Healthcare information systems are becoming ubiquitous and thus increasingly subject to attack, misuse and abuse. Mechanisms are needed to help analysts disambiguate regulations so that they may be clearly specified as software requirements. In addition, regulations are increasingly requiring organizations to comply with the law and account for their actions. We propose a requirements management framework that enables executives, business managers, software developers and auditors to distribute legal obligations across business units and/or personnel with different roles and technical capabilities.

This research focuses on how society uses, values, and protects citizens? personal information. From the perspective of system design, software engineers need methods and tools to enable them to design systems that reflect those values and protect personal information, accordingly. This research examines how privacy considerations and value systems influence the design, deployment and consequences of IT. The goal is to develop concepts, tools and techniques that help IT professionals and policy makers bring policies and system requirements into better alignment. An action-oriented set of conceptual tools, including guidelines and privacy- relevant policy templates will be constructed and validated.

Privacy is increasingly a major concern that prevents the exploitation of the Internet's full potential. Consumers are concerned about the trustworthiness of the websites to which they entrust their sensitive information. Although significant industry efforts are seeking to better protect sensitive information online, existing solutions are still fragmented and far from satisfactory. Specifically, existing languages for specifying privacy policies lack a formal and unambiguous semantics, are limited in expressive power and lack enforcement as well as auditing support. Moreover, existing privacy management tools aimed at increasing end-users' control over their privacy are limited in capability or difficult to use.

Since 1995, the Triangle Computer Science Distinguished Lecturer Series (TCSDLS) has been hosting influential university researchers and industry leaders from computer-related fields as speakers at the three universities within the Research Triangle Area. The lecturer series, sponsored by the Army Research Office (ARO), is organized and administered by the Computer Science departments at Duke University, NC State University, and the University of North Carolina at Chapel Hill. This proposal argues for continuation, for an additional 3 years, of this highly successful lecturer series.

The goal of this project is to develop an extensible framework for designing and using derived data in answering database queries efficiently. The outcomes of the project are expected to be general and independent of a specific data model (e.g., relational or XML), while giving guarantees with respect to query-performance improvement. The approach consists of developing and evaluating mathematical models and algorithms for common types of queries on relational and XML data. Expected outcomes of the project include automated tuning of data-access characteristics in a variety of applications, thus enhancing the quality of user interactions with data-intensive systems.

This research proposal builds upon recent advances in Semantic Web and information integration. The objective is to produce the formalism and lay the groundwork for a broad spectrum of information integration, that facilitate rapid development of efficient and effective integration systems. We concentrate on information integration from XML sources. A user query is translated and executed on XML data, which is the native model of the source or can be obtained using wrappers to represent sources' data in XML. We consider, in particular, the XML pipeline approach as the preferred architecture for the implementation of an extensible system for integration/mashup.

This project aims to better understand the process by which one can change the beliefs and preferences of populations in different cultures. We propose to develop formal models of belief change in individuals and populations. We will use the ongoing MURI effort on "Computational Models for Belief Revision, Group Decisions, and Cultural Shifts" as a starting point, but will focus on developing models that exploit formal notions of entrenchment, habit, and mental inertia.

The goal of this project is to develop and deploy a GENI instrumentation framework, integrate it into one of thecfive control framework prototypes, and develop a set of experimenter capabilities to enable cross-layer experimentation in the optical substrate.

This proposal proposes to build an outdoor wireless mesh testbed comprised of a large number of low-cost experimental fabricated nodes and a small number of commercially available nodes. The testbed will be built in two stages: in the first stage, nodes placed on pushcarts will be temporarily placed outdoors for trials and tests; in the second phase, permanent antenna placements will be installed on equipment poles over a large area of the Centennial Campus of North Carolina State University. The testbed will leverage experience of, as well as enable the research of NCSU researchers participating in the Secure Open Systems Institute (SOSI), currently engaged in DoD, NSF, and other projects. Current and envisaged research activities of SOSI researchers address secure and redundant routing, energy-efficient routing, topology control, localization, cross-layer optimization, security and performance of SIP and VoIP, secure virtualization of network and compute resources, social networking. The proposed testbed will provide realistic large-scale outdoor wireless network environments for evaluating and validating the ideas, protocols and systems conceived from these activities. The data and experience gained from operating and managing a real network environment will also provide practical insights for students and researchers on the operation of large-scale heterogeneous mesh networks that help identify new security and performance problems and develop their practical solutions.

Core counts will increase in the next several years to tens of cores per processor, however, memory bandwidth will increase but only for blocks of data. Consequently, existing applications will have to be significantly modified to take advantage of multiple cores. This research program, a collaboration between NCSU and RENCI, creates an adaptive, parallel runtime system for efficiently executing applications on processors with numerous cores.

NOAA awarded $12.5 Million to fund the Interdisciplinary Scientific Environmental Technology (ISET) Cooperative Research and Education Center. NC A&T State University is the lead institution. The team includes a diverse network of scientists, and engineers from A&T, NC State University, University of Minnesota, University of North Carolina at Pembroke, City University of New York, University of Alaska Southeast, California State University-Fresno, Fisk University as well as industrial, state and federal government partners. NC State University is the lead university for the research thrust on the analysis of global observing systems that includes numerical and physical research and analysis of hurricanes.

We propose to explore the new computing model of offering computation- and/or data-intensive cloud services on active nodes serving on-demand utility computing users. More specifically, we plan to (1) assess the efficacy of resource sharing between foreground interactive utility computing workloads and background high-throughput cloud computing workloads on multi-core servers, in terms of energy saving and performance interference; (2) develop a scheduling and load management middleware that performs dynamic background workload distribution considering the energy-performance tradeoff; and (3) exploits the use of GPGPUs for cloud services on active nodes running foreground workloads mainly on the CPUs.

The goal of this project is to develop efficient and light-weight VM management techniques to greatly improve the scalability and resource-efficiency of the distributed virtual computing infrastructure.

