I am a PhD. student in the Department of Computer Science at The University of Iowa. I am being advised by Prof. Rishab Nithyanand. I also have had the pleasure of working under the mentorship of Prof. Zubair Shafiq previously.

At a high level, I am interested in improving the way content is generated and delivered between publishers and end-users across the Internet. My research interests span Content Delivery, Quality-of-Experience (QoE) and Internet Measurements. Specifically, I am interested in building data-driven tools to measure the performance of network and web applications, and design approaches to improve them. To this end, I have explored the following areas:
  1. Optimizing web page performance (HTTP/2 server push, HTML resource loading attributes)
  2. Measuring Quality-of-Service and facilitating the deployment of prior network measurement techniques (e.g. PathNeck)
  3. Studying the performance aspects of the ways networks connect across the Internet
  4. Measuring Quality-of-Experience for live video and identifying impairments in real-time

Publications

HTML Resource Attributes in the Wild
Adnan Ahmed, Rishab Nithyanand, Octav Chipara
(In Progress)

ATOMIX : Data-driven HTTP/2 Server Push
Adnan Ahmed, Harkeerat Bedi, Zubair Shafiq
(In Progress)

FlowTrace : A Framework for Active Bandwidth Measurements Using In-band Packet Trains
Adnan Ahmed, Ricky Mok, Zubair Shafiq
International Conference on Passive and Active Network Measurement (PAM) 2020

Peering vs. Transit: Performance Comparison of Peering and Transit Interconnections
Adnan Ahmed, Zubair Shafiq, Harkeerat Bedi, Amir Khakpour
IEEE International Conference on Network Protocols (ICNP) 2017

Suffering from Buffering? Detecting QoE Impairments in Live Video Streams
Adnan Ahmed, Zubair Shafiq, Harkeerat Bedi, Amir Khakpour
IEEE International Conference on Network Protocols (ICNP) 2017

QoE Analysis of a Large-Scale Live Video Streaming Event (Extended Abstract)
Adnan Ahmed, Zubair Shafiq, Amir Khakpour
Proceedings of the ACM SIGMETRICS (SIGMETRICS) 2016

Sneak-Peek: High Speed Covert Channels in Data Center Network
Rashid Tahir, Mohammad Taha Khan, Xun Gong, Adnan Ahmed, Amiremad Ghassami, Hasanat Kazmi, Matthew Caesar, Fareed Zaffar, Negar Kiyavash
IEEE International Conference on Computer Communications (INFOCOM) 2016

On the Co-existence of Transport Protocols in Data Centers
Syed Mohammad Irteza, Adnan Ahmed, Sana Farrukh, Babar Naveed Memon, Ihsan Ayyub Qazi
IEEE International Conference on Communications (ICC) 2014

Projects

HTML Resource Attributes in the Wild

Resource loading attributes, such as async and preload, have been around for a while for web developers to control how their web pages are loaded and improve web performance. In this work, we study the prevalence and performance impact of resource loading attributes.

Data Driven HTTP/2 Server Push

HTTP/2 server push has been standardized for a while and yet not widely used on the web pages across the Internet. In this research, we facilitate the deployment for HTTP/2 server push across web pages by designing a data-driven approach to discover pushable resources on a web page.

Leveraging Application Traffic to Conduct Active Measurements

One of the major hurdles in conducting large-scale network measurements using techniques like PathNeck, is the network overheads associated with these techniques. In this research, we design a framework to mitigate these overheads by leveraging useful application traffic as payload data for these techniques.

Detecting QoE Impairments in Live Video

The popularity of large-scale live video streaming has dramatically increased over the last few years. In this research, we analyze the relationship between Quality-of-Experience (QoE) and user engagement for a large-scale live video streaming event. We further design a data-driven approach to detect QoE impairments across viewers in real-time, allowing content providers to detect and mitigate QoE impairments on the fly instead of relying on post-hoc analysis.

Performance Aspects of Network Interconnection Strategies

More and more ISPs are engaging in direct peering relationships with bilateral traffic exchange agreements, instead of the traditional approach to buy global transit from upstream ISPs. In this research, we study the performance comparison of the peering and transit interconnection strategies to complement the previously understood economic aspects.

High-Speed Covert Channels in Data Center Networks

Modern cloud vendors host and process large volumes of sensitive customer information across a multitude of mutually untrusting tenants, increasing the propensity for data leakage. In this work, we present a novel and highspeed network-based covert channel that is robust and circumvents a broad set of security mechanisms currently deployed by cloud vendors.

Coexistence of Transport Protocols in Data Centers

Transport protocols such as Data Center TCP (DCTCP) are designed to improve the performance of cloud applications. But, in practice, these protocols may have to compete and coexist with TCP or other transport protocols. In this research, we present a comprehensive study of the coexistence of DCTCP and TCP, and evaluate their bandwidth sharing properties under different active queue management schemes (AQM). We also present a modified version of CHOKe AQM aimed to improve fairness among co-existing protocols by more accurately penalizing dominating flows.

Service

Peer Review

ACM IMC 2022 (External Reviewer)
TMA 2019 (External Reviewer)
ACM IMC 2017 (Shadow Program Committee Member)
IEEE INFOCOM 2016 (External Reviewer)
ACM SIGCOMM 2014 (External Reviewer)

Teaching

Intro to Networks and their Applications
Operating Systems
Networking and Security
Computer Organization
Discrete Structures
Intro to Computer Science

Updates