Our research aims at evaluating the performance of contemporary multi-core platforms, the quad-core AMD Barcelona and the dual-core Intel Woodcrest, for scientific applications. We used the High Performance Computing Challenge (HPCC) benchmark suite, which contains test cases for quantifying performance of the memory and communication sub-systems. Analyzing the performance of multi-core devices allows us to identify system parameters and configurations that could yield optimal performance. Since these devices are building blocks of the fastest supercomputing systems, the insights into performance behavior provide guidance for designing future high performance computing resources. Our results demonstrate that without multi-core aware optimization the achievable performance decreases as the number of cores per socket increases.
This paper describes a novel multimedia tool to facilitate visual assessment of Functional Magnetic Resonance Imaging (fMRI) activation patterns by human experts. A great effort is placed by radiologists and neurologists to present a consistent methodology to provide assessment for brain activation map images. Since each radiologist has his own way to perform the visual analysis on the images and present the findings, rating a large and heterogeneous group of images is a hard task. Although this tool is focused on assessing fMRI activation patterns related to brain language network paradigms, the tool can be extended to other brain activation maps, such as motor, reading, and working memory. Moreover, the same tool can be used for assessing images acquired using different recording modalities as long as these images are saved in standard image formats such as JPEG, BMP, or PNG. The use of this tool is independent of the methodology used to generate the brain activation map, which can be done using specialized software tools such as Statistical Parametric Mapping (SPM) or fMRI Software Library (FSL). The main benefits of using this tool for brain activation image scoring are the systematic approach for rating the activation maps, the automatic descriptive statistics applied to the results and the reduction of assessment time from several minutes to seconds. For each study, the proposed system presents the activation pattern image, based on which the rater is asked to indicate the level and type of activation observed in general, and in specific on the following areas: frontal, temporal, and supplemental motor area.
In Delay Tolerant Networks (DTN), the performance heavily depends on the association between mobile users, which decides how frequently the users meet with each other. It is therefore very important to understand the association. In this paper, by analyzing a number of wireless network traces, we verified the exponential distribution of the association of mobile users, which means that most mobile users are not closely associated. Furthermore, we examined the distribution of the inter-connectivity time of those mobile users more closely associated.
Grid computing promises improvements in collaboration. This includes sharing of computational resources as well as improved collaboration amongst professionals of different areas of expertise. Several mature software applications are available for simplifying the deployment of an arbitrary grid. In this work, we share our experience with some of these applications for collaboration between a consortium of hospitals and our research lab, which specializes in neuroscience and image processing applications. We explain the suitability of the Grid tools through extensions and enhancements made to an existing Grid Computing Software platform and the visualization mechanisms of the display wall. This paper describes our grid computing prototype infrastructure, which uses Globus Toolkit-4 (GT4) and third-party components.
In order to safeguard an organization's networked assets, a network administrator must decide how to harden the network. To aid the decision-making process, network administrators may use attack graphs, which, through analysis, yield network hardening suggestions. A critical drawback of currently available analyses is the lack of consideration for the network administrator's defense budget. We overcome this shortcoming by modeling the problem of choosing security measures given a finite budget as a combinatorial optimization problem. We call this problem the Security Measures Choosing Problem (SMCP). Dynamic programming is used to provide optimal solutions.
Statistics for underrepresented minority groups and women continue to show low numbers in enrollment and rates of retention in academic computer science programs. A new approach to increase student interest in computer science in a first year program is introduced. Laboratory modules for an introductory programming course have been developed at the University of Alabama with the goal to increase student motivation and understanding of fundamental programming concepts. The course utilizes robots and Alice, a 3D graphical programming environment. The drag and drop interface of Alice allows students to program real robots using instructions that correspond to statements of programming languages such as Java, C++, and C#. Students gain programming experience that is transferable to upper level courses by engaging in a stimulating and less frustrating environment using Alice interfaced with robots.
In this work, we present some preliminary ideas about the development of an approach to mining different perspectives of a business process. Process mining, to date, has been narrowly concerned with mining the control-flow of a business process. There are very few process mining algorithms aimed at mining different business process perspectives. We believe one of the primary reasons for the paucity of process mining algorithms in perspectives other than control-flow is that there has been no general definition of what a business process perspective is. With this work, we provide a formal and general definition of a business process perspective, and present an approach to mine other business process perspectives using this definition.
