Antipolis



C3D: A distributed raytracer for benchmarking Java RMI & Serialization

C3D is a Java benchmark application that measures the performance of a 3D ray tracer renderer distributed over several Java virtual machines using Java RMI. This benchmark gives indication of the performance of the serialization process and Java RMI itself. The benchmark is an automated version of C3D, which is both a collaborative application and a distributed ray tracer: users can interact
through messaging and voting facilities in order to choose a scene that is renderer using a set of distributed rendering engines working in parallel. Near-linear speedup is achieved with up to half a dozen rendering engines running in parallel and communicating using Java RMI.


contact person:

Cleve Moler


National Institute of Standards and Technology



 JAMA is a basic linear algebra package for Java. It provides user-level classes for constructing and manipulating real, dense matrices. It is meant to provide sufficient functionality for routine problems, packaged in a way that is natural and understandable to non-experts. It is intended to serve as the standard matrix class for Java, and will be proposed as such to the Java Grande Forum and then to Sun. A straightforward public-domain reference implementation has been developed by the MathWorks and NIST as a strawman for such a class. We are releasing this version in order to obtain public comment. There is no guarantee that future versions of JAMA will be compatible with this one.

JAMA is comprised of six Java classes: Matrix, CholeskyDecomposition, LUDecomposition, QRDecomposition, SingularValueDecomposition and EigenvalueDecomposition.

 contact person:

Dennis Gannon

Indiana University


 DPSE-CAT is a distributed problem solving environment component architecture toolkit. The component model for the CAT is based on a java bean-like architecture that allows programmers to dynamically compose scientific computations written in either Java or HPC++ into a single distributed application. the individual software components are objects that interoperate by means of the Java RMI semantics. However, the actual implementation of RMI is over the Globus/Nexus communication system. The CAT also contains a Java based information browser that lets the user have access a distributed directory of Java and HPC++ components.

Future version of DPSE-CAT will integrate Java Jini technology with Globus serevices and will support the DOE2000 Common Component Architecture model currently under design.



contact person:

Roldan Pozo

National Institute of Standards and Technology



SciMark is a composite Java benchmark measuring the performance of numerical kernels occurring in scientific and engineering applications. It consists of five kernels which typify computational routines commonly found in numeric codes: FFT, Jacobi relaxation, sparse matrix-multiply, Monte Carlo integration, and dense LU factorization. These kernels are chosen to provide an indication of how well the underlying JVM/JITs perform on applications utilizing these types of algorithms. The problems sizes are purposely chosen to be small in order to isolate the effects of memory hierarchy and focus on internal JVM/JIT and CPU issues. SciMark scores are normalized to a Sun SPARC 10 running Netscape Navigator 4.04 (SciMark = 1.0) A SciMark of 3.0 indicates execution time that is 3 times faster than the reference system.

SciMark is packaged as a Java applet which can be downloaded from the SciMark Web page. Results can be returned to NIST for display in a list of contributed results.

Java Grande Forum Technical Reports

JGF-TR-1: Making Java Work for High-End Computing PostScript - PDF
JGF-TR-2: Desktop Access to Remote Resources HTML
JGF-TR-3: MPI for Java: Position Document and Draft API Specification PostScript

Return to top of document...