<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>CS 440: Advanced Networking | Aqualab - Northwestern University</title><link>https://aqualab.cs.northwestern.edu/course/cs440-f26/</link><atom:link href="https://aqualab.cs.northwestern.edu/course/cs440-f26/index.xml" rel="self" type="application/rss+xml"/><description>CS 440: Advanced Networking</description><generator>Hugo Blox Builder (https://hugoblox.com)</generator><language>en-us</language><image><url>https://aqualab.cs.northwestern.edu/media/logo_hu_9be9516131db575.png</url><title>CS 440: Advanced Networking</title><link>https://aqualab.cs.northwestern.edu/course/cs440-f26/</link></image><item><title>Calendar</title><link>https://aqualab.cs.northwestern.edu/course/cs440-f26/calendar/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://aqualab.cs.northwestern.edu/course/cs440-f26/calendar/</guid><description>&lt;div class="course-tabs"&gt;&lt;a href="https://aqualab.cs.northwestern.edu/course/cs440-f26/"
class="course-tab"&gt;
Syllabus
&lt;/a&gt;&lt;a href="https://aqualab.cs.northwestern.edu/course/cs440-f26/calendar/"
class="course-tab course-tab-active"&gt;
Calendar
&lt;/a&gt;&lt;a href="https://aqualab.cs.northwestern.edu/course/cs440-f26/reading/"
class="course-tab"&gt;
Reading Blog &amp;amp; Presentations
&lt;/a&gt;&lt;a href="https://aqualab.cs.northwestern.edu/course/cs440-f26/project/"
class="course-tab"&gt;
Reproduction Project
&lt;/a&gt;&lt;/div&gt;
&lt;p&gt;&lt;strong&gt;[B]&lt;/strong&gt; = background paper (no blog post required) · &lt;strong&gt;[R]&lt;/strong&gt; = research paper (blog post required)&lt;/p&gt;
&lt;hr&gt;
&lt;div class="week-row"&gt;
&lt;div class="week-label"&gt;Week 1&lt;/div&gt;
&lt;div class="week-cards"&gt;
&lt;details class="day-card"&gt;
&lt;summary class="day-card__summary"&gt;
&lt;span class="day-card__date"&gt;Tue&lt;/span&gt;
&lt;span class="day-card__topic"&gt;Course introduction&lt;/span&gt;
&lt;span class="day-card__due-badge"&gt;Form teams&lt;/span&gt;
&lt;/summary&gt;
&lt;div class="day-card__body"&gt;
&lt;p&gt;&lt;strong&gt;Background&lt;/strong&gt;
&lt;a href="http://ccr.sigcomm.org/online/files/p83-keshavA.pdf" target="_blank" rel="noopener"&gt;How to Read a Paper,&lt;/a&gt;
S. Keshav, SIGCOMM CCR 2007&lt;/p&gt;
&lt;p&gt;&lt;a href="https://people.inf.ethz.ch/troscoe/pubs/review-writing.pdf" target="_blank" rel="noopener"&gt;Writing Reviews for Systems Conferences,&lt;/a&gt;
T. Roscoe, SOSP 2007&lt;/p&gt;
&lt;p&gt;&lt;a href="https://simon.peytonjones.org/great-research-talk/" target="_blank" rel="noopener"&gt;How to Give a Great Research Talk,&lt;/a&gt;
S. Peyton Jones et al., SIGPLAN 1993&lt;/p&gt;
&lt;p&gt;&lt;em&gt;No blog post due. Presenter: Fabián.&lt;/em&gt;&lt;/p&gt;
&lt;/div&gt;
&lt;/details&gt;
&lt;details class="day-card"&gt;
&lt;summary class="day-card__summary"&gt;
&lt;span class="day-card__date"&gt;Thu&lt;/span&gt;
&lt;span class="day-card__topic"&gt;Network design &amp;amp; architecture&lt;/span&gt;
&lt;span class="day-card__due-badge"&gt;Team formation confirmed&lt;/span&gt;
&lt;/summary&gt;
&lt;div class="day-card__body"&gt;
&lt;p&gt;&lt;strong&gt;Background&lt;/strong&gt;
&lt;a href="https://dl.acm.org/doi/10.1145/52325.52336" target="_blank" rel="noopener"&gt;The Design Philosophy of the DARPA Internet Protocols,&lt;/a&gt;
D. Clark, SIGCOMM 1988&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;Research paper&lt;/strong&gt;
&lt;a href="https://nines-conference.org/papers/p022-Misa.pdf" target="_blank" rel="noopener"&gt;There is More to Internet Invariants Than Meets the Eye,&lt;/a&gt;
C. Misa, W. Willinger, R. Durairajan, R. Rejaie — NINeS 2026&lt;/p&gt;
&lt;p&gt;&lt;em&gt;Presenter: Fabián.&lt;/em&gt;&lt;/p&gt;
&lt;/div&gt;
&lt;div class="day-card__note"&gt;
Opens the course with a foundational provocation: what properties of the Internet can we actually rely on? Watch the pre-recorded NINeS talk before class.
