PET FOCUSED EFFORT PROPOSAL Focused Effort Title: Object Web RTI Prototype Thematic Area(s): Scalable Computing Migration HPC Training and DoD User Productivity HPC Performance Metrics/Tools Management and Interpretation of Large Data Sets SciVis for Very Large Problems DoD Challenge Applications PI Name: David Bernholdt PI EMail Address: bernhold@npac.syr.edu PI Telephone: 315 443 3857 PI Fax: 315 443 1973 Project Description: DMSO introduced recently new integration framework for advanced simulation networking called High Level Architecture (HLA) and based on Run-Time Infrastructure (RTI) software bus model. RTI enables federations of real-time/time-stepped and logical-time/event-driven simulations/federates and it optimizes communication via event filtering and publisher/subscriber region/interest matching, supported by the Data Distribution Management (DDM) service. Full and rapid DoD-wide transition to the HLA is strongly advocated by DMSO and facilitated by open public specifications of all HLA components, extensive nation-wide tutorial programs and prototype RTI implementations. In particular, early prototype codes for RTI F.0 ("familiarization" release) and RTI 1.0 subset were released by DMSO to the DoD simulation community in spring/fall '97. The complete RTI 1.0 implementation, including full DDM support is expected by the end of '97. Given the systematic shift of the DoD training, testing and wargaming activities from the physical to synthetic environments, and the ever increasing computational demands imposed on advanced modeling and simulation systems, high performance distributed computing support for HLA will likely play the crucial role in the DoD Modernization Program. Based on our expertise at NPAC in Web/Commodity Software based HPCC, we propose a focused project within the HPCMP/PET program that will adress a critial aspect of rapid deployment of high performance HLA systems: World-wide distributed RTI on top of Java/CORBA based Object Web and focused on Web based collaborative HLA training The DIS->HLA transition process is rapid and, for some people often complex or/and confusing. DMSO provides significant help to the community in terms of high level tutorials and prototype RTI codes. Additional help, as proposed in this project, could be provided by an open, source level RTI subset implementation using Object Web/Commodity technologies such as CORBA and Java. Such RTI would be made broadly accessible both at the source code and runtime/collaborative simulation level, and integrated into the existing and forthcoming HLA training and tutorial suites. At NPAC, we are currently developing Java based Web Object Request Broker (WORB) server that will support HTTP and IIOP protocols and will act as a univeral node of our HPcc (High Performance commodity computing) environment. Given that RTI object bus model is strongly influenced by CORBA and DMSO is in fact interacting with OMG towards proposing HLA as CORBA simulation facility/framework, an early Java/CORBA based RTI prototype seems to be a natural effort in the domain of interactive HLA training. Our Object Web/WORB based RTI subset would support and integrate Web DIS (Java and VRML based) applications under development at the Naval Postgradual School at Monterey, CA, as well as more traditional and substantial simulation codes such as ModSAF and perhaps also SPEEDES, TEMPO or IMPORT, currently at the planning stage as possible FMS training targets for our PET activities at ARL. By the end of this project, we will deliver a prototype object web (CORBA) based RTI kernel (subset), capable of running a a simple demonstration application to be developed locally. This would serve as a demonstration of the integration of DMSO and web technologies and provide a freely available tool. A follow-on project could further develop the system into a full FTI implementation, at which point it would be possible to run real RTI applications as a demonstration of this tool. Benefits: This project will enhance the ability of the FMS community to make use of HPCC resources. Deliverables: CORBA/HPCC based implementation of RTI kernel (subset) with simple application for demonstration purposes (March 1998) Notes: Concurrently with this project, ARL will be supporting an analysis of DMSO RTI with regard to parallelization. Both projects should be completed by or before March 1998, then one can decide on zero one or both of: o Implement Parallelization of DMSO RTI o Complete Public Domain Corba HPCC based RTI which should be able to run significant FMS applications The relationship between these projects might be understood by analogy with MPI (RTI being a runtime infrastructure to support general event and time stepped simulations). There are a variety of different MPI implementations, both freely available and commercial. They have different target platforms (networks of workstations, MPPs, etc.), different performance, and different strengths. Similarly, the Corba/HPCC RTI project shows the relationship with emerging distributed HPCC commodity technologies, while the other focuses adapting DMSO RTI to work more efficiently on a large HPC platform. We emphasize that FMS and HPCC are not easy to mix as there has been little work in this area and so we need upfront technology development. There are no obvious (to us ) HPCC FMS activities which can deliver real results very quickly. Note we have extensively interacted with Wasilausky and understand his needs. We are going to NRaD early November to continue this interaction. We believe continued work with Wasilausky is essential and we will continue to do this. The work we propose is not without risk as we are essentially addressing issues that have not been studied. We believe our approaches are technically sound and this evaluation is based on our experience in both fields. We cannot give the same certainity in our work as say porting MPI would allow! Customer Commitment/Involvement: The FMS CTA lead has requested parallelization of RTI and examination of web technologies in this context.