Kai Hwang - US grants

This project investigates two design methodologies, dynamic systolization and macro-dataflowing, useful in developing multiprocessor supercomputers. The concept of multiprocessor element systolization is generalized from pipeline chaining implemented in Cray X-MP. This technique offers a new network approach to designing reconfigurable systolic arrays for fast evaluation of compound functions and special algorithms. Macro-dataflowing refers to an event-driven approach to developing dataflow supercomputers. Macro-dataflow is effective to support multitasking with variable granularity. The studies include principles of dynamic systolization and macro-dataflowing, development of functional mechanisms, multiprocessing software supports, system reconfiguration techniques, multitasking benchmark studies through simulation experiments, and supercomputer performance evaluations. The major application area emphasizes the solution of Partial Differential Equation (PDE) problems.

This is a project to build a mesh multiprocessor, in which memory busses run along the rows and columns. A dual-ported memory at each grid point in the square mesh is connected to its row bus and its column bus. There is one processor for each element of the main diagonal, which can be connected either to its row or its column. The multiprocessor is to be used in image processing, and may also have some application to the matrix computations used for simulating neural networks.

A new trust model will be developed for grid metacomputing over multiple administrative domains. The GridSec enhances grid operations with seamless security, assured privacy, data integrity, confidentiality, and optimized resource allocations. Distributed micro firewalls and intrusion repelling libraries for protecting grid resources will be generated as a part of this project. The new security system will be designed to adjust itself dynamically with changing threat patterns and network conditions. Fine-grain resource-access control at the file, device, and storage levels will be designed to enhanced the trusted aspect of the system.
The results of this project will benefits all grid applications and offers protection of
shared grid resources. The eventual construction of a production grid platform dedicated for global emergency response and crisis management is one of the project's goals. This work integrates advanced security research with higher education. The project enhances Internet and grid security, reduces the vulnerability of our society, and protects the global economy as a whole. The broader impacts are far reaching in science, education, business, and governments.

Abstract

Program: NSF 04-588 CISE Computing Research Infrastructure
Title: CRI: Reconfigurable computing infrastructure for high end and embedded computing applications
Proposal: CNS 0454407
PI: Prasanna, Viktor K.
Institution: University of Southern California

The investigators will acquire a reconfigurable computer comprised of general purpose processors, field programmable gate arrays (FPGAs), a common memory, and an interconnect fabric joined under a programming model that works with all the parts. The acquisition of this machine will enable research at a realistic scale on actual reconfigurable machines for performance testing, validation, and applications demonstrations. This infrastructure will be robust enough to implement application "kernels" such as (e,g, an LU implementation or n-body simulation) that give realistic scale experimental results. Applications that will be explored include matrix operations, computational genomics, molecular dynamics, density functional theory, and finite element methods. The team will also be able to work on energy efficiency for embedded FPGAs. Broader impacts of this project include the potential impact on reconfigurable systems, use of FPGAs for applications, and discoveries in the applications areas. The investigators participate in USC's Minority Opportunities in Research (MORE) program.

