BURGs - Broadening Undergraduate Research Groups
To recruit, retain, and develop a diverse undergraduate research cohort to: cultivate an inclusive culture and work environment that reflects our community; empower individuals to meaningfully contribute computer-driven solutions in service to society; and promote the progress of computer systems research.
Computer Systems Genome
Since the dawn of computing, the world has tracked system performance. Yet, computer system performance data is still primarily siloed by benchmark, system, or system component. The Mission of the Computer Systems Genome Project (CSGenome) is to conduct the first scientific effort to catalog the lineage of computer system performance over time to enable knowledge discovery and further understanding of the impact of computing innovations on transformative technologies, the knowledge-based economy, and societal change. Browse visualizations and access repository at csgenome.org.
Undergraduate Students: CSGenome About
Equitable access to the Internet can help bridge digital divides, and supporting infrastructure deployment to un- and underserved communities is a policy priority for state and national governments. High-quality data on the current availability of broadband is critical for making effective policy decisions and for helping local communities determine how best to use their limited resources to improve their own access to the Internet. This project is building a toolset for broadband mapping and network planning for un- and underserved communities throughout the US, with a particular focus on Tribal communities.
Faculty Lead: Shaddi Hasan
Computing Performance for Machine Learning
PyTorch enables the use of OpenMP allowing the use of libraries that exploit parallelism on multicore processors and multi-processor systems. Specifically. The goal of this project is to explore the performance implications of OpenMP multithreading in PyTorch inference codes. Specifically, the project conducts rigorous performance characterization, analytical modeling, and experimentation to investigate how the different forms of parallelism available in inference (intra-operation and inter-operation parallelism) can best be exploited by OpenMP, so that inference performance is maximized.
Faculty Lead: Dimitrios Nikolopoulos
Security and Privacy in Augmented Reality Systems
This project aims to develop a defense in augmented reality (AR) systems against side-channel attacks where the attacker is able to access and analyze the performance counters data as a background app to identify an application being launched by the victim user. Specifically, we will leverage eye gaze data available on the AR headset devices (e.g., Microsoft HoloLens 2) to design a countermeasure that can determine which app the user is focused on in real-time, thus not permitting any background apps (which are potentially malicious) to access the performance counters data.
Faculty Lead: Bo Ji
Cyber attacks have emerged as a pressing issue in recent years, causing substantial financial and operational damage to both businesses and individual users. Enhancing the protection is therefore critical. This project aims to comprehensively study the behaviors of ransomware across diverse families. Utilizing execution traces and machine learning models, the study aims to differentiate between benign and malicious activities.
Faculty Lead: Daphne Yao
Graduate Student Lead: Wenjia Song
Kirk W. Cameron, Ph.D.
Kirk W. Cameron, Professor of Computer Science and a Research Fellow in the College of Engineering at Virginia Tech, is Co-Founder and Executive Director of BURGs. The central theme of his research is to improve power and performance efficiency in high performance computing (HPC) systems and applications. Accolades for his work include NSF and DOE Career Awards, IBM and AMD Faculty Awards, and being named Innovator of the Week by Bloomberg Businessweek Magazine. In 2017-2018, Prof. Cameron held a Distinguished Visiting Fellowship at Queen’s University Belfast from the U.K. Royal Academy of Engineering. He is Director of the stack@cs Center for Computer Systems and the Scalable Performance Laboratory. Kirk is overall VarSys project lead PI and heads the HPC Systems Team.
Margaret Ellis, M.S.
Margaret Ellis, Associate Professor of the Practice of Computer Science in the College of Engineering at Virginia Tech, is Co-Founder and Director of BURGs. As a passionate teacher of key courses such as problem solving and data structures, her work lies at the intersection of computer science research, computer science education, and challenges of diversity and inclusion in the field. Among other accolades and accomplishments, she was awarded the College of Engineering Dean's Excellence in Teaching Award in 2020 and Engineering Inclusive Teaching (EIT) Inclusive Educators Award by Women in Engineering ProActive Network (WEPAN) in 2015.
Godmar Back, Ph.D.
Godmar Back, Associate Professor of Computer Science in the College of Engineering at Virginia Tech, is Co-Founder and Technical Advisor of BURGs. His research primarily focuses on computer operating systems while his interests also include collaborative work in computer science education and the library sciences. Accolades for his research include the NSF Career Award. For his educational efforts, he was awarded the 2018 ICPC Foundation Coach Award recognizing mentoring excellence resulting in 5 teams in 6 years competing in the World Finals. He leads the Systems Software Laboratory at Virginia Tech.
BURGs is affiliated with stack@cs Center for Computer Systems
Photos provided by Peter Means