This paper describes our successful leveraging of the National Security Agency/Department of Homeland Security Center of Academic Excellence in Information Assurance program and Auburn University's highly successful partnership with three (Historically Black College and Universities (HBCU)) universities through the National Science Foundation's Scholarship for Service Program. This paper will describe this ongoing and highly successful program that has been publicly praised by the National Science Foundation as "a model for innovative collaboration and community building. It demonstrates how majority institutions and minority serving institutions can effectively build mutually beneficial partnerships which will increase diversity in the information assurance community."
This Partnership for International Research and Education (PIRE) is a 5-year long project funded by the National Science Foundation that aims to provide 196 international research and training experiences to its participants by leveraging the established programs, resources, and community of the Latin American Grid (LA Grid, an international academic and industry partnership designed to promote research, education and workforce development at major institutions in the USA, Mexico, Argentina, Spain, and other locations around the world). In return, PIRE will take LA Grid to the next level of research and education excellence. Top students, particularly underrepresented minorities, are engaged and each participant will receive multiple perspectives in each of three different aspects of collaboration as they work with (1) local and international researchers, in (2) academic and industrial research labs, and on (3) basic and applied research projects. PIRE participants will engage not only in computer science research topics focused on transparent cyberinfrastructure enablement, but will also be exposed to challenging scientific areas of national importance such as meteorology, bioinformatics, and healthcare. During the first year of this project, 18 students out of a pool of 68 applicants were selected; they participated in complementary PIRE research projects, visited 7 international institutions (spanning 5 countries and 4 continents), and published 9 papers.
Although some work has been done to better predict the outcome of sporting events, it has focused on mainstream sports such as football and has typically employed forecasting or machine learning techniques. This work focuses on the sport of cross-country, and uses feature selection and evolutionary computation to better predict National Meet results. Feature Selection is utilized to find the most optimal feature set and a Particle Swarm Optimizer (PSO) to find the most optimal weight set. The best results are attained using the PSO, with an improvement over the current system of 2.5% for Women and 0.3% for Men.
Recent work has focused on removing explicit network identifiers (such as MAC addresses) from the wireless link layer to protect users' privacy. However, despite comprehensive proposals to conceal all information encoded in the bits of the headers and payloads of network packets, we find that a straightforward attack on a physical layer property yields information that aids in the profiling of users. In this paper, a statistical technique is developed to associate wireless packets with their respective transmitters solely using the signal strengths of overheard packets. Through experiments conducted in a real indoor office building environment, we demonstrate that packets with no explicit identifiers can be grouped together by their respective transmitters with high accuracy. We next show that this technique is sufficiently accurate to allow an adversary to conduct a variety of complex traffic analysis attacks. As an example, we demonstrate that one type of traffic analysis--a website fingerprinting attack--can be successfully implemented after packets have been associated with their transmitters. Finally, we propose and evaluate techniques that can introduce noise into the measurements of such physical layer phenomena to obfuscate the identifiers derived from them.
Genomics has reached the stage at which the amount of DNA sequence information in existing databases is quite large. Synthetic biology is now using these databases to catalog sequences according to their functionality thus creating a system of standard biological parts. Flexible tools are needed which both permit access and modification to that data and also allow one to perform meaningful, intelligent manipulation. A Platform-Based Design approach views genetic information as having a particular functionality and assembles platforms (collections of DNA elements) to perform this functionality. Specifically this paper presents the Clotho toolset which uses these concepts to create a complete design environment for standardized biological parts.
Software security concerns are frequent, widespread, and with potentially harmful consequences. We believe that security concerns should not only be specified as part of software requirements, but should also be supported during later stages of development (architecture, design, implementation, testing, and maintenance). This paper focuses on security requirements and the support available for them past their creation. As part of ongoing research we surveyed 12 approaches to security requirements engineering and identified the level of support each approach provides on a variety of areas related to later stages support. We show that support for security requirements after they are specified is lacking at best, creating opportunities for significant improvement and further research in this area.
In this paper, we describe the innovative approach of using formal computational models to guide design efforts and evaluate software interfaces for usability. Decades of work in the field of HCI looked to using task analyses, live user testing and GOMS modeling techniques to help design for usability. While these methods proved helpful, guidance came after the fact. Only after the prototype interface had been designed were the traditional techniques useful in evaluating the interface. This research seeks to be innovative in two ways. First, it seeks to examine techniques that can directly impact design at the very first stages where it is most critical. Second, it demonstrates the utility of a not so often used approach, that of using formal computational models to impact design. In this research, formal computational models were used to map sequences of operations of an activity or task. A candidate interface was then designed around the formal model's recommendations for what would be most facilitative for this activity. The result was an interface that performed optimally per model predictions. Additionally, live user testing was employed and empirical observations verified and reinforced the model predictions.