&lt;/div&gt;
&lt;/details&gt;
&lt;/div&gt;
&lt;/div&gt;
&lt;div class="week-row"&gt;
&lt;div class="week-label"&gt;Week 2&lt;/div&gt;
&lt;div class="week-cards"&gt;
&lt;details class="day-card"&gt;
&lt;summary class="day-card__summary"&gt;
&lt;span class="day-card__date"&gt;Tue&lt;/span&gt;
&lt;span class="day-card__topic"&gt;Internet routing &amp;amp; reachability&lt;/span&gt;
&lt;/summary&gt;
&lt;div class="day-card__body"&gt;
&lt;p&gt;&lt;strong&gt;Background&lt;/strong&gt;
&lt;a href="https://www.cs.princeton.edu/~jrex/papers/sigmetrics00.pdf" target="_blank" rel="noopener"&gt;Stable Internet Routing Without Global Coordination,&lt;/a&gt;
L. Gao and J. Rexford, SIGMETRICS 2000&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;Research paper&lt;/strong&gt;
&lt;a href="https://nines-conference.org/papers/p004-Baltra.pdf" target="_blank" rel="noopener"&gt;Understanding Partial Reachability in the Internet Core,&lt;/a&gt;
G. Baltra, T. Saluja, Y. Pradkin, J. Heidemann — NINeS 2026&lt;/p&gt;
&lt;/div&gt;
&lt;div class="day-card__note"&gt;
From policy design (Gao-Rexford) to an underexplored question: not all reachability failures are total. What do partial failures reveal about routing system design?
&lt;/div&gt;
&lt;/details&gt;
&lt;details class="day-card"&gt;
&lt;summary class="day-card__summary"&gt;
&lt;span class="day-card__date"&gt;Thu&lt;/span&gt;
&lt;span class="day-card__topic"&gt;WAN architecture &amp;amp; traffic engineering&lt;/span&gt;
&lt;span class="day-card__due-badge"&gt;Paper selection &amp;#43; project proposal&lt;/span&gt;
&lt;/summary&gt;
&lt;div class="day-card__body"&gt;
&lt;p&gt;&lt;strong&gt;Background&lt;/strong&gt;
&lt;a href="https://dl.acm.org/doi/10.1145/2486001.2486019" target="_blank" rel="noopener"&gt;B4: Experience with a Globally-Deployed Software Defined WAN,&lt;/a&gt;
S. Jain et al., SIGCOMM 2013&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;Research paper&lt;/strong&gt;
&lt;a href="https://dl.acm.org/doi/10.1145/3718958.3754348" target="_blank" rel="noopener"&gt;Raha: A General Tool to Analyze WAN Degradation,&lt;/a&gt;
B. Arzani, S. Taheri, P. Namyar, R. Beckett, S. K. Kakarla, E. Jalilipour — SIGCOMM 2025&lt;/p&gt;
&lt;/div&gt;
&lt;/details&gt;
&lt;/div&gt;
&lt;/div&gt;
&lt;div class="week-row"&gt;
&lt;div class="week-label"&gt;Week 3&lt;/div&gt;
&lt;div class="week-cards"&gt;
&lt;details class="day-card"&gt;
&lt;summary class="day-card__summary"&gt;
&lt;span class="day-card__date"&gt;Tue&lt;/span&gt;
&lt;span class="day-card__topic"&gt;SDN &amp;amp; programmable networks&lt;/span&gt;
&lt;/summary&gt;
&lt;div class="day-card__body"&gt;
&lt;p&gt;&lt;strong&gt;Background&lt;/strong&gt;
&lt;a href="https://dl.acm.org/doi/10.1145/2602204.2602219" target="_blank" rel="noopener"&gt;The Road to SDN: An Intellectual History of Programmable Networks,&lt;/a&gt;
N. Feamster, J. Rexford, E. Zegura, SIGCOMM CCR 2014&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;Research paper&lt;/strong&gt;
&lt;a href="https://www.usenix.org/conference/nsdi25/presentation/liu-yadong" target="_blank" rel="noopener"&gt;Unlocking ECMP Programmability for Precise Traffic Control,&lt;/a&gt;
Y. Liu, Y. Xiao, X. Zhang et al. — NSDI 2025&lt;/p&gt;
&lt;/div&gt;
&lt;/details&gt;
&lt;details class="day-card"&gt;
&lt;summary class="day-card__summary"&gt;
&lt;span class="day-card__date"&gt;Thu&lt;/span&gt;
&lt;span class="day-card__topic"&gt;Network verification &amp;amp; formal methods&lt;/span&gt;
&lt;/summary&gt;
&lt;div class="day-card__body"&gt;
&lt;p&gt;&lt;strong&gt;Background&lt;/strong&gt;
&lt;a href="https://www.usenix.org/conference/nsdi12/technical-sessions/presentation/kazemian" target="_blank" rel="noopener"&gt;Header Space Analysis: Static Checking for Networks,&lt;/a&gt;
P. Kazemian, G. Varghese, N. McKeown, NSDI 2012&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;Research paper&lt;/strong&gt;
&lt;a href="https://www.usenix.org/conference/nsdi25/presentation/li-zechun" target="_blank" rel="noopener"&gt;NDD: A Decision Diagram for Network Verification,&lt;/a&gt;
Z. Li, P. Zhang, Y. Zhang, H. Yang — NSDI 2025 &lt;em&gt;(Outstanding Paper Award)&lt;/em&gt;&lt;/p&gt;
&lt;/div&gt;
&lt;/details&gt;
&lt;/div&gt;
&lt;/div&gt;
&lt;div class="week-row"&gt;
&lt;div class="week-label"&gt;Week 4&lt;/div&gt;
&lt;div class="week-cards"&gt;
&lt;details class="day-card"&gt;
&lt;summary class="day-card__summary"&gt;
&lt;span class="day-card__date"&gt;Tue&lt;/span&gt;
&lt;span class="day-card__topic"&gt;Congestion control: signals &amp;amp; theory&lt;/span&gt;
&lt;/summary&gt;
&lt;div class="day-card__body"&gt;
&lt;p&gt;&lt;strong&gt;Background&lt;/strong&gt;
&lt;a href="https://dl.acm.org/doi/10.1145/3365609.3365855" target="_blank" rel="noopener"&gt;Beyond Jain&amp;rsquo;s Fairness Index,&lt;/a&gt;
R. Ware, M. Mukerjee, S. Seshan, J. Sherry, HotNets 2019&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;Research paper&lt;/strong&gt;
&lt;a href="https://nines-conference.org/papers/p012-McClure.pdf" target="_blank" rel="noopener"&gt;No Signal to Rule Them All: A Systematic Analysis of In-Network Congestion Signals,&lt;/a&gt;
S. McClure, N. Dukkipati, S. Ratnasamy, S. Shenker — NINeS 2026&lt;/p&gt;
&lt;p&gt;&lt;em&gt;Alt: Agarwal, Arun, Seshan, &lt;a href="https://nines-conference.org/papers/p008-Agarwal.pdf" target="_blank" rel="noopener"&gt;Contracts: A Unified Lens on Congestion Control&lt;/a&gt;, NINeS 2026.&lt;/em&gt;&lt;/p&gt;
&lt;/div&gt;
&lt;div class="day-card__note"&gt;
Argues no single in-network congestion signal can satisfy all desiderata simultaneously. Dukkipati also delivers a NINeS keynote on CC — watch both paper and talk.
