NAVOCEANO Project Proposal from NPAC/Syracuse U.
High Performance Commodity Computing (HPcc) Based HLA-Compliant Meta-Cluster Management and Meta-Federation Support for NAVO Applications
Project POC
David Bernholdt, NPAC/Syracuse University
Email: bernhold@npac.syr.eduTelephone: 315 443 3857Fax: 315 443 1973
Background
The advantages of building a complex system (hardware or software) using commodity components to the extent possible are now well recognized. The explosive growth of "web-related" technologies gives this idea new meaning in the context of complex software systems. In parallel with these emerging Web/commodity technologies, DoD mandates now new promising HLA (High Level Architecture) interoperability and federation standards for advanced distributed simulations. This project addresses syneriges between Web/Commodity and DoD/HLA standards and offers an innovating Metacomputing platform for NAVO applications.
HPcc systems are built from distributed object technologies, such as CORBA, Microsoft's COM, Javabeans and Web/XML. One can abstract these to a three-tier model with largely independent clients connected to a distributed network of servers. The latter host various services including object and relational databases and of course parallel and sequential computing. High performance can be obtained by combining concurrency at the middle server tier with optimized parallel back end services. The resultant system combines the needed performance for large-scale HPCC applications with the rich functionality of commodity systems. Further, the architecture, with distinct interface, server and specialized service implementation layers, naturally allows advances in each area to be easily incorporated. The HPcc approach is particularly well suited to the support of multidisciplinary interactions, and in situations were resources (i.e. database system, parallel computers, graphical workstations) can only be accessed in distributed fashion on a network. DoD-mandated HLA standards can be also naturally incorporated in such layered losely coupled framework, and in fact HLA interoperatibility model can play an essential integrative role by offering a universal metacomputing glue as in our WebHLA based meta-federation framework discussed below.
HPcc applications will typically run on heterogeneous distributed hardware platforms ranging from high end MPPs such as Origin2000 to UNIX workstation clusters to commodity PC clusters running a mixture of NT, Linux and Solaris operating systems. Efficient management of such resources and seamless interoperability between individual software components that cross boundaries of operating systems and programming paradigms, is of critical relevance for the success of modern HPC applications.
Project Description
In this project, we will address these challenges, both at the core infrastructure and the application levels. In the first stage of the project, we will analyze NAVO computational resources and application needs, and we will adapt to and install our HPcc infrastructure software on the NAVO facilities. This includes:
Our Object Web RTI is being now certified by DMSO for full HLA compliance and it will assure interoperability of NAVO applications with other DoD systems within the broad DoD Modeling and Simulation area. Furthermore, HLA is being now adopted by other federal agencies (e.g. DoE), large industry (e.g. Boeing) and overseas (France, Germany, Japan, Russia, U.K.) and it has a growing potential to become a global interoperability standard for integrating and federating heterogeneous world-wide distributed computing resources. Our JWORB based WebHLA framework builds HLA on top of Web/commodity base and offers an attractive new HPcc and Metacomputing framework.
In the first stage of this project (Jan-May 99) we will custom tailor and install at NAVO the core WebHLA support. This will allow configuration of NAVO resources selected for this project (likely including a mix of Origins, UNIX workstations and PC clusters) as a coordinated commodity cluster, with individual nodes represented as system level federates and managed by a Cluster FOM (Federation Object Model) to be designed jointly with NAVO researchers. We will also perform analysis of potential applications (to be addressed in the second phase discussed below) and we will address interoperability of our approach with other metacomputing models of interest at NAVO, most notably Legion. Since Legion is a distributed object based system and it develops now a CORBA interface, it should be possible to establish close connectivity with this system both in the JWORB and WebHLA layers. This would allow us to federate with Legion applications and would facilitate HLA-compliance for the Legion domain.
In the second phase of the project (in the period June 99 – June 00 and to be quantified in another proposal) we will select (in consultation with NAVO PET management), adapt and port to our JWORB/HWebHLA platform one or more of NAVO applications. If appropriate, we will also work with the Legion group to implement and demonstrate interoperability on a production Legion installation at NAVO. The chosen applications will be able to operate either internally within the NAVO cluster/federation or they can export NAVO capabilities, packaged as an HLA-compliant Meta-Federate to other MSRCs. We will be conducting similar WebHLA based Metacomputing experiments in CEWES and ARL MSRCs in the area of large scale battlefield / wargaming simulations. Our current focus is on Army components such as terrain, minefields, land vehicles etc. and NAVO could contribute valuable new meta-federates in the areas of weather, ocean or atmosphere simulation domains.
We have organized this project as a 17 month effort divided in to two phases
Phase 1 Jan 99 – May 99, 5 months, NAVO PET Contract Year 3
Phase 2 Jun 99 – May 00, 12 months, NAVO PET Contract Year 4
Because of the as-yet-undefined nature of the user application(s) we will work with, we are deferring the submission of a formal proposal for the second phase (Year 4) of this project. We believe that the Phase 1 effort is substantial and stands on its own for purposes of evaluation.
Project Deliverables
(Timeline denoted as months relative to the start date "S". Schedule
assumes a single demonstration project is chosen. If two are chosen,
work may either proceed concurrently or sequentially, with appropriate
modifications to the schedule.)
Project Benefits
This project will result in specific DoD research teams/projects receiving direct assistance in the incorporation of HPcc ideas into their software development efforts, and at the same time provide experience and examples which will facilitate the wider dissemination of this approach through trainings and other means. JWORB technology will introduce integrated support for several leading commodity technologies for distributed objects such as Java, CORBA, COM and Web/XML. DMSO HLA compliance assured via the WebHLA technology will enable the DoD-mandated style of federating NAVO applications with other systems, labs and centers. Interoperability with Legion would enable federating with Legion applications and it would facilitate the addition of HLA-compliance to the Legion model.
Collaboration
While we are not yet in a position to identify specific research teams/projects which fit this proposal, the willingness and responsiveness of the team to work jointly on this project will be an important factor in the choice. The customers of this project will, of necessity, have a high level of involvement in this effort if it is to be successful.
Collaboration with the UVa Legion group and those involved in its deployment/operation at NAVO will also be required.
Budget Justification