- M-Ware
The Middleware (M-Ware) project in Georgia Tech's College of
Computing addresses future distributed systems that are composed of
computational data streams, exemplified by (1) high end data
generators, transformers, and data viewers employed by end users --
termed Distributed Laboratories in our previous research -- or (2)
sets of sensors, data transformers and interpreters, and actuators (or
end users).
- Turbulent Combustion
Turbulent combustion research is carried out under the direction of Suresh Menon of the Computational Combustion Laboratory, Georgia Tech School of Aerospace Engineering.
- Massively Parallel Quantum Chemistery (MPQR) at the Ab Initio limit
The linear R12 versions of the conventional ab initio electron
correlation methods have recently emerged as one of the most promising
next-generation approaches to highly accurate (better than 0.1
kcal/mol) theoretical predictions for molecular systems. The
practical success of such methods is due to the explicit use of the
interelectronic distances in the wave function expression and the
approximate treatment of complicated three- and higher electron
integrals. Unfortunately, computer implementations of linear R12
methods exist but are extremely scarce and do not yet take advantage
of the modern parallel computing environments. This project deals with
adding capability to compute molecular energies with linear R12
versions of the second-order Moller-Plesset (MP2-R12) and coupled
cluster singles and doubles (CCSD-R12) methods to Sandia's
Massively Parallel Quantum Chemistry (MPQC) program.
- Computational Materials Science
Research in the Center for Computational Materials Science (CCMS) focuses on the development of analytical models and novel computer-based classical and quantum molecular dynamics simulations for investigations of a wide range of condensed matter phenomena. The Center is headed by Regents' Professor Uzi Landman.
- Materials Research
Simulations of heterogeneous material structure, often at the scale of microstructure or mesoscopic scales, are essential to constructing the next generation of constitutive laws for material deformation and damage.
- ASAN Project
The Active System Area Network project focuses on the development of
both hardware and software. The "active" part of the name emphasizes
the application-specific as well as system-level computations used in
the optimizing data intensive inter-processor communications.
- Large-scale Optimization
Large-scale optimization research is conducted by The Logistics Institute of the School of Industrial and Systems Engineering under the direction of Ellis Johnson.
- Adaptive Intrusion Detection
An adaptive intrusion detection
system (IDS) has multiple dynamically configured front-end and back-end
detection modules, and a monitor. Its run-time performance is measured
in terms of both accuracy and delay in detection. In overload and
crashed situations, detection workload can be shed from the front-end to
the back-end according to resource constraints and cost-effectiveness.
- Soft Architecture
A soft architecture is a computer built by integrating dynamic compiler software with simple, fast processing hardware. The dynamic compiler is used to replace hardware structures to enable higher speed, lower power or greater functionality.
- 3D Models from Vision Data
The goal of this BORG lab project, headed by Prof. Frank Dellaert, is to
build a 3D world model from thousands of 2D images acquired by a team of
autonomous robots.
- Semiconductor Materials
The Computational Electronics Group at Georgia Tech headed by Prof. Brennan
uses the Intel Cluster to support extensive numerical simulation of
semiconductor materials and related devices. Through these simulations the
basic transport properties and device workings of new, emerging
semiconductor materials is studied.
-
The Stampede Cluster System
Emerging application domains such as interactive vision and animation
require the efficient management of "temporally evolving" data, which
often translates to providing "soft real time" guarantees for such
applications. Stampede system provides a higher level data abstraction
called Space-time Memory which is well-suited to providing the common
parallel programming requirements found in most interactive
applications, namely, buffer management, inter-task synchronization,
and meeting real-time constraints.
- DEOS: Distributed Extensible Open Systems
The purpose of the DEOS project is to create a basis for the
runtime creation of novel system services across the wide range of
wired, wireless, high and low end platforms in common use for next
generation applications. This necessitates the creation of quality of
service (QoS) support including OS- and application-level mechanisms for
online monitoring and for system adaptation. Applications targeted
have in common their real-time and dynamic nature.
- MSREC/PADS (simulation center)
MSREC is the Modeling & Simulation Research & Education
Center. It focuses on large-scale simulations for cross-disciplinary
research. Current projects include air transportation models, network
simulations, and defense simulations.
|