&lt;/div&gt;
&lt;/details&gt;
&lt;details class="day-card"&gt;
&lt;summary class="day-card__summary"&gt;
&lt;span class="day-card__date"&gt;Thu&lt;/span&gt;
&lt;span class="day-card__topic"&gt;Transport &amp;amp; QUIC&lt;/span&gt;
&lt;span class="day-card__due-badge"&gt;In-class proposal pitches (5 min per team)&lt;/span&gt;
&lt;/summary&gt;
&lt;div class="day-card__body"&gt;
&lt;p&gt;&lt;strong&gt;Background&lt;/strong&gt;
&lt;a href="https://dl.acm.org/doi/10.1145/3098822.3098842" target="_blank" rel="noopener"&gt;QUIC at Google: Deployed at Scale,&lt;/a&gt;
A. Langley et al., SIGCOMM 2017&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;Research paper&lt;/strong&gt;
&lt;a href="https://dl.acm.org/doi/10.1145/3768988" target="_blank" rel="noopener"&gt;Waiting for QUIC: Passive Measurements to Understand QUIC Deployments,&lt;/a&gt;
J. Mücke, M. Nawrocki, R. Hiesgen et al. — CoNEXT 2025&lt;/p&gt;
&lt;/div&gt;
&lt;/details&gt;
&lt;/div&gt;
&lt;/div&gt;
&lt;div class="week-row"&gt;
&lt;div class="week-label"&gt;Week 5&lt;/div&gt;
&lt;div class="week-cards"&gt;
&lt;details class="day-card"&gt;
&lt;summary class="day-card__summary"&gt;
&lt;span class="day-card__date"&gt;Tue&lt;/span&gt;
&lt;span class="day-card__topic"&gt;Datacenter transport: rethinking the model&lt;/span&gt;
&lt;/summary&gt;
&lt;div class="day-card__body"&gt;
&lt;p&gt;&lt;strong&gt;Background&lt;/strong&gt;
&lt;a href="https://dl.acm.org/doi/10.1145/1402958.1402967" target="_blank" rel="noopener"&gt;A Scalable, Commodity Data Center Network Architecture,&lt;/a&gt;
M. Al-Fares, A. Loukissas, A. Vahdat, SIGCOMM 2008&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;Research paper&lt;/strong&gt;
&lt;a href="https://nines-conference.org/papers/p019-Straussman.pdf" target="_blank" rel="noopener"&gt;Simulate Before Sending: Rethinking Transport in Datacenter Networks,&lt;/a&gt;
D. Straussman, I. Keslassy, A. Shpiner, L. Liss — NINeS 2026&lt;/p&gt;
&lt;/div&gt;
&lt;div class="day-card__note"&gt;
Proposes that endpoints simulate network behavior before committing to a send rate — a genuine departure from reactive transport design.
&lt;/div&gt;
&lt;/details&gt;
&lt;details class="day-card"&gt;
&lt;summary class="day-card__summary"&gt;
&lt;span class="day-card__date"&gt;Thu&lt;/span&gt;
&lt;span class="day-card__topic"&gt;RDMA &amp;amp; high-performance datacenter networking&lt;/span&gt;
&lt;/summary&gt;
&lt;div class="day-card__body"&gt;
&lt;p&gt;&lt;strong&gt;Background&lt;/strong&gt;
&lt;a href="https://dl.acm.org/doi/10.1145/1851275.1851192" target="_blank" rel="noopener"&gt;DCTCP: Data Center TCP,&lt;/a&gt;
M. Alizadeh et al., SIGCOMM 2010&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;Research paper&lt;/strong&gt;
&lt;a href="https://dl.acm.org/doi/10.1145/3718958.3750480" target="_blank" rel="noopener"&gt;Revisiting RDMA Reliability for Lossy Fabrics,&lt;/a&gt;
W. Li, X. Liu, Y. Zhang et al. — SIGCOMM 2025 &lt;em&gt;(Best Student Paper, Honorable Mention)&lt;/em&gt;&lt;/p&gt;
&lt;/div&gt;
&lt;/details&gt;
&lt;/div&gt;
&lt;/div&gt;
&lt;div class="week-row"&gt;
&lt;div class="week-label"&gt;Week 6&lt;/div&gt;
&lt;div class="week-cards"&gt;
&lt;details class="day-card"&gt;
&lt;summary class="day-card__summary"&gt;
&lt;span class="day-card__date"&gt;Tue&lt;/span&gt;
&lt;span class="day-card__topic"&gt;ML training networks&lt;/span&gt;
&lt;span class="day-card__due-badge"&gt;Midpoint report&lt;/span&gt;
&lt;/summary&gt;
&lt;div class="day-card__body"&gt;
&lt;p&gt;&lt;strong&gt;Background&lt;/strong&gt;
&lt;a href="https://dl.acm.org/doi/10.1145/3458817.3476209" target="_blank" rel="noopener"&gt;Efficient Large-Scale Language Model Training on GPU Clusters Using Megatron-LM,&lt;/a&gt;
D. Narayanan et al., SC 2021&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;Research paper&lt;/strong&gt;
&lt;a href="https://dl.acm.org/doi/10.1145/3718958.3750468" target="_blank" rel="noopener"&gt;InfiniteHBD: Building Datacenter-Scale High-Bandwidth Domain for LLM with Optical Circuit Switching Transceivers,&lt;/a&gt;