Large-scale hosting infrastructures have become important platforms for many real-world systems such as cloud computing, virtual computing lab, enterprise data centers, and web hosting services. However, system administrators are often overwhelmed by the tasks of correcting various system anomalies such as performance bottlenecks, resource hotspots, and service level objective (SLO) violations. The goal of this project is to develop novel online anomaly prediction and diagnosis techniques to achieve robust continuous system operation. The major contributions will be an integrated framework consisting of three synergistic techniques: i) self-compressing information tracking to achieve low-cost continuous system monitoring; ii) online anomaly prediction that can raise advance alerts to impending anomalies; and iii) just-in-time anomaly diagnosis that can perform online anomaly diagnosis while the system approaches the anomaly state.

In the research component of my career development program, I focus on strategies for addressing the challenges and opportunities that face the deployment of structured P2P systems. In particular, I introduce new schemes to locate resources and strategies to serve them. I also introduce new schemes for topology interference, integration, and organization in order to optimize content distribution. The proposed educational aspect of my career development program focuses on (1) enhancing our department�s networking curriculum, (2) extending opportunities for women, under-represented minorities, and undergraduates in research, (3) encouraging students to participate in the computer science community outside the university.

This proposal describes a project to use scientific visualization to display, monitor, and analyze network-based data. Interest in flexible methods to visualize network environments has grown in recent years, particularly with the recent emphasis on network and data security and reliability. Previous work was conducted primarily with Cisco System, with presentations to MCNC. We have partnered with CACC members to identify network-related problems that will benefit from advanced visual representations.

Due to its interdisciplinary nature and rapid pace, Bioinformatics is a challenging task for students and teachers. Despite many excellent text books and tutorials, there are hardly any supplementary educational tools such as visualizations, animations, or simulation games available. We will address this lack of resources by developing a library of animations for Bioinformatics algorithms and applications, organizing a symposium about Bioinformatics Education with focus on educational tools, and developing an online Bioinformatics education resource portal.

Alternative splicing (AS) is an important mechanism of gene regulation that contributes to transcriptome and proteome diversity. The role of AS in specific biological processes is largely uninvestigated. Our goal is to measure extent and functional significance of AS in Arabidopsis thaliana defense against the bacterial pathogen Pseudomonas syringae pv tomato (Pst). Specific objectives include 1) discover Pst-induced AS via long-read mRNA-Seq, 2) characterize genome-wide AS and gene expression patterns associated with Arabidopsis-Pst interactions via short-read mRNA-Seq, and 3) determine the impact of AS on protein structure, validate selected AS isoforms and perform functional analysis of selected AS genes.

Research in virtualization technologies has gained significant momentum in recent years. One of the basic yet powerful enabling function in many virtualization research efforts is virtual machine introspection or VMI: Observing a VM's states and events from outside the VM. The goal of this project is to develop OpenVMI: a software-based research infrastructure for VMI, which is expected to enable new research and education opportunities, including, but not limited to, safe malware experiments, intelligent virtual infrastructure management etc.

Botnets are recognized as one of the most serious threats to today's Internet. To combat them, one key step is to effectively understand how the botnet members communicate with each other. Unfortunately, the trend of adopting various obfuscation schemes (e.g., encryption) in recent bots greatly impedes our understanding. The main thrust of this research is the investigation of several interrelated key techniques to overcome the above challenges and significantly enrich the understanding of botnet command and control.

There is an alarming trend that elusive malware is armed with techniques that detect, evade, and subvert malware detection facilities of the victim. On the defensive side, a fundamental limitation of traditional host-based anti-malware systems is that they run inside the very hosts they are protecting, making them vulnerable to malware's counter-detection and subversion. To address this limitation, solutions using virtual machine (VM) technologies advocate placing the malware detection facility outside of the protected VM. However, a dilemma exists between these two approaches: The "out of the box" approach gains tamper resistance at the cost of losing the native, semantic view of the host enjoyed by the "in the box" approach. To resolve the above dilemma, a new approach called OBSERV ("Out of the Box with SEmantically Reconstructed View") is introduced to achieve the advantages of both camps by reconstructing the semantic internal view of a VM from external, low-level observations. OBSERV enables two exciting malware defense opportunities: (1) malware detection by view comparison and (2) real-time detection and stoppage of kernel-level rootkits. The broader impact of this research is two-fold: (1) It will enhance the trustworthiness and effectiveness of widely deployed anti-malware systems. Moreover, OBSERV is expected to be viewed favorably by the anti-virus software industry because of its support for existing off-the-shelf anti-virus software. (2) Results from this research will lead to the development of education materials for undergraduate and graduate courses and for professional training sessions.

To realize the vision of high assurance computing architectures, it is important to have the capability of monitoring, controlling, and manipulating a target machine with high trustworthiness and tamper-resistance. In this project, we propose to investigate and develop high-assurance, active introspection capability for virtual machines (VMs). Such capability is transparent to the guest VM running in the host, whereas the host will have a control interface to affect the guest VM's operation. Moreover, we will explore the feasibility of moving such capability into a hardware-based hypervisor for higher assurance and tamper-resistance. If successful, our research will be a solid step toward high-assurance computing with applications in trusted cloud computing, virtual cybercraft, and honeypot/honeyfarm systems.

Cyberinfrastructures are facing increasingly stealthy and sophisticated malware threats. For example, recent reports have suggested that new computer worms and viruses deliberately avoid fast massive propagation. Instead, they lurk in infected machines and inflict contaminations over time, such as rootkit and backdoor installation, botnet creation, and private data theft. Current methods for detection and investigation do not fully exploit the use of information flows tracked at the operating system level. We argue that OS-level information flow is currently an under-utilized tool for malware investigation. In this project, we propose to use operating system information flows to propagate malware break-in provenance information and will demonstrate that provenance preservation can help achieve more efficient and effective malware investigation. We will also show that this technique can be used to produce live alerts for malware that existing tools are unable to provide.