DESCRIPTION (provided by applicant): Cognitive control is a cognitive capacity that allows us to adaptively regulate sensory, perceptual, motor, and other mental processes for goal-directed behaviors. A prominent theory of cognitive control suggests that the frontal cortex is the source of top-down control signals that bias lower-order brain processes to generate voluntary behaviors. However, how control signals are communicated to distribute cortical regions to enhance task-relevant and inhibit task-irrelevant processes remains much debated. The goal of this proposal is to test a hypothesis that the thalamus serves as a hub to mediate top-down biasing signals transmitted from the frontal cortex to targeted cortical sites within sensorimotor networks. This hypothesis is drawn from anatomical evidence indicating that there are transthalamic, cortico-thalamo-cortical pathways that provide widespread and diffuse linkages between distributed cortical regions, and that thalamo-cortical pathways innervate cortical GABAergic inhibitory neurons that are critical for modulating neocortical processes. To test my hypothesis, the role of thalamus in cognitive control will be assessed by systematically examining its functional interaction with the cortex. In Aim 1, I will use resting-state functional connectivity magnetic resonance imaging and graph theoretic measures such as network centrality and participation coefficient to quantify the contribution of the thalamus to functional brain network communications. Studying patients with focal thalamic lesions will further enhance my inferential power, and I will evaluate how thalamic lesions negatively affect network interactions. Completing Aim 1 will demonstrate that the thalamus serves as a pivotal node to support network communication between cognitive control and sensorimotor networks, allowing task-specific biasing signals to reach their appropriate cortical targets. In Aim 2, we will employ a cutting-edge simultaneous transcranial magnetic stimulation (TMS) and functional MRI (fMRI) study to map the direction of information flow within fronto-thalamo-cortical pathways involved in suppressing distracting sensory information. Coupled with a behavioral task, fMRI can detect local and remote TMS effects and how these vary with task conditions. This novel approach will enable me to map task-related network dynamics. We hypothesized that applying TMS pulse over the frontal cortex during distracter suppression will increase thalamic activity, and enhance thalamo- cortical connectivity to suppress cortical activities associated with task-irrelevant representations. Completing Aim 2 will demonstrate that a fronto-thalamo-cortical pathway mediates top-down biasing signals for cognitive inhibition. Many neurologic and psychiatric disorders such as traumatic brain injury, dementia, schizophrenia, depression, and attention-deficit disorders are characterized by deficient cognitive control. Basic knowledge about the thalamus and cognitive control can provide substantial insights into the nature of a large number of disorders that are increasingly acknowledged to be affected by aberrant thalamo-cortical connectivity.

PROJECT SUMMARY Cognitive control is an umbrella term that describes processes that regulate cognitive, perceptual, and motor functions needed to perform adaptive, goal-directed behaviors. Patients with neurological and psychiatric disorders, for example schizophrenia, attention-deficit hyperactivity disorder (ADHD), traumatic brain injury and stroke, suffer from impairments in cognitive control. Although the function of frontoparietal and striatal circuits in cognitive control have been extensively studied, increasing evidence suggests a role for the thalamus. Thalamic nuclei have reciprocal and non-reciprocal connections with multiple cortical regions, integrate modulatory inputs from other subcortical structures, and innervate both excitatory and inhibitory cortical neurons. Despite this prominent connectivity profile, the role of the human thalamus in cognitive control and effects of its dysfunction are not well understood. The objective of the proposed research is to determine the cognitive control functions of the human thalamus. Our central hypothesis, based on our extensive preliminary data and anatomical properties of thalamocortical circuits in animal models, is that thalamic nuclei support cognitive control by modulating cortical activities that include evoked responses, cortical network interactions, and neural oscillations. These cortical activities instantiate neurocognitive processes that select or maintain task-relevant information, and they can be selectively amplified through targeted increases in thalamocortical interactions. Consequently, thalamocortical dysfunction can lead to cognitive rigidity, increased distraction, and poor planning. To test our hypothesis, we propose a novel approach that integrates cognitive behavioral tasks that specifically manipulate processes that select and maintain task-relevant information (for example, working memory and set switching), multimodal neuroimaging (fMRI and EEG), and human thalamic lesions studies. We will first determine the functional organization (topography) of thalamocortical functional connectivity for cognitive control (Aim 1), which has not yet been systematically mapped in humans. We will then determine how thalamocortical interactions select and maintain task-relevant information by examining its relationship with, and modulatory effects on, task- related cortical activities (Aims 2 and 3). For all studies, we will recruit healthy individuals, patients with focal thalamic lesions, and control patients from a lesion comparison group. By collecting multimodal neuroimaging data from patients with focal thalamic lesions, we will determine how the disruption of thalamocortical interactions affects task-related cortical neural activities and behavior. Results from our proposed research will establish how the distribution of thalamocortical connectivity enables thalamic nuclei to participate in multiple cognitive control functions, and specify the cognitive and neural repercussions of thalamocortical dysfunction observed in patients with thalamic stroke, as well as other disorders such as ADHD and schizophrenia.