AI and Cloud Systems Lab

The AI and Cloud Systems Lab is dedicated to conducting transformative research at the intersection of AI, Cloud, and Edge Computing. Our work explores the synergies between artificial intelligence and computing systems to address the growing complexity of cloud and edge infrastructures, along with their associated software applications. We aim to advance robust performance, energy efficiency, and security in these environments. Our research spans two complementary directions: AI for Systems - leveraging AI to automate and optimize system operations; and Systems for AI - designing middleware and system software to enable efficient deployment of AI models for low-latency, energy-efficient inference in resource-constrained settings.

People

Faculty

Current Students

PhD Students: Masters Students: Undergraduate Students:

Past Students

PhD Students: Masters Students: Undergraduate Students:

Sponsored Projects

Collaborative Research: SaTC:EDU: Small: Integrating Cybersecurity in Computing Curricula: A Software PBLDriven Approach with Focus on Identity and Access Management (IAM). (Sponsor: NSF, Role: Co-PI, Total Award: $100,001, Period: 09/2023 - 08/2026)

The stability and well-being of virtually every facet of our society, ranging from national security, financial markets, and power systems to education, is contingent on the rapid and sustained development of a capable cybersecurity workforce. This long-recognized need is only increasing in importance, as the impact of vulnerabilities to our nation's cyberinfrastructures, whether exploited by malicious actors or the consequences of extreme weather events, becomes all too apparent. The goals of this education research project are (1) to iteratively develop a series of course projects and supporting modules and demonstrate the successful deployment of the developed projects in multiple courses in existing computer science and software engineering curricula and (2) to gather preliminary evidence of this approach's promise to (i) improve student learning outcomes in personal competencies, motivation, engagement, and overall satisfaction and mastery of cybersecurity skills and concepts, and (ii) promote the development and transferability of higher-order critical thinking skills, specifically analysis, synthesis, and evaluation in Bloom's taxonomy.

Secure Federated Learning at the Tactical Edge. (Sponsor: Army Research Office, Role: PI, Total Award: $149,850, Period: 11/2022 - 10/2024)

Federated Learning (FL) is a decentralized privacy-preserving approach that allows edge devices to collaboratively train machine learning (ML) and deep learning (DL) models without sharing the large amounts of data generated at the edge. However, FL can be vulnerable to data poisoning, model poisoning, and targeted model poisoning (backdoor attacks), where a malicious client influences model behavior without being detected. Our research focuses on developing robust techniques based on statistical analysis to detect malicious behavior and defend FL against adversarial attacks in a tactical edge environment.

Automated Techniques for Cyber Risk Detection and Mitigation in the Presence of Malicious AI Attacks. (Sponsor: NSA, Role: Co-PI, Total Award: $494,702, Period: 09/2021 - 08/2023)

The goal of this project is to develop automated and integrated techniques to detect and mitigate cyber risks and compromises, especially in the presence of AI-based cyberattacks.

CNS Core: Small: Robust Performance Guarantee of Containerized Microservices in the Cloud. (Sponsor: NSF, Role: PI, Total Award: $243,940, Period: 10/2019 - 9/2022)

Large-scale web services are increasingly being built with many small modular components (microservices), which can be deployed, updated and scaled seamlessly. These microservices are packaged to run in a lightweight isolated execution environment (containers) and deployed on compute resources rented from cloud providers. However, the complex interactions and the contention of shared hardware resources in cloud datacenters pose significant challenges in managing web service performance. This project will develop novel performance models and resource management solutions that can enable cloud platforms to provide robust performance guarantee for large-scale web services.

Detection and Visualization of DDoS Attacks on Software-Defined Networks (Sponser: UTSA & ITESM Seed Funding Program, Role: Co-PI, Total Award: $80,000, Period: 09/2019 - 08/2020)

This project investigates the vulnerabilities of network communication protocols and resource bottlenecks of software-defined network architecture, which are exploited by DDoS attacks. These results will be helpful in designing new techniques to detect DDoS attacks accurately using machine learning (ML), algorithmic, and statistical methods.

VENOM-Aided Adaptive Denial of Service Attacks on Sofware Defined Networks (Sponser: National Security Agency, Role: Co-PI, Total Award: $215,142, Period: 09/2018 - 08/2019)

This project investigates the vulnerabilities of software-defined networks architecture, and focuses on designing new DDoS attacks (cyber offence) which can automatically adapt itself to overcome existing cyber defense mechanisms.

CREST Center for Security and Privacy Enhanced Cloud Computing (C-SPECC) (Sponsor: NSF, Role: Senior Personnel, Total Award: $5 million, Period: 9/2017 - 8/2022)

The Centers of Research Excellence in Science and Technology (CREST) program supports the enhancement of research capabilities of minority-serving institutions through the establishment of centers that effectively integrate education and research. The Center articulates three research thrusts: Protection, Detection and Policy. The Protection sub-project will develop access control, private computing and protected computing technologies for cloud computing. The Detection sub-project will focus on system and host monitoring techniques to detect anomalous activity in a cloud along with digital forensic techniques for cloud-based systems. The Policy sub-project will research policy specification, composition and verification techniques for secure cloud computing.

Analysis and Training for the Defense of Biological and Digital Threats (Sponsor: Department of Homeland Security, Role: Co-PI, Total Award: $400,000, Period: 9/2014 - 8/2020)

The project focuses on enabling and improving cloud based informatics capabilities that are needed to support both biological and digital threat assessment activities. It involves introducing undergraduate DHS scholars/students to cross-disciplinary teaching and research on biological and digital pathogens, informatics techniques and procedures useful for pathogenic outbreak investigations.

Collaborative Research: Chameleon: A Large-Scale, Reconfigurable Experimental Environment for Cloud Research (Sponsor: NSF, Role: Key Personnel. 10/2014 - 09/2017)

This project focuses on building a large-scale platform to the open research community allowing them to explore transformative concepts in programmable cloud services, design, and core technologies.

Locality-Aware Fair Scheduling of ZeroVM in Multi-tenant Cloud (Sponsor: RackSpace, Role: Student Advisor. 08/2014 - 08/2017)

This project focuses on advancing an open source hypervisor technology, ZeroVM, to accelerate big data processing in the Cloud.

[HOME]