NCSU Center for Advanced Computing and Communication (CACC) plans to host a workshop July 31, 2007 –August 1, 2007 at the William and Ida Friday Institute for Innovational Education on NCSU’s Centennial Campus in Raleigh, NC. The goal of the workshop is to bring together select individuals from industry, government, and academia to develop a national series of workshops to address problem of stimulating interest in STEM K-12.

Center for Advanced Computing and Communication in association with the Friday Institute, National Science Foundation, and NSA High Confidence Software and Systems (HCSS) seek to advance outcomes from the recent National Workshop on Stimulating and Sustaining Excitement and Discovery in K-12 STEM (Science, Technology, Engineering, Mathematics) Education by supporting a new online computer science curriculum pilot, aiding the NC State Kenan Fellows program to expand to national audience, and supporting existing residential middle grade math and related science summer camp sponsored by UMES.

This program will establish a national Secure Open Systems Institute (SOSI), located on North Carolina State’s premier Centennial Campus that will be a global center for Open Source security research and development. The goals are twofold. First, SOSI will significantly contribute to strengthening mission critical information technology infrastructures vital to the Department of Defense, state and nation. Second, SOSI will accelerate the creation and growth of high tech industries in North Carolina and beyond by providing a centralized repository of research results, testing tools and qualification services.

It has long been recognized that affect is central to cognition and learning. Affect plays a particularly important role in the rich interactions offered by game-based learning environments. Devising computational models of affect recognition and affect expression could significantly increase the efficacy of interactive learning environments. In this project, we propose an inductive approach to affect modeling that induces models of affect recognition and affect expression from students' physiological signals and from event streams generated by game-based learning environments. The resulting models will be empirically evaluated with respect to their accuracy, efficiency, and contribution to student learning gains and motivation.

The project will explore and identify the key technical challenges involved in producing justifications for recommendations regarding objectives and plans using spoken or textual natural language. Existing decision support systems generally provide little or no such justification for their actions, making it difficult for even well-trained users to trust them. JOUST-enabled systems will provide justifications that are relevant to the user's objectives and mission, aware of the user's context and environment, and expressed naturally using language appropriate to both the task and the user.

Multiple representations are central to the creative process. The objective of the project is to design and empirically evaluate an interactive creativity environment that supports the automatic mapping of one representation to another that is fundamentally different but complementary. In particular, the proposed work will focus on the design, implementation, and evaluation of the Narrative Theatre, an interactive narrative-centered creativity environment. Rigorous comparative studies using both quantitative and qualitative methods will explore the hypothesis that the multiple representations supported by the Narrative Theatre will significantly enhance the creative process in measurable ways.

Pedagogical agents are embodied software agents that have emerged as a promising vehicle for promoting effective learning. The proposed work has two complementary technology and learning thrusts. First, it will develop a full suite of Bayesian pedagogical agent technologies that leverage probabilistic models of inference to systematically reason about the multitude of factors that bear on tutorial decision making in dynamic high-performance inquiry-based science learning environments. Second, it will provide a comprehensive account of the cognitive processes and results of interacting with Bayesian pedagogical agents in inquiry-based science learning by conducting extensive empirical studies of learning processes and outcomes.

Despite the great promise offered by game-based learning environments for elementary science education, realizing its potential poses significant technological challenges. In this project we will develop a full suite of intelligent game-based learning environment technologies for elementary science education. To promote effective science learning, we will create intelligent game-based learning environment technologies that leverage the rich interactive 3D game environments provided by commercial game engines and the inferential capabilities of intelligent tutoring systems. We will also provide a comprehensive empirical account of the cognitive processes and results of elementary students interacting with intelligent game-based learning environments for science education.

While individual workstations in scientific research environments have become more powerful, they cannot meet the needs of today's interactive data processing tasks. Meanwhile, idle desktop resources are not efficiently utilized. This project aims at harnessing the collective idle resources within institutional boundaries to speed up computation- or data-intensive tasks routinely executed on desktop machines. We will build a novel desktop parallel computing framework, which will integrate distributed computing and storage resources to create an execution platform similar to that provided by a parallel computer, while maintaining the comfort and responsiveness of desktop sequential computing and the autonomy of resource donors.

Xioasong Ma's joint work with NCSU and Oak Ridge National Laboratories (ORNL) will bridge the gap between the two organizations in a practical manner to cooperatively research parallel I/O in conjunction with the Genomes to Life (GTL) and Scientific Data management projects within the Computer Science and Mathematics Division at ORNL.

In this proposal, we address the I/O stack performance problem with adaptive optimizations at multiple layers of the HEC I/O stack (from high-level scientific data libraries to secondary storage devices and archiving systems), and propose effective communication schemes to integrate such optimizations across layers. Our proposed PATIO (Parallel AdapTive I/O) framework will coordinate storage resources ranging from processors to tape archiving systems.

We propose to explore the new computing model of offering computation- and/or data-intensive cloud services on active nodes serving on-demand utility computing users. More specifically, we plan to (1) assess the efficacy of resource sharing between foreground interactive utility computing workloads and background high-throughput cloud computing workloads on multi-core servers, in terms of energy saving and performance interference; (2) develop a scheduling and load management middleware that performs dynamic background workload distribution considering the energy-performance tradeoff; and (3) exploits the use of GPGPUs for cloud services on active nodes running foreground workloads mainly on the CPUs.

Assuring deadlines of embedded tasks for contemporary multicore architectures is becoming increasingly difficult. Real-time scheduling relies on task migration to exploit multicores, yet migration actually reduces timing predictability due to cache warm-up overheads and increased interconnect traffic. We propose a fundamentally new approach to increase the timing predictability of multicore architectures aimed at task migration in embedded environments making three major contributions. 1. We develop novel strategies to guide migration based on cost/benefit tradeoffs exploiting both static and dynamic analyses. 2. We devise mechanisms to increase timing predictability under task migration providing explicit support for proactive and reactive real-time data movement across cores and their caches. 3. We propose rate- and bandwidth-adaptive mechanisms as well as monitoring capabilities to increase predictability under task migration. Our work aims at initiating a novel research direction investigating the benefits of interactions between hardware and software for embedded multicores with respect to timing predictability.