C. Shou, G. Liu, H. Nie et al. — SIGCOMM 2025&lt;/p&gt;
&lt;p&gt;&lt;em&gt;Alt: Z. Zhang et al., &lt;a href="https://dl.acm.org/doi/10.1145/3718958.3750472" target="_blank" rel="noopener"&gt;DistTrain: Addressing Model and Data Heterogeneity with Disaggregated Training for Multimodal LLMs&lt;/a&gt;, SIGCOMM 2025.&lt;/em&gt;&lt;/p&gt;
&lt;/div&gt;
&lt;/details&gt;
&lt;details class="day-card"&gt;
&lt;summary class="day-card__summary"&gt;
&lt;span class="day-card__date"&gt;Thu&lt;/span&gt;
&lt;span class="day-card__topic"&gt;Sustainable &amp;amp; energy-efficient networking&lt;/span&gt;
&lt;/summary&gt;
&lt;div class="day-card__body"&gt;
&lt;p&gt;&lt;strong&gt;Background&lt;/strong&gt;
&lt;a href="https://doi.org/10.2200/S00874ED3V01Y201906CAC046" target="_blank" rel="noopener"&gt;The Datacenter as a Computer&lt;/a&gt; (power and energy chapters),
L. A. Barroso, U. Hölzle, P. Ranganathan, 3rd ed., 2019&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;Research paper&lt;/strong&gt;
&lt;a href="https://www.usenix.org/conference/nsdi25/presentation/xu-kaiqiang" target="_blank" rel="noopener"&gt;GREEN: Carbon-Efficient Resource Scheduling for Machine Learning Clusters,&lt;/a&gt;
K. Xu, D. Sun, H. Tian, J. Zhang, K. Chen — NSDI 2025&lt;/p&gt;
&lt;p&gt;&lt;em&gt;Alt: Z. Liu, S. Oh, B. Tao et al., &lt;a href="https://nines-conference.org/papers/p006-Liu.pdf" target="_blank" rel="noopener"&gt;EcoCell: Energy Conservation through Traffic Shaping in Cellular Radio Access Networks&lt;/a&gt;, NINeS 2026.&lt;/em&gt;&lt;/p&gt;
&lt;/div&gt;
&lt;/details&gt;
&lt;/div&gt;
&lt;/div&gt;
&lt;div class="week-row"&gt;
&lt;div class="week-label"&gt;Week 7&lt;/div&gt;
&lt;div class="week-cards"&gt;
&lt;details class="day-card"&gt;
&lt;summary class="day-card__summary"&gt;
&lt;span class="day-card__date"&gt;Tue&lt;/span&gt;
&lt;span class="day-card__topic"&gt;Naming &amp;amp; DNS infrastructure&lt;/span&gt;
&lt;/summary&gt;
&lt;div class="day-card__body"&gt;
&lt;p&gt;&lt;strong&gt;Background&lt;/strong&gt;
&lt;a href="https://dl.acm.org/doi/10.1145/3387514.3405881" target="_blank" rel="noopener"&gt;Akamai DNS: Providing Authoritative Answers to the World&amp;rsquo;s Queries,&lt;/a&gt;
K. Schomp et al., SIGCOMM 2020&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;Research paper&lt;/strong&gt;
&lt;a href="https://nines-conference.org/papers/p020-Bharadwaj.pdf" target="_blank" rel="noopener"&gt;OpenFLAME: A Federated Spatial Naming Infrastructure,&lt;/a&gt;
S. Bharadwaj, Z. Ma, I. Liang, M. Farb, A. Rowe, S. Seshan — NINeS 2026&lt;/p&gt;
&lt;/div&gt;
&lt;div class="day-card__note"&gt;
Akamai DNS shows naming at scale for the current Internet; OpenFLAME asks what naming should look like for future networked systems — location-aware, federated, built for IoT and physical spaces.
&lt;/div&gt;
&lt;/details&gt;
&lt;details class="day-card"&gt;
&lt;summary class="day-card__summary"&gt;
&lt;span class="day-card__date"&gt;Thu&lt;/span&gt;
&lt;span class="day-card__topic"&gt;LEO satellites &amp;amp; new access infrastructure&lt;/span&gt;
&lt;/summary&gt;
&lt;div class="day-card__body"&gt;
&lt;p&gt;&lt;strong&gt;Background&lt;/strong&gt;
&lt;a href="https://dl.acm.org/doi/10.1145/3544216.3544237" target="_blank" rel="noopener"&gt;Analyzing Starlink from the Ground,&lt;/a&gt;
Q. Michel et al., SIGCOMM 2022&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;Research paper&lt;/strong&gt;
&lt;a href="https://dl.acm.org/doi/10.1145/3718958.3750491" target="_blank" rel="noopener"&gt;LeoCC: Making Internet Congestion Control Robust to LEO Satellite Dynamics,&lt;/a&gt;
Z. Lai, Z. Li, Q. Wu et al. — SIGCOMM 2025&lt;/p&gt;
&lt;p&gt;&lt;em&gt;Alt: B. Kataria, H. Bin Tanveer, R. Nithyanand, R. Singh, &lt;a href="https://nines-conference.org/papers/p007-Kataria.pdf" target="_blank" rel="noopener"&gt;What Obstructed Skies Teach Us about Satellite Internet&lt;/a&gt;, NINeS 2026.&lt;/em&gt;&lt;/p&gt;
&lt;/div&gt;
&lt;div class="day-card__note"&gt;
A systems paper: rather than measuring LEO performance, LeoCC proposes a new CC algorithm aware of satellite-induced path changes.
