The Molecular Modeling Team is working with the biology department on a structural model of the ATP binding site of the KATP channel Kir6.1 subunit. This funded research is intended to build on current research and extend our understanding of such channels and of protein homology, leading to improved drug design for conditions such as heart disease or diabetes.
Molecular Modeling Page New Molecular Modeling PageThe Neuron Bank project is an NIH funded joint project between members of the DiMoS group and the Neurons & Networks research group at GSU. The goal of the NeuronBank project is to create a flexible system for managing, analyzing, sharing, and extending our knowledge of identified neurons and the circuits they form. The DiMoS group is primarily concerned with developing infrastructure and distributed software for this project.
Motivated by configuration management systems, such as Revision Control System (RCS), that employ
pessimistic concurrency techniques to avoid the problems of merging conflicting, concurrent
changes made to shared documents, we revisit the idea that pessimistic locks offer potential
collaborative editing system (CES) research opportunities. We recognize that using pessimistic
locks reduces the concurrent access, thus the CES community has lately focused on the optimistic
approach that requires OT and similar merging techniques; but what if the locks were dynamic and
could grow and shrink automatically? By dynamically managing lock granularity, we may allow
maximum concurrent access among many authors and avoid the need to merge disparate versions of
the shared document, which can not be resolved occasionally.
This research focuses on developing an open-system architecture via Web services to combine
heterogeneous client editing and server repository tools. Further, we develop a set of
distributed, peer-to-peer algorithms to dynamically manage pessimistic locks for multiple users
within the CES. Our techniques also allow for the incorporation of operational transformation
(OT) approaches. The consistency, causality preservation, and intention preservation (CCI) model
is assured using delayed/relaxed intermediate consistency among the distributed peers, and all
users are guaranteed to have the most current copy of the portion(s) of the shared document that
they are currently viewing.
Link to Dr. Jon Preston's Work
Information and wireless technologies are introduced in healthcare services in an effort to reduce costs, minimize errors related to data entry and prescriptions, and to provide anytime anywhere service. The limited computation and communication capabilities of devices, vulnerable wireless networks, and fear of security and privacy breaches make mobile healthcare service a challenge to design and implement. We take up emergency health care as our case study and envision that communication, computation, and data resources needed for emergency healthcare applications are available on ‘emergency grid’.
The existing solutions for defining and executing applications in a security-aware manner are limited. Currently, there is no standardized notion of security level and an intelligent scheduler to find the right match of security service providers from among the pool of grid resources. To correctly identify and communicate among a coalition; group key and group communication have to be secure.
This research work focus is therefore to: (i) Define the notion of security level for health applications and scheduling algorithms to execute applications satisfying desired level of security, (ii) Design resource monitor and broker to monitor the available resources and match the application with the best security service providing domains, and (iii) Securing group applications with proper identification and secure communication.
Link to Srilaxmi Malladi's page
We have designed and implemented a middleware for hand-held and other
devices to enable seamless embedded software development and their
deployment on a heterogeneous set of devices across multiple networking
protocols and data formats.
Documentation and Code
Bondflow is an initiative in Service Oriented Computing
Bondflow Web Site
iC2MPI is a framework for generating parallel code from iterative serial code. iC2MPI is meant for the class of iterative graph-structured programs with dynamically varying computational loads. It is expected to provide a valuable testing ground and benchmarking tool, as well as allowing non-specialists to harness some power of distributed systems.
DiMos Laboratory Last modified: Thu Jan 4 18:56:48 EST 2007