The objective of this work is to provide functionality for the Berkeley Lab Checkpoint/Restart (BLCR) Linux module to support advanced fault-tolerant capabilities, which are of specific value in the context of large-scale computational science codes running on high-end clusters. We have developed a set of techniques to reduce this checkpoint/restart overhead. We propose to integrate a job pause mechanism, live migration support and an incremental checkpoiting mechanism into the latest BLCR version.

The objective of this work is to systematically model and study large-scale reliability, to develop novel, scalable mechanisms in support of proactive FT and to significantly enhance reactive FT. The proposed work promises contributions in modeling reliability, monitoring system health and devising novel reactive as well as proactive schemes to tolerate faults. Overall, this work is targeted at alleviating limitations of current reactive FT schemes. It contributes to fault modeling and health monitoring, it significantly advances transparent checkpointing techniques, and it combines them with novel proactive approaches.

Dr. Alex Huang received an award from NSF for the ERC at NC State University. Funds were distributed to Drs. Mueller and McMillin for this segment of the project. The primary objective of this work is to define the system management requirements for the Distributed Grid Intelligence (DGI) within the Intelligent Energy/Fault Management (IEM/IFM). Requirements are determined from and generated with the System Theory, Modeling and Control (SMC) thrust for control algorithms, the Reliable and Secured Communication (RSC) subthrust for communications, and with computer science and engineering knowledge of system management of distributed resources. The results of the project are implementations of coordination algorithms for DGI within the testbed and requirements for the overall project and demonstration thrust.

Real-time embedded systems require known bounds on the worst-case execution time (WCET) of tasks. Static timing analysis provides such bounds, yet cannot keep pace with architectural innovations and hardware performance variation due to chip fabrication scaling. Instead of simulating execution, this work promotes actual execution in hardware to bound WCETs. This renders tedious hardware modeling unnecessary while guaranteeing correct behavior regardless of complexity or variation of hardware. The approach will be evaluated by FPGA synthesis to assess its feasibility and to validate a prototype. Advanced architectural features are studied in co-design space exploration to combine predictability and tight WCET bounds.

Today, high-performance clusters of shared-memory multiprocessors (SMPs) are employed to cope with large data sets for scientific applications. On these SMPs, hybrid programming models combing message passing and shared memory are often less efficient than pure message passing although the former fits SMP architectures more closely. For more information on this project check Dr. Mueller's Web Page

This work seeks to construct data streaming abstractions for clusters composed of graphics processing units (GPUs). The proposed work covers development of stream operations, such as abstract split/join operations, demonstrated through concrete text application, such as text search, for massive data processing requirements in clusters of GPUs. We propose to demonstrate the feasibility of GPU clustering for data-dependent programming problems with fixed input data as well as live data streams.

This work seeks to assess the viability of text mining on clusters composed of graphics processing units (GPUs). The proposed work covers development of optimized text search algorithms for GPUs and novel program paradigms for massive data processing requirements in GPUs. To this end, this work will leverage and build on previous research (at ORNL and NCSU) on: - customizing text search for the Nvidia CUDA programming platform for GPUs and - high-level prototyping of data-parallel programming models. We propose to demonstrate the feasibility of a flow-based abstraction for data-dependent iterative problems on a local CUDA cluster. Text mining will be in the center of our experiments that are to assess the suitability of massive data-centric iterative algorithms for this text search application domain.

This project addresses issues of adaptive, reliable,and efficient operating and runtime system solutions for ultra-scale high-end scientific computing with the following goals: (1)Create a modular and configurable Linux system based on the application / runtime requirements. (2)Build runtime systems that leverage the OS modularity and configurability to improve efficiency, reliability, scalability,ease-of-use. (3)Advance computer reliability, availability and serviceability management systems to work cooperatively. (4)Explore the use of advanced monitoring and adaptation to improve application performance and predictability of system interruptions. Our focus is on developing scalable algorithms for high-availability without single points of failure and without single points of control.

The focus of this project is to develop tool support to provide the ability for scientific programmers to inquire about scalability problems and correlate this information back to source code. Furthermore, we believe that tools should be able to suggest and evaluate optimizing transformations to alleviate these problems. This would constitute a significant improvement over current performance analysis practice. The key intellectual merit is in providing an automatic framework for detecting scalability problems and correlating them back to source code. We will experiment with our framework on the ASCI codes, which is intended to stress high-performance clusters.

Parallel I/O benchmarks are crucial for application developers, I/O software/hardware designers, and center administrators. However, currently there lack portable and comprehensive I/O benchmarks for high-end storage systems. We address this gap by proposing automatic generation of parallel I/O benchmarks. More specifically, we target the automated creation of application I/O benchmarks.

Energy infrastructure is a critical underpinning of modern society that any compromise or sabotage of its secure and reliable operation will have a prominent impact on people's daily lives and the national economy. Past failures such as the massive northeastern power blackout of August 2003 have revealed serious defects in both system-level management and device-level designs. This project proposes a hardware-in-the-loop reconfigurable system with embedded intelligence and resilient coordination schemes to tackle the vulnerabilities of the power grid. As a part of the collaborative research project, the research efforts at NCSU will focus on the threats to existing state estimation algorithms and their defenses.

Sensor networks are ideal candidates for a wide range of applications, such as monitoring of critical infrastructures, data acquisition in hazardous environments, and military operations. It is usually necessary to reprogram sensor nodes after they are deployed through wireless links. In this project, we will investigate secure, robust, and DoS-resilient remote program of sensor nodes through wireless links. We expect to develop three groups of fundamental techniques as a result, including secure and proactively robust encoding of binary code images, DoS-resilient mechanisms for authenticating binary images, and efficient and effective techniques for remote sensor programming in hybrid sensor networks.