The Commonwealth Alliance for Information Technology Education (CAITE) is an alliance of 15 Massachusetts public campuses that focuses on community colleges because of their role as a gateway to careers and further higher education for underserved populations. CAITE outreach extends into four regions that have high percentages of students who are under-represented in the knowledge and innovation economy. Building on two years' experience and extensive data, we are strengthening our efforts to develop nurturing educational pathways and to ensure that students are adequately prepared to enter them. CAITE can serve as a transferable model for statewide collaborations.
This work seeks to contribute to software development education by motivating the use of engaging in-class and laboratory assignments. Ideally, these assignments should involve considerable student buy-in and should also evolve throughout the course to mimic real-world software development. Prior research is discussed, as well as several specific examples from two introductory programming classes. The ultimate contribution is a convincing argument to spend the extra effort to design better student projects.
In this paper, we describe our proof of concept system that uses genetic algorithms to generate choreography for the waltz, a ballroom dance. We detail the representation of the dance steps and sequences our system manipulates, and our design of the fitness function to guide the algorithm. Preliminary results show that we have successfully incorporated several rudimentary choreography principles and that our system thus can generate effective waltz sequences. There are many potential areas of future development for our system, such as extending it to generate more sophisticated choreography for a variety of ballroom dances.
According to Computing Research Association, during each year between 2003 and 2007, fewer than 3% of the US's Ph.D.s graduates in computer science and computer engineering were Hispanic or African American and fewer than 20% were women. Such an under-representation precludes the benefits of diversity in computer sciences research and industry and consequently compromises the competitiveness of the US economy. It is therefore imperative that undergraduate institutions introduce students from these groups to research at an early stage of their academic careers and to provide them with the tools necessary for success in graduate school. The School of Computing and Information Sciences (SCIS) at Florida International University (FIU) has been working to strengthen the pipeline of underrepresented students to graduate work in computer science by hosting an NSF sponsored Research Experiences for Undergraduates (REU) site for the past three years. Our REU site has hosted 30 undergraduate students, 23 of them were underrepresented including 8 females, 16 Hispanics, and 4 African Americans, who published 13 technical papers. Six of the ten students who have already graduated, have started their graduate studies.
If we are to attract more women and minorities to computing we must engage students at an early age. As part of its mission to increase participation of women and underrepresented minorities in computing, the Increasing Student Participation in Research Development Program (INSPIRED) conducts computing academies for high school students. The academies are designed to increase students’ knowledge of and interest in computing and to encourage females and minorities to participate in computing. INSPIRED academies differ from others in several ways. They are relatively easy to organize and require relatively few resources; they focus on computing concepts and object-oriented programming; they expose students to successful female and minority computer scientists; and they actively engage university students from underrepresented groups to organize, coordinate, teach, and help assess the academies. This not only provides role models for the high school students but also helps engage the university students and promote their professional development. Our assessment results show that high school student participants have gained significant knowledge and interest in computing through participation in the academies. This article describes the organization, coordination, content, and assessment of the academies, along with suggestions for those who would like to design academies like these. It also discusses how to prepare university students for their roles in the academies and how their participation has helped in their professional development. It includes pointers to sites from which the instructional and assessment materials can be downloaded for those who wish to replicate or adapt these materials.
We propose a jointly opportunistic source coding and opportunistic routing (OSCOR) protocol for correlated data gathering in wireless sensor networks. OSCOR improves data gathering efficiency by exploiting opportunistic data compression and cooperative diversity associated with wireless broadcast advantage. The design of OSCOR involves several challenging issues across different network protocol layers. At the MAC layer, sensor nodes need to coordinate wireless transmission and packet forwarding to exploit multiuser diversity in packet reception. At the network layer, in order to achieve high diversity and compression gains, routing must be based on a metric that is dependent on not only link-quality but also compression opportunities. At the application layer, sensor nodes need a distributed source coding algorithm that has low coordination overhead and does not require the source distributions to be known. OSCOR provides practical solutions to these challenges incorporating a slightly modified 802.11 MAC, a distributed source coding scheme based on network coding and Lempel-Ziv coding, and a node compression ratio dependent metric combined with a modified Dijkstra's algorithm for path selection. We evaluate the performance of OSCOR through simulations, and show that OSCOR can potentially reduce power consumption by over 30% compared with an existing greedy scheme, routing driven compression, in a 4 times 4 grid network.
