Goals and Approach. My research exploits computing resources available in the network infrastructure to develop network management tools that can profoundly impact how networks operate. A key aspect to my work is a practical approach to advanced research informed by extensive experience converting academic research projects into functional solutions. Throughout my career, I have striven to build software that 1) validates the proposed system designs via real-world rigorous evaluation, 2) that can be verified, reused, and extended by the research community, and 3) that can achieve important broader implications for both ISPs and consumers at large.
The MINT project seeks to develop techniques to regain visibility and insights into modern network traffic.
The PARFAIT project develops theoretical foundations for distributed and scalable resource allocation schemes on edge computing infrastructures tailored for AI-intensive processing tasks.
We developed a new framework and system that enables a joint evaluation of both the conventional notions of machine learning performance (e.g., model accuracy) and the systems-level costs of different representations of network traffic.
We developed a system that accurately infers video streaming quality metrics in real time, such as startup delay or video resolution, by using just a handful of features extracted from passive traffic measurement. Network Microscope passively collects a corpus of network features about the traffic flows of interest in the network and directs those to a real-time analytics framework that can perform more complex inference tasks. Network Microscope enables network operators to determine degradations in application quality as they happen, even when the traffic is encrypted.
Service Traceroute is a tool that allows the discovery of individual application flows paths. Service Traceroute leverages the ideas from paratrace, Sidecar, and 0trace, to passively listen to application traffic to then issue traceroute probes that pretend to be part of the application flow. Service Traceroute extends this idea to work for modern Internet services and support tracing of multiple concurrent TCP flows, as well as for UDP flows. Service Traceroute is available both as command-line tool and library versions. Check out our paper for more details on how Service Traceroute works.
The MobilityFirst project is funded by the National Science Foundation's Future Internet Architecture (FIA) program started in Sept 2010. The FIA program is aimed at design and validation of comprehensive new architectures for the next-generation Internet. This is a three-year project (2010-13) with scope including network design, performance evaluation, large-scale prototyping and end-user application trials.
Design of a Name-Based Virtual Networking (NVN) framework for end-to-end virtualization and service enhancement. Development of a working prototype using Click modular router.
Design of an in-network caching system aimed at improving quality of experience over wireless attached clients. Development of a OpenFlow-SDN based pre-fetching system for video streaming in Information Centric Networks.
Design of an SDN based framework for network-assisted dynamic spectrum access towards service enhancement in emerging 5G Radio Access Network scenarios. Development of a working prototype using OpenFlow and OpenAirInterface based software components.
Design and development of an hybrid CDN/P2P solution for Content Delivery Networks for the FP7 funded project COntent Aware Search retrieval and sTreaming (COAST).