Sensor networks are ideal candidates for a wide range of applications such as critical infrastructure protection. It is necessary to guarantee the trustworthiness and resilience of sensor networks as well as the sensing applications. The objective of this project is to develop practical techniques for building trustworthy and resilient sensor networks as well as instructional materials that facilitate the education of these techniques. The research activities are focused on practical broadcast authentication, trustworthy and resilient clock synchronization, and light-weight and collaborative intrusion detection in sensor networks, seeking effective integration of cryptographic techniques, application semantics, and other knowledge or constraints.

This project is to test and evaluate selected lab exercises developed by Dr. Wenliang Du at Syracuse University. Starting from Year 2, he will select 2-3 labs (on average) each year and use them in the course he teaches. After students finish each lab, Dr. Ning will ask students to fill out a survey questionnaire provided by the PI. The surveys will be sent back to Dr. Du, along with Dr. Ning's evaluation on students' performance

This project consists of three technical thrusts: (1) Thrust 1 -- new security architecture and services that better isolate different customers' workloads and enhance their trustworthiness; (2) Thrust 2 -- protection of management infrastructure against malicious workloads; and (3) Thrust 3 -- protection of hosted workloads from potentially malicious management infrastructure. The first thrust explores new opportunities to enhance the trustworthiness of virtual cloud computing against mutual threats between workloads as well as external security threats, while the last two address the service providers' security concerns for customers' workloads and customers' security concerns for the service providers, respectively.

The Web is evolving from a resource for finding/verifying facts, to one used to support complex problem solving and exploratory tasks in a variety of domains. Thus, the traditional search paradigm targeted primarily at fact-finding, and is predicated on users knowing what they want and how to search for it, is unsuitable for such situations. This project focuses on developing support for advanced query models that capture the iterative process typical of problem solving and exploratory tasks where at each step users may only be able to formulate vague queries. It proposes to develop semantic techniques for integrating information created across the multiple steps of such tasks as and also predict and recommend other potentially relevant and informative data that the user may have missed.

We propose to design and implement a passive network monitoring tool that will be able to characterize novel applications, provide early warning for security incidents and provide traffic measurements available to the community through a web-based NCSU/NCREN network map. More specifically our focus will be on: (1) conducting a series of detailed trials to capture data from different places across research and education campuses and identify the emerging traffic patterns; (2) using the trial results to deconstruct and analyze the intersecting networking and social distances of next generation of users (K-20); (3) incorporating these traffic patterns into a queueing model of a flow of IP packets with a view to calculating the end-to-end delay and packet loss rate. The results from this model will be used in cost-based parametric model in order to obtain the optimal network topology architecture.

There are many ways that computers attached to the Internet can be infected by malicious software. Virus and computer worm writers go to great pains to make their software difficult to detect. We have developed a method for identifying infectious software, before it succeeds, that is fast and very general. This method has been tested on a wide variety of software and shown to be effective. We propose now to automate this method to a greater degree. Essentially every method of detection relies upon human intelligence to guide the search for uniquely identifying properties of infectious software. We propose to instead use techniques of data mining that will automatically search for and evaluate such properties. A key characteristic that is exploited is that there are few true innovations in the design of infectious software, but many imitations or variations. Our method looks for the unvarying, common properties of such software. The benefit will be automated defenses that adapt rapidly to changing threats, including previously-unknown, or "zero-day", threats.

Internet Worms are software that propagate from computer to computer across the network, without intervention by or knowledge of users, for the purpose of compromising the defenses of those machines against unauthorized access or use. Worms have the property that they can spread very quickly to the vulnerable population of hosts, sometimes in only seconds, to achieve worldwide penetration. This speed allows them to bypass conventional methods of positive identification and human response.

This project develops a high-level notion of context that exploits the capabilities of next genera-tion networks to enable applications that deliver better user experiences. In particular, it exploits mobile devices always with a user to capture key elements of context: the user's location and, through localization, characteristics of the user's environment.

We propose to do a study to determine the minimal set of information that network domains and GOLEs must advertise about their resources to allow independent entities to perform coursegrain global path computation on interdomain links including technology (layer) adaptations. Although, we will initially focus on network resources, our proposed architecture will be extendable to include other GRID type resources, such as compute, storage, and high-end instruments. In this architecture, other entities will exist that mimic the behavior of the NRMs but for other resources (i.e. a Compute Resource Manager (CRM) will manage compute resources, and Instruments Resource Manager (IRM), will manage Instruments, much the same way NRMs function today).

Traffic quantization is a new approach to supporting per-flow functionality in packet-switched networks in an efficient and scalable manner. We propose the concept of tiered service to alleviate the complexity associated with supporting per-flow QoS: a quantized network offers a small set of service tiers, and each flow is mapped to the tier that guarantees its QoS. Research will consist of four components: develop novel quantized implementations of weighted fair queueing (WFQ); develop Linux implementations of quantized WFQ to validate the theoretical results; extend the quantization approach to multiple traffic parameters; and investigate efficient constraint-based routing algorithms for quantized traffic.

The focus of this project is to streamline pathways through which students receive an education that equips them with the computing tools necessary for them to serve as future computing leaders of society. To this end, we will assemble a community of individuals, each of whom is invested in their own unique way to revitalizing the undergraduate computing education. The community will involve faculty representatives from several academic departments and delegates from industry partner organizations, and will open up meaningful channels for dialogue to flow from industry to the university, leading to a more diverse, flexible workforce of computing professionals.

The objective of this project is to formulate a formal framework for a non-layered internetworking architecture in which complex protocols are composed from elemental functional blocks in a configurable manner, and to demonstrate its potential by developing proof-of-concept prototypes. We propose a new internetworking architecture that represents a significant departure from current philosophy. The proposed architecture is flexible and extensible so as to foster innovation and accommodate change, it supports network convergence, it allows for the integration of security features at any point in the networking stack, and it is positioned to take advantage of hardware-based performance-enhancing techniques.

