Systems and Networking at Khoury College of Computer Sciences
Developing the knowledge essential to making computer systems and networks trustworthy, reliable, and secure
Systems and Networking research at Khoury College of Computer Sciences focuses on building knowledge and practices to identify the vulnerabilities in the digital technology that runs our world. Research is focused on finding new ways to look at all levels of these systems, network architecture, protocols, services, and security and privacy, finding ways to detect and respond to attacks and building stronger systems in the future.
Attacks on computer systems and networks are constant, but AI has increased their intensity and frequency as well as the challenges of mitigating them. At the same time, AI has also opened exciting research vistas; Khoury College faculty and students are developing cutting-edge, AI-based approaches for detecting vulnerabilities and using automation to increase accuracy, speed, and scale of response, as well as making AI systems themselves more resistant to hacks.
New risks, innovative solutions
One of the critical impacts of Systems and Networking research is exploring how to keep intermediary networks that deliver web content securely. Unlike the early days of the web, when individual users connected directly to a page on a company’s server, the majority of web content now comes through companies that serve as intermediaries, helping deliver vast amounts of data and balance the demand on the networks. Understanding how to keep these services safe has a direct impact on keeping the internet functional and secure.
The internet is no longer just on computers or phones; it’s part of numerous household and personal devices. This brings new risks that defy simple fixes. Khoury College researchers are developing solutions to make these devices and the networks that serve them secure.
Sample research areas
- Cloud computing
- Cyber-physical systems
- Distributed systems
- High performance computing
- Internet measurement
- Mobile/wireless systems
- Networking
- Parallel computing
- Security and privacy
- Storage systems
- Privacy
- Cryptographic risk
- Cybersecurity
Khoury researchers: At the forefront
Current project highlights
Systems and Networking faculty and students participate in the International Secure Systems Lab (iSecLab), a collaboration with academic institutions in the US and abroad focusing on web security, malware analysis, cybercrime, and vulnerability analysis.
The Northeastern University Systems Security Lab conducts research on a wide range of practical security problems, with recent projects comparing cybersecurity and privacy on MacOS and Windows.
Recent research publications
No Root Store Left Behind
Authors: James Larisch, Waqar Aqeel, Taejoong Chung, Eddie Kohler, Dave Levin, Bruce M. Maggs, Bryan Parno, Christo Wilson
The paper discusses the problem of inconsistent trust levels for digital certificates across different computer systems, something that can cause security vulnerabilities. It proposes a new system to manage these trust levels more effectively, ensuring that all systems have the same up-to-date information about which certificates should be trusted and for what purposes.
Implementation-Oblivious Transparent Checkpoint-Restart for MPI
Authors: Yao Xu, Leonid Belyaev, Twinkle Jain, Derek Schafer, Anthony Skjellum, Gene Cooperman
This research developed a new tool to automatically save the progress of complex computer programs that run across multiple computers using message passing interfaces (MPI). This tool improves system reliability by allowing programs to resume from a saved state after unexpected interruptions, such as power failures or software crashes. It also enhances network efficiency by reducing the need to re-compute large amounts of data.
Problematic Advertising and its Disparate Exposure on Facebook
Authors: Muhammad Ali, Angelica Goetzen, Alan Mislove, Elissa M. Redmiles, Piotr Sapiezynski
This research into social networks and human computer interactions (HCI) found that certain groups of people, such as older adults and minorities, are more likely to see misleading ads on Facebook. The study suggests that both advertisers and algorithms are responsible —essentially, that online advertising doesn’t provide all users equivalent experiences and that some people are exposed to more potentially exploitative or untruthful ads than others.