PET FOCUSED EFFORT PROPOSAL from NPAC/Syracuse University

Focused Effort Title:
	High Performance Commodity Computing (HPcc) for NAVO Applications

Thematic Area(s):  
	Scalable Computing Migration
	HPC Training and DoD User Productivity

PI Name: David Bernholdt

PI EMail Address: bernhold@npac.syr.edu

PI Telephone:	315 443 3857
PI Fax:		315 443 1973

Project Description:

The advantages of building a complex system (hardware or software)
using commodity components to the extent possible is now well
recognized.  The explosive growth of "web-related" technologies gives
this idea new meaning in the context of complex software systems.

HPcc systems are built from distributed object technologies, such as
CORBA, Microsoft's COM, Javabeans, and less sophisticated web and
networked approaches. 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.

One or two NAVO applications (depending on their complexity) will be
the target for this project, most likely in the CWO area, which is
very prominent at NAVO. The application will provide a single,
comprehensive access to all application services, including a seamless
access to DoD HPC resources. This web-based system will empower local
and remote technology use, provide navigation aids for users with
varying levels of familiarity with the system, and give access to
remote data for browsing and visualization.  This new system will be
based on industry standard technologies and will reuse legacy as well
as commodity software. Syracuse will implement a prototype system
that, through a web interface, will allow execution of simulations on
a geographically distributed system, including a local host.  The
front-end of the system will allow the user to specify computational
resources to be used, location of the input data sets and the format
of the output. The system will upload the specified data to the
compute server and launch the application(s) there. The user will
control the simulation using the original application's GUI displayed
on his/her local display device, and if requested, the output will be
feed to a local visualization engine.

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.

Required Resources:
	$75,000 over 12 months

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.)

* Seminar at NAVO on HPcc (S+1)
* Meetings with PET & MSRC staff as well as NAVO users to identify a
  suitable project. (S to S+3)
* Specific project defined (S+3)
* Prototype demonstration (S+7, or as fits chosen project)
* Final HPcc demonstration (S+11)
* Training at NAVO on HPcc (S+12)
* Technical report on results of effort (S+12)

Notes:

A related effort is underway at CEWES, focusing on DoD projects in the
EQM and FMS areas.  This project will leverage the understanding and
some of the software developed so far for the CEWES project.

At present, the overall security model for all of the MSRCs is in some
conflict with the HPcc idea.  However it is important to note that
HPcc is NOT inherently insecure.  Rather, the security model being
implemented by the MSRCs has not given any consideration to the needs
fo new technologies such as HPcc.  As there is interest in these
approaches at all of the MSRCs, we believe that the security model
will have to be re-examined in the near future on a program-wide
basis.  We expect to be invoved in these discussions, and will work
with the current security model to the greatest extent possible,
however we believe that the most effective use of our effort at this
time is to focus on developing the technology, to increase its utility
and user interest, so that it will ultimately be worth the effort to
revise the security model to accomodate these approaches.

Customer Commitment/Involvement:

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.