Dr. Nagiza Samatova's joint work with NC State University and Oak Ridge National Laboratory (ORNL) will provide the interface between the two organizations aiming to collaboratively address computational challenges in the Scientific Data Management, and the Large-Scale Analysis of DOE-mission applications. (Supplement)

Dr. Nagiza Samatova's joint work with NC State University and Oak Ridge National Laboratory (ORNL) will provide the interface between the two organizations aiming to collaboratively address computational challenges in the Scientific Data Management, Data-Intensive Computing for Understanding Complex Biologicial Systems, Knowledge Integration for the Shewanella Federation, and the Large-Scale Analysis of Biologicial Networks with Applications to Bioenergy Production.

Terascale computing and high-throughput experiments enable studies of complex natural phenomena, on a scale not possible just a few years ago. With this opportunity, comes a new problem - the massive quantities of complex so data produced. However, answers to fundamental science questions remain largely hidden in these data. The goal of this work is to provide a scalable high performance data analytics technologies to help application scientists extract knowledge from these raw data. Towards this goal, this project will research and develop methodologies for addressing key bottlenecks, and provide proof-of-principle demonstrations on the DOE applications.

This project proposes to research and test a novel revenue-generating product and media recommendation system. We will architect and experimentally test a recommendation system that will facilitate the social process of preference formation and the diffusion of innovations by adapting the currently socially embedded architectures that facilitate these social processes as frictionless information architectures made for an increasingly internet-enabled society. Our social architecture will provide new product and media recommendations to consumers by acting as an automatic identifier and coordinator of opinion leaders, their discoveries, and consumers, modeling our process from breakthrough sociological models in consumer preference formation.

The goal of this workshop is to engage mathematical scientists and applications researchers to define a research agenda for developing the next-generation mathematical techniques needed to meet the challenges posed by petascale data sets. Specific objectives are to: * understand the needs of various scientific domains, * delineate appropriate mathematical approaches and techniques, * determine the current state-of-the-art in these approaches and techniques, and * identify the gaps that must be addressed to enable the effective analysis of large, complex data sets in the next five to ten years.

Physical science applications such as nanoscience, fusion science, climate and biology generate large-scale data sets from their simulations and high throughput technologies. This necessitates scalable technologies for processing and analyzing this data. We plan to research and develop advanced data mining algorithms for knowledge discovery from this complex, high-dimensional, and noisy data. We will apply these technologies to DOE-mission scientific applications related to fusion energy, bioenergy, understanding the impacts of climate extremes, and insider threat detection and mitigation.

The solution of linear diophantine inequalities is at the heart of problems in many areas of mathematics. In recent years there has been an evolution of mathematical techniques and software tools to solve them. Nevertheless, many important problems remain beyond the scope of these methods. The focus of this proposal is the enumeration of combinatorial structures defined by linear diophantine constraints. A primary goal is the development of mathematical techniques for diophantine enumeration that are able to exploit the symmetry, recursion, and structure appearing in combinatorial families.

Mobile ad-hoc networks (MANETs) have been the focus of significant research activity in the past decade. Thousands of algorithms and protocols for MANETs have been proposed, evaluated and compared. One of the defining characteristics of MANETs is their mobility. We propose to develop and evaluate a hybrid mobility model that is relatively easy to generate and, at the same time, produces realistic mobility traces, that in turn, result in meaningful simulation results for MANET simulations. The proposed model has the desirable characteristics that it is customizable to match any scenario, while allowing the users to vary key parameters.

This project will develop policy-based service governance modules for the Oceanographic Observatories Initiative (OOI) Cyberinfrastructure. The main objectives of the proposed project include (1) formulating the key conceptual model underlying the patterns of governance; (2) formalizing "best practices" patterns familiar to the scientific community and seeding the cyberinfrastructure with them; (3) understanding user requirements for tools that support creating and editing patterns of governance

This project will develop a computational approach to trust geared toward enhancing the quality of information in tactical networks. In particular, this project will develop a trust model that takes into account various objective and subjective qualities of service as well as the social relationships among the parties involved in a network who originate, propagate, or consume information. The proposed approach will build an ontology for quality of information and its constituent qualities, and will expand existing probabilistic techniques to multivalued settings. The project will develop a prototype software module that realize the techniques for producing trust assessments regarding the information exchanged.

This project will develop policy-based service governance modules for the Oceanographic Observatories Initiative (OOI) Cyberinfrastructure. The main objectives of the proposed project include (1) formulating the key conceptual model underlying the patterns of governance; (2) formalizing "best practices" patterns familiar to the scientific community and seeding the cyberinfrastructure with them; (3) understanding user requirements for tools that support creating and editing patterns of governance.

Tool use is an agent's manipulation of objects in the environment to transform the interaction between the environment and the agent's sensors or actuators such that its goals are more efficiently achieved. We propose four core capabilities for a habile agent: the ability to generalize existing effectivities to those provided by a tool under the agent's control; the ability to perform detailed internal simulations based on hypothetical application of tool-using abilities; the use of symmetry in recognizing opportunities for tool use; and the use of a general image-schematic representation to control tool-using behavior.

NCSU graduate student will develop a machine learning approach to linking extreme atmospheric ridging events with extreme surface temperatures, employing a Gaussian Process (GP)-based predictive analysis tool that leverages the predictive value of spatial and temporal correlations and builds on ORNL’s past success in spatial classification, temporal change prediction, and parallelizing GP for large spatiotemporal extents.

With the increasing volume and complexity of data produced by ultra-scale simulations and high-throughput experiments, understanding the science is largely hampered by the lack of comprehensive, end-to-end data management solutions ranging from initial data acquisition to final analysis and visualization. The SciDAC-1 Scientific Data Management (SDM) Center succeeded in bringing an initial set of advanced data management technologies to DOE application scientists in astrophysics, climate, fusion, and biology. Building on our early successes, we will improve the SDM framework to address the needs of ultra-scale science.

In partnership with industry and faculty from across the country, this project will develop a transformative approach to developing the communication abilities (writing, speaking, teaming, and reading) of Computer Science and Software Engineering students. We will integrate communication instruction and activities throughout the curriculum in ways that enhance rather than replace their learning technical content and that supports development of computational thinking abilities of the students. We will implement the approach at two institutions. By creating concepts and resources that can be adapted by all CS and SE programs, this project also has the potential to increase higher education's ability nationwide to meet industry need for CS and SE graduates with much better communication abilities than, on average, is the case today. In addition, by using the concepts and resources developed in this project, CS and SE programs will be able to increase their graduates' mastery of technical content and computational thinking.