&lt;/div&gt;
&lt;/details&gt;
&lt;/div&gt;
&lt;/div&gt;
&lt;div class="week-row"&gt;
&lt;div class="week-label"&gt;Week 8&lt;/div&gt;
&lt;div class="week-cards"&gt;
&lt;details class="day-card"&gt;
&lt;summary class="day-card__summary"&gt;
&lt;span class="day-card__date"&gt;Tue&lt;/span&gt;
&lt;span class="day-card__topic"&gt;Video QoE &amp;amp; real-time congestion control&lt;/span&gt;
&lt;/summary&gt;
&lt;div class="day-card__body"&gt;
&lt;p&gt;&lt;strong&gt;Background&lt;/strong&gt;
&lt;a href="https://dl.acm.org/doi/10.1145/2398776.2398799" target="_blank" rel="noopener"&gt;Video Stream Quality Impacts Viewer Behavior,&lt;/a&gt;
S. Krishnan and R. Sitaraman, IMC 2012&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;Research paper&lt;/strong&gt;
&lt;a href="https://nines-conference.org/papers/p009-Karimi.pdf" target="_blank" rel="noopener"&gt;Tight Loops, Smooth Streams: Responsive Congestion Control for Real-Time Video,&lt;/a&gt;
P. Karimi, S. Fouladi, V. Sivaraman, M. Alizadeh — NINeS 2026&lt;/p&gt;
&lt;/div&gt;
&lt;div class="day-card__note"&gt;
Argues TCP-style CC is architecturally too slow for real-time video. Watch the pre-recorded NINeS talk before class.
&lt;/div&gt;
&lt;/details&gt;
&lt;details class="day-card"&gt;
&lt;summary class="day-card__summary"&gt;
&lt;span class="day-card__date"&gt;Thu&lt;/span&gt;
&lt;span class="day-card__topic"&gt;Mobile &amp;amp; wireless systems design&lt;/span&gt;
&lt;/summary&gt;
&lt;div class="day-card__body"&gt;
&lt;p&gt;&lt;strong&gt;Background&lt;/strong&gt;
&lt;a href="https://dl.acm.org/doi/10.1145/3012426.3022184" target="_blank" rel="noopener"&gt;BBR: Congestion-Based Congestion Control,&lt;/a&gt;
N. Cardwell, Y. Cheng, C. S. Gunn, S. H. Yeganeh, V. Jacobson, ACM Queue 2016&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;Research paper&lt;/strong&gt;
&lt;a href="https://nines-conference.org/papers/p015-Larrea.pdf" target="_blank" rel="noopener"&gt;BISCAY: Practical Radio KPI Driven Congestion Control for Mobile Networks,&lt;/a&gt;
J. Larrea, T. Shreedhar, A. Niemi, A. Sefiane, M. K. Marina — NINeS 2026&lt;/p&gt;
&lt;p&gt;&lt;em&gt;Alt: M. Tariq, Y. Chen, H. Hassanieh, R. Mittal, &lt;a href="https://nines-conference.org/papers/p002-Tariq.pdf" target="_blank" rel="noopener"&gt;Performance Isolation for 5G RAN Slices Across Multiple Interfering Cells&lt;/a&gt;, NINeS 2026.&lt;/em&gt;&lt;/p&gt;
&lt;/div&gt;
&lt;div class="day-card__note"&gt;
BBR uses network-layer signals to estimate state; BISCAY adds radio-layer KPIs as an additional signal source — extending model-based CC into the mobile context.
&lt;/div&gt;
&lt;/details&gt;
&lt;/div&gt;
&lt;/div&gt;
&lt;div class="week-row"&gt;
&lt;div class="week-label"&gt;Week 9&lt;/div&gt;
&lt;div class="week-cards"&gt;
&lt;details class="day-card"&gt;
&lt;summary class="day-card__summary"&gt;
&lt;span class="day-card__date"&gt;Tue&lt;/span&gt;
&lt;span class="day-card__topic"&gt;Anonymity &amp;amp; privacy infrastructure&lt;/span&gt;
&lt;/summary&gt;
&lt;div class="day-card__body"&gt;
&lt;p&gt;&lt;strong&gt;Background&lt;/strong&gt;
&lt;a href="https://www.usenix.org/conference/13th-usenix-security-symposium/tor-second-generation-onion-router" target="_blank" rel="noopener"&gt;Tor: The Second-Generation Onion Router,&lt;/a&gt;
R. Dingledine, N. Mathewson, P. Syverson, USENIX Security 2004&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;Research paper&lt;/strong&gt;
&lt;a href="https://nines-conference.org/papers/p017-Ikram.pdf" target="_blank" rel="noopener"&gt;Don&amp;rsquo;t get caught, keep your Onions in a Vault,&lt;/a&gt;
H. Ikram, R. Habib, M. Ali, Z. A. Uzmi — NINeS 2026&lt;/p&gt;
&lt;/div&gt;
&lt;div class="day-card__note"&gt;
From the original Tor design to a new approach to anonymity infrastructure. Discussion should address both the systems design choices and the ethical implications.
