A Computational Infrastructure for Algorithmic Implementation: A Step Closer to Theater Deployment

'In this proposed effort, we plan on building a parallel computational infrastructure' 'for the VISSTA lab that will enable us to address ongoing Big Data research problems as well as prepare for future ones. To address the increasing demand in online, low cost computations that such applications require, we will be seeking to set up a Dual Processor multi-core (12) server, with the appropriate distributed memory and high end, input and output capabilities. This will complement and match the input modalities already present at the lab, and expand our capabilities in aggregating, parsing, fusing and ultimately analyzing, big data relevant to the following applications: (a) Sensor Networks of various modalities (static or mobile), (b) Social Networks with information stored locally (c) Bioinformatics data, specifically related to the brain connectome' '(d) Inference complexes of various modalities. The proposed addition of computational power by way of the multi-process architecture will enable us to investigate and pursue a genuine parallel implementation of our many already successful centralized algorithms. It will also provide us with the ability to design a graphics user interface that can support more than one user, with full access to all the processes of the analysis at any given moment. Parallel to the central server, we will be putting in place a secondary peripheral system of high caliber individual PCÕs which will be linked to the Computational platform, the available sensors and amongst themselves, increasing thus the overall computational efficiency.'