Over the past seven years, we have collected data on 250 high achieving young women, ages 11-20 for an intervention project and an ITWF project. High achieving is defined as those girls selected/electing to take Algebra 1 in middle grades, putting them on track to take calculus in high school. The proposed research provides an opportunity to extend and redirect the current database for a new study. By 2009 we expect to have 100 longitudinal records to inform post-undergraduate analysis, 200 longitudinal records to inform the undergraduate analysis, and 300 longitudinal records to inform the high school analysis.

NC State's Virtual Computing Environment (VCE) is a computing environment in use at NC State, which is very attractive to the NC Community College System (NCCCS). It is expected to help them deliver educational services in a manner both superior to and with less resource expenditure than their current computer labs. After investigating the VCE the NCCCS put it into their plans. They then requested and received a Legislative appropriation to fund a production pilot using NC State's resources. The services provided under this agreement by NC State will support pilots at a number of NC Community College campuses.

Drastic improvements in the speed of 3D graphics rendering hardware have been accompanied by even more drastic increases in the size of displayed models. Researchers trying to display these models have been forced to reduce display speed and interactivity or reduce the fidelity of the displayed views of their models. What are the best methods for preserving visual fidelity as model complexity is automatically reduced? What is the most effective way of striking the display speed vs. visual fidelity compromise? Our research will take examine these questions, resulting in prototype systems and investigations of their effectiveness with user studies.

Participants of this interactive designed technology hothouse for undergraduate researchers and designers will work with computer science and design faculty and industry on projects spanning artificial intelligence, graphics, visualization as well as visual and interactive design. Sample projects include: advanced AI for interactive narratives and games; including camera control, and story planning and level design; automated tours through virtual and visualized environments; visualizing streaming news feeds using swarming sprites, and interactive, ambient display walls; PDA-based art installations, and real-world navigation tools. Students will gain the cross-disciplinary and cross-cultural teamwork and communications skills so important in designed technology research and industry.

Real-time systems often exhibit some level of non-determinism with regard to defect observability. Testers frequently must run tests multiple times in order to have some assurance that the test actually passes. This costs industry significant time and effort. No one has yet studied these defects to understand why they exhibit non- determinism and what techniques or tools can be used to better observe the software and control the non-determinism.

The work outlined in this proposal focuses on the use of security metrics to prioritize risk-based software engineering for security. Specifically, the objective of this proposal is to build a predictive model based upon security metrics obtained from code artifacts, inspections, and testing to highlight vulnerability-prone and attack-prone components for the risk-based prioritization of re-design, inspection, and testing efforts. The model will be built and validated through detailed analysis of industrial code and data of CACC members: automated static analysis alerts, other static metrics to be determined, inspection records, testing records, and customer-reported problems.

Our nation's critical infrastructure demands that our current and future IT professionals have the knowledge, tools, and techniques to produce reliable and trustworthy software. The objective of this research is to extend, validate, and disseminate a software development practice to aid in the prevention of computer-related disasters. The practice is based upon test-driven development (TDD), a software development technique with tight verification and validation feedback loops. The proposed work extends the TDD practice and provides a supportive open-source tool for explicitly situating security and reliability as primary attributes considered in these tight feedback loops.

We propose to build, evolve, and validate a statistical prediction model whereby security-related ASA alerts from one or more tools and other software metrics are used to predict the actual overall security of a system. Our research involves collecting and analyzing a significant amount of data on software programs including security-related ASA alerts and actual security vulnerabilities and exploits, based upon inspections, testing failures, field failures, and reported exploits.

National efforts are focused on improving medical quality and reducing medical costs by implementing standardized electronic medical record keeping. However, rural health care providers and those who have small offices may not have the financial resources or expertise to purchase and maintain necessary expensive hardware and software applications which are necessary to join in this move toward electronic medical records (EMRs). We propose that the electronic medical records (EMR) application needs of rural and small-practice ambulatory health care providers be satisfied via open source EMR applications.

The proposed project is an extension of the BPC-A STARS Alliance. Our goal continues to be to broaden participation in computing (BPC) through the STARS Alliance, a Southeastern partnership to implement, institutionalize and disseminate effective practices for recruiting, bridging, retaining and graduating women, under-represented minorities and persons with disabilities into computing disciplines. With this proposal, additional academic institutions will join the STARS Alliance. NC State will continue to work with our own Student Leadership Corps, and with our demonstration and evaluation tasks, and it will act as the initial mentor for the new regional institutions.

Our goal is to broaden participation in computing by developing a Southeastern partnership to implement, institutionalize and disseminate effective practices for recruiting, bridging, retaining and graduating women, under-represented minorities and persons with disabilities into computing disciplines. The Alliance will implement a comprehensive set of activities to provide high-quality opportunities to a large audience of post secondary students, including a Student Leadership Corps, pair programming, a Web portal, and Marketing and Careers Campaign, summer REUS, and a STARS Celebration Conference and Exchange.

This ITWF award to NCSU, NCA&T, and Meredith College will support a three-year study of the collaborative aspects of agile software development methodologies. We believe the collaboration and the social component inherent in these methodologies is appealing to people whose learning and training models are socially oriented, such as some minority groups, women, and men. The project?s objective is to perform extensive, longitudinal experimentation in advanced undergraduate software engineering college classes at the three institutions to examine student success and retention in the educational and training pipeline when the classes utilize an agile software development model.

Developer testing has been widely recognized as an important, valuable means of improving software reliability. However, manual developer testing is often tedious and not sufficient. Automated testing tools can be used to reduce manual testing efforts. This project develops a systematic framework for cooperative developer testing to enable effective, synergetic cooperation between developers and testing tools. This framework centers around test intentions (i.e., what testing goals to satisfy) and consists of four components: intention specification, test generation, test abstraction, and intention inference. The project also includes integrated research and educational plans.