&lt;/div&gt;
&lt;/details&gt;
&lt;details class="day-card"&gt;
&lt;summary class="day-card__summary"&gt;
&lt;span class="day-card__date"&gt;Thu&lt;/span&gt;
&lt;span class="day-card__topic"&gt;BGP security &amp;amp; routing integrity&lt;/span&gt;
&lt;span class="day-card__due-badge"&gt;Final report (including critique section)&lt;/span&gt;
&lt;/summary&gt;
&lt;div class="day-card__body"&gt;
&lt;p&gt;&lt;strong&gt;Background&lt;/strong&gt;
&lt;a href="https://dl.acm.org/doi/10.1145/1851182.1851198" target="_blank" rel="noopener"&gt;How Secure Are Secure Interdomain Routing Protocols?,&lt;/a&gt;
S. Goldberg, M. Schapira, P. Hummon, J. Rexford, SIGCOMM 2010&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;Research paper&lt;/strong&gt;
&lt;a href="https://nines-conference.org/papers/p014-Sengupta.pdf" target="_blank" rel="noopener"&gt;Passive Data-Plane Telemetry to Mitigate Long-Distance BGP Hijacks,&lt;/a&gt;
S. Sengupta, H. Kim, D. Jubas, M. Apostolaki, J. Rexford — NINeS 2026&lt;/p&gt;
&lt;/div&gt;
&lt;div class="day-card__note"&gt;
The background paper asks how secure BGP security proposals are; this paper proposes bypassing the BGP control plane entirely for hijack detection using in-band telemetry. Note the Rexford co-authorship across both papers — 15 years of progress on the same problem.
&lt;/div&gt;
&lt;/details&gt;
&lt;/div&gt;
&lt;/div&gt;
&lt;div class="week-row"&gt;
&lt;div class="week-label"&gt;Week 10&lt;/div&gt;
&lt;div class="week-cards"&gt;
&lt;details class="day-card"&gt;
&lt;summary class="day-card__summary"&gt;
&lt;span class="day-card__date"&gt;Tue&lt;/span&gt;
&lt;span class="day-card__topic"&gt;CDN architecture &amp;amp; Internet infrastructure&lt;/span&gt;
&lt;/summary&gt;
&lt;div class="day-card__body"&gt;
&lt;p&gt;&lt;strong&gt;Background&lt;/strong&gt;
&lt;a href="https://dl.acm.org/doi/10.1145/3098822.3098853" target="_blank" rel="noopener"&gt;Engineering Egress with Edge Fabric,&lt;/a&gt;
B. Schlinker, H. Kim, T. Cunha et al., SIGCOMM 2017&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;Research paper&lt;/strong&gt;
&lt;a href="https://nines-conference.org/papers/p023-Kumar.pdf" target="_blank" rel="noopener"&gt;Who Holds the Steering Wheel? Opacity and Consolidation in CDN Replica Selection,&lt;/a&gt;
R. Kumar, F. E. Bustamante, M. Flores — NINeS 2026&lt;/p&gt;
&lt;p&gt;&lt;em&gt;Companion reading: M. Weaver, D. Veitch, P. Barford, F. E. Bustamante, E. Carisimo, &lt;a href="https://nines-conference.org/papers/p016-Weaver.pdf" target="_blank" rel="noopener"&gt;Monitoring Latency on Submarine Cables&lt;/a&gt;, NINeS 2026.&lt;/em&gt;&lt;/p&gt;
&lt;p&gt;&lt;em&gt;Presenter: Fabián.&lt;/em&gt;&lt;/p&gt;
&lt;/div&gt;
&lt;div class="day-card__note"&gt;
Edge Fabric shows how content leaves a major network; this paper asks who controls which CDN replica serves you — and argues that consolidation is making that answer increasingly opaque.
&lt;/div&gt;
&lt;/details&gt;
&lt;details class="day-card"&gt;
&lt;summary class="day-card__summary"&gt;
&lt;span class="day-card__date"&gt;Thu&lt;/span&gt;
&lt;span class="day-card__topic"&gt;Student project presentations&lt;/span&gt;
&lt;span class="day-card__due-badge"&gt;Blog posts due&lt;/span&gt;
&lt;/summary&gt;
&lt;div class="day-card__body"&gt;
All teams present (15 min + Q&amp;amp;A).