Testing is essential for database applications to function correctly and with acceptable performance when deployed. In practice, it is often necessary for a database software vendor to test their software completely before selling or integrating their package to the database owner. In this proposal, we focus on two bottlenecks in database application testing: functional testing, which is to test whether the applications can perform a set of predefined functions correctly, and performance testing, which is to test whether the applications can function with acceptable performance when deployed.

Improving software quality is becoming an important yet challenging task in software development, especially for those mission-critical or safety-critical software systems. Many software defects related to correctness, security, and robustness are caused by the incorrect usage of system or application programming interfaces (APIs). We propose to develop new approaches for mining API properties for static verification from the API client call sites in existing code repositories, and then detect violations of these mined API properties to find defects in code repositories.

Security policies such as access control and firewall policies are one of the most fundamental and widely used privacy and security mechanisms. Assuring the correctness of security policies has been a critical and yet challenging task. In this proposal, we propose to develop a uniform representation of security policies across application domains such as XACML access control policies and firewall policies, and a set of novel techniques for testing and verification of both static and stateful policies based on the uniform representation.

Existing techniques for the design of tests typically target at designing tests for testing the current version of a software system. Designing tests for evolving software has rarely been explored but is of great importance in advancing science of design. The goal of this research is to address the test-suite augmentation problem by defining novel techniques for: (1) determining whether an existing regression test suite adequately exercises the changes between two versions of a software product; and (2) providing automated support for designing and developing test cases that target the software changes not adequately exercised by the existing tests.

An increasing number of applications are set within narrative-oriented 3D virtual worlds. Current research on the generation of activities within these worlds holds the promise of tailored experiences customized to individual users? needs. The work described in this project seeks to expand the computational models of narrative being used to AI researchers, specifically to explore formal, plan-based models of actions to create stories that demonstrate complex conflict, rising action, dynamism and intentionality. The work will proceed both formally and empirically, with models being developed motivated by work from narrative theory and cognitive psychology and evaluated using experimental methods.

Through innovative application of computational thinking, this project will build the necessary cyber infrastructure to provide the next generation platform for multi-disciplinary and multi-agency collaboration in crime scene investigation (CSI). Since Daubert v. Merrell Dow Pharmaceuticals, CSI is both a highly visual and quantitative analysis characterized by a time-sensitive need to gather, organize, analyze, model, and visualize large, multi-scale, heterogeneous and context-rich data. CSI is also characterized by a fundamental need for rapid coordination and data translation across disciplines, agencies and levels of expertise as crime scenes are processed, reconstructed, solved and ultimately prosecuted over time, often critically in front of lay-people comprising a jury. Current methods of CSI are hindered by a lack of cyber infrastructure and protocols for virtual access to expertise and inadequate repositories of key data. From a computational e-Science perspective, forensic science is ripe for revolution through the development of a cyber infrastructure that will provide both new core data resources and collaboration capabilities in CSI for analysis and communication. Through remote access to data, tools and experts, as well as holistic integration of diverse data streams to virtually reconstruct and preserve actual crime scenes, the application of computational thinking to CSI will enable meta-analysis of evidentiary data and transform research and education for CSI professionals, legal professionals, forensic scientists, and K-20 students. The transformative research goal of this project is to develop a pioneering platform for interdisciplinary, cyber-enabled crime reconstruction through innovative methodology and engagement (IC-CRIME). The IC-CRIME platform will enable collaborative engagement in a 3D virtual reconstructed crime scene that incorporates multi-layer, scale-variant data and objects, including new critical data resources. The proposed cyber infrastructure will allow novice users to embed, interact with and analyze multi-layered data and objects within reconstructed crime scene in such a way that provides lucid spatial insight for users while simultaneously preserving quantitative geospatial relationships among evidentiary components as meaning is gleaned from data. The transformative educational goal of this project is to develop a team-based inter-disciplinary educational tool for professionals and K-20 that provides experiential, problem-based learning opportunities in a data-intensive virtual environment.

Researchers in science education, computer science, distance education, and the NC Department of Public Instruction are partnering with the Kenan Fellows Program to harness the untapped potential of inexpensive, online multi-user competitive simulation software in improving the science achievement and IT skills of NC's grade 6-12 students. Over three years, teacher participants will learn how to use this technology to increase student science achievement and motivation to enter IT-related science careers. Intellectual merits of the project entail rigorous assessment and evaluation of how these environments most effectively improve the IT skills and science content mastery of students.

Online social networks not only greatly expand the scale of people's social connections, but also have the potential to become an open computing platform, where new types of services can be quickly offered and propagated through existing social structures. Mechanisms for trust management of privacy protection are integral to the future success of online social networks. In this project, we develop theoretical and practical techniques for the management of trust and privacy for social networks. Some of the innovative expected results include a formal trust model and trust policy languages for social networks, privacy preserving feedback management, and graph anonymization techniques for the sharing of social network data.

The correct and reliable operation of an information system relies not only on users' capabilities, but oftentimes on users' obligations. The management of obligations in security policies imposes significant challenges since obligations bear different properties from access control. This project develops a comprehensive framework for the management of obligations, including obligation modeling, specification, analysis, monitoring and discharges. Though the framework is formal in nature, and is designed on purpose to be general, the evaluation of its usefulness and effectiveness is firmly grounded on real applications, in particular, in the context of cross-domain data sharing systems and privacy policy enforcement systems.

This is a grant request to support the series of workshops on security in emerging areas that are affiliated with the 2009 ACM Conference on Computer and Communications Security (CCS), and will be held on November 9 and 13, 2009 in Chicago, IL. CCS workshops are active forums for researchers to form focus groups, discuss and collaborate on emerging and critical security problems, and disseminate fresh, revolutionary (and sometimes even controversial) ideas. These workshops also serve as natural venues to bring together researchers from multiple disciplines to address security issues in specific domains, such as health care, cloud computing and national critical infrastructures.