&lt;/div&gt;
&lt;/details&gt;
&lt;/div&gt;
&lt;/div&gt;
&lt;div class="week-row"&gt;
&lt;div class="week-label"&gt;Week Finals&lt;/div&gt;
&lt;div class="week-cards"&gt;
&lt;details class="day-card"&gt;
&lt;summary class="day-card__summary"&gt;
&lt;span class="day-card__date"&gt;&lt;/span&gt;
&lt;span class="day-card__topic"&gt;Reproduction project presentations (date TBD)&lt;/span&gt;
&lt;/summary&gt;
&lt;div class="day-card__body"&gt;
&lt;/div&gt;
&lt;/details&gt;
&lt;/div&gt;
&lt;/div&gt;
&lt;hr&gt;
&lt;h2 id="paper-index"&gt;PAPER INDEX&lt;/h2&gt;
&lt;h3 id="background-papers"&gt;Background papers&lt;/h3&gt;
&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;#&lt;/th&gt;
&lt;th&gt;Paper&lt;/th&gt;
&lt;th&gt;Venue&lt;/th&gt;
&lt;th&gt;Week&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;B1&lt;/td&gt;
&lt;td&gt;Keshav, &lt;em&gt;How to Read a Paper&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;SIGCOMM CCR 2007&lt;/td&gt;
&lt;td&gt;1 Tue&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;B2&lt;/td&gt;
&lt;td&gt;Roscoe, &lt;em&gt;Writing Reviews for Systems Conferences&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;SOSP 2007&lt;/td&gt;
&lt;td&gt;1 Tue&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;B3&lt;/td&gt;
&lt;td&gt;Peyton Jones et al., &lt;em&gt;How to Give a Great Research Talk&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;SIGPLAN 1993&lt;/td&gt;
&lt;td&gt;1 Tue&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;B4&lt;/td&gt;
&lt;td&gt;Clark, &lt;em&gt;Design Philosophy of the DARPA Internet Protocols&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;SIGCOMM 1988&lt;/td&gt;
&lt;td&gt;1 Thu&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;B5&lt;/td&gt;
&lt;td&gt;Gao &amp;amp; Rexford, &lt;em&gt;Stable Internet Routing Without Global Coordination&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;SIGMETRICS 2000&lt;/td&gt;
&lt;td&gt;2 Tue&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;B6&lt;/td&gt;
&lt;td&gt;Jain et al., &lt;em&gt;B4: Globally-Deployed Software Defined WAN&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;SIGCOMM 2013&lt;/td&gt;
&lt;td&gt;2 Thu&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;B7&lt;/td&gt;
&lt;td&gt;Feamster, Rexford, Zegura, &lt;em&gt;The Road to SDN&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;SIGCOMM CCR 2014&lt;/td&gt;
&lt;td&gt;3 Tue&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;B8&lt;/td&gt;
&lt;td&gt;Kazemian, Varghese, McKeown, &lt;em&gt;Header Space Analysis&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;NSDI 2012&lt;/td&gt;
&lt;td&gt;3 Thu&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;B9&lt;/td&gt;
&lt;td&gt;Ware et al., &lt;em&gt;Beyond Jain&amp;rsquo;s Fairness Index&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;HotNets 2019&lt;/td&gt;
&lt;td&gt;4 Tue&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;B10&lt;/td&gt;
&lt;td&gt;Langley et al., &lt;em&gt;QUIC at Google: Deployed at Scale&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;SIGCOMM 2017&lt;/td&gt;
&lt;td&gt;4 Thu&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;B11&lt;/td&gt;
&lt;td&gt;Al-Fares, Loukissas, Vahdat, &lt;em&gt;Scalable Commodity Datacenter Architecture&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;SIGCOMM 2008&lt;/td&gt;
&lt;td&gt;5 Tue&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;B12&lt;/td&gt;
&lt;td&gt;Alizadeh et al., &lt;em&gt;DCTCP: Data Center TCP&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;SIGCOMM 2010&lt;/td&gt;
&lt;td&gt;5 Thu&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;B13&lt;/td&gt;
&lt;td&gt;Narayanan et al., &lt;em&gt;Efficient LLM Training: Megatron-LM&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;SC 2021&lt;/td&gt;
&lt;td&gt;6 Tue&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;B14&lt;/td&gt;
&lt;td&gt;Barroso, Hölzle, Ranganathan, &lt;em&gt;The Datacenter as a Computer&lt;/em&gt; (power ch.)&lt;/td&gt;
&lt;td&gt;Book 2019&lt;/td&gt;
&lt;td&gt;6 Thu&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;B15&lt;/td&gt;
&lt;td&gt;Schomp et al., &lt;em&gt;Akamai DNS&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;SIGCOMM 2020&lt;/td&gt;
&lt;td&gt;7 Tue&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;B16&lt;/td&gt;
&lt;td&gt;Michel et al., &lt;em&gt;Analyzing Starlink from the Ground&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;SIGCOMM 2022&lt;/td&gt;
&lt;td&gt;7 Thu&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;B17&lt;/td&gt;
&lt;td&gt;Krishnan &amp;amp; Sitaraman, &lt;em&gt;Video Stream Quality Impacts Viewer Behavior&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;IMC 2012&lt;/td&gt;
&lt;td&gt;8 Tue&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;B18&lt;/td&gt;
&lt;td&gt;Cardwell et al., &lt;em&gt;BBR: Congestion-Based Congestion Control&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;ACM Queue 2016&lt;/td&gt;
&lt;td&gt;8 Thu&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;B19&lt;/td&gt;
&lt;td&gt;Dingledine, Mathewson, Syverson, &lt;em&gt;Tor: The Second-Generation Onion Router&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;USENIX Sec. 2004&lt;/td&gt;
&lt;td&gt;9 Tue&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;B20&lt;/td&gt;
&lt;td&gt;Goldberg, Schapira, Hummon, Rexford, &lt;em&gt;How Secure Are Secure Routing Protocols?&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;SIGCOMM 2010&lt;/td&gt;
&lt;td&gt;9 Thu&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;B21&lt;/td&gt;
&lt;td&gt;Schlinker et al., &lt;em&gt;Engineering Egress with Edge Fabric&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;SIGCOMM 2017&lt;/td&gt;
&lt;td&gt;10 Tue&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;h3 id="research-papers-blog-post-required"&gt;Research papers (blog post required)&lt;/h3&gt;
&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;#&lt;/th&gt;
&lt;th&gt;Paper&lt;/th&gt;
&lt;th&gt;Venue&lt;/th&gt;
&lt;th&gt;Week&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;R1&lt;/td&gt;
&lt;td&gt;Misa, Willinger, Durairajan, Rejaie — &lt;em&gt;Internet Invariants&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;NINeS 2026&lt;/td&gt;
&lt;td&gt;1 Thu&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;R2&lt;/td&gt;
&lt;td&gt;Baltra, Saluja, Pradkin, Heidemann — &lt;em&gt;Partial Reachability in the Internet Core&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;NINeS 2026&lt;/td&gt;
&lt;td&gt;2 Tue&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;R3&lt;/td&gt;
&lt;td&gt;Arzani et al. — &lt;em&gt;Raha: Analyzing WAN Degradation&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;SIGCOMM 2025&lt;/td&gt;
&lt;td&gt;2 Thu&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;R4&lt;/td&gt;
&lt;td&gt;Liu, Xiao, Zhang et al. — &lt;em&gt;Unlocking ECMP Programmability&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;NSDI 2025&lt;/td&gt;
&lt;td&gt;3 Tue&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;R5&lt;/td&gt;
&lt;td&gt;Li, Zhang, Zhang, Yang — &lt;em&gt;NDD: Network Decision Diagram&lt;/em&gt; &lt;em&gt;(Outstanding Paper)&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;NSDI 2025&lt;/td&gt;
&lt;td&gt;3 Thu&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;R6&lt;/td&gt;
&lt;td&gt;McClure, Dukkipati, Ratnasamy, Shenker — &lt;em&gt;No Signal to Rule Them All&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;NINeS 2026&lt;/td&gt;
&lt;td&gt;4 Tue&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;R7&lt;/td&gt;
&lt;td&gt;Mücke, Nawrocki, Hiesgen et al. — &lt;em&gt;Waiting for QUIC&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;CoNEXT 2025&lt;/td&gt;
&lt;td&gt;4 Thu&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;R8&lt;/td&gt;
&lt;td&gt;Straussman, Keslassy, Shpiner, Liss — &lt;em&gt;Simulate Before Sending&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;NINeS 2026&lt;/td&gt;
&lt;td&gt;5 Tue&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;R9&lt;/td&gt;
&lt;td&gt;Li, Liu, Zhang et al. — &lt;em&gt;Revisiting RDMA Reliability&lt;/em&gt; &lt;em&gt;(Best Student Paper HM)&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;SIGCOMM 2025&lt;/td&gt;
&lt;td&gt;5 Thu&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;R10&lt;/td&gt;
&lt;td&gt;Shou, Liu, Nie et al. — &lt;em&gt;InfiniteHBD: Optical Circuit Switching for LLM&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;SIGCOMM 2025&lt;/td&gt;
&lt;td&gt;6 Tue&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;R11&lt;/td&gt;
&lt;td&gt;Xu, Sun, Tian, Zhang, Chen — &lt;em&gt;GREEN: Carbon-Efficient Scheduling for ML Clusters&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;NSDI 2025&lt;/td&gt;
&lt;td&gt;6 Thu&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;R12&lt;/td&gt;
&lt;td&gt;Bharadwaj, Ma, Liang, Farb, Rowe, Seshan — &lt;em&gt;OpenFLAME: Federated Spatial Naming&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;NINeS 2026&lt;/td&gt;
&lt;td&gt;7 Tue&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;R13&lt;/td&gt;
&lt;td&gt;Lai, Li, Wu et al. — &lt;em&gt;LeoCC: CC Robust to LEO Dynamics&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;SIGCOMM 2025&lt;/td&gt;
&lt;td&gt;7 Thu&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;R14&lt;/td&gt;
&lt;td&gt;Karimi, Fouladi, Sivaraman, Alizadeh — &lt;em&gt;Tight Loops, Smooth Streams&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;NINeS 2026&lt;/td&gt;
&lt;td&gt;8 Tue&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;R15&lt;/td&gt;
&lt;td&gt;Larrea, Shreedhar, Niemi, Sefiane, Marina — &lt;em&gt;BISCAY: Radio KPI Driven CC&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;NINeS 2026&lt;/td&gt;
&lt;td&gt;8 Thu&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;R16&lt;/td&gt;
&lt;td&gt;Ikram, Habib, Ali, Uzmi — &lt;em&gt;Don&amp;rsquo;t get caught, keep your Onions in a Vault&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;NINeS 2026&lt;/td&gt;
&lt;td&gt;9 Tue&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;R17&lt;/td&gt;
&lt;td&gt;Sengupta, Kim, Jubas, Apostolaki, Rexford — &lt;em&gt;Passive Telemetry for BGP Hijacks&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;NINeS 2026&lt;/td&gt;
&lt;td&gt;9 Thu&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;R18&lt;/td&gt;
&lt;td&gt;Kumar, Bustamante, Flores — &lt;em&gt;Who Holds the Steering Wheel?&lt;/em&gt;&lt;/td&gt;
&lt;td&gt;NINeS 2026&lt;/td&gt;
&lt;td&gt;10 Tue&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;</description></item><item><title>Reading Blog &amp; Presentations</title><link>https://aqualab.cs.northwestern.edu/course/cs440-f26/reading/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://aqualab.cs.northwestern.edu/course/cs440-f26/reading/</guid><description>&lt;div class="course-tabs"&gt;&lt;a href="https://aqualab.cs.northwestern.edu/course/cs440-f26/"
class="course-tab"&gt;
Syllabus
&lt;/a&gt;&lt;a href="https://aqualab.cs.northwestern.edu/course/cs440-f26/calendar/"
class="course-tab"&gt;
Calendar
&lt;/a&gt;&lt;a href="https://aqualab.cs.northwestern.edu/course/cs440-f26/reading/"
class="course-tab course-tab-active"&gt;
Reading Blog &amp;amp; Presentations
&lt;/a&gt;&lt;a href="https://aqualab.cs.northwestern.edu/course/cs440-f26/project/"
class="course-tab"&gt;
Reproduction Project
&lt;/a&gt;&lt;/div&gt;
&lt;h2 id="reading-blog"&gt;READING BLOG&lt;/h2&gt;
&lt;p&gt;Instead of submitting individual critiques on Canvas, each student will maintain a &lt;strong&gt;running reading blog&lt;/strong&gt; — a personal website where you post a short critique of the research paper before each class. Over the quarter, your blog becomes a cumulative record of how your understanding of the field develops.&lt;/p&gt;
&lt;h3 id="why-a-blog"&gt;Why a blog?&lt;/h3&gt;
&lt;p&gt;A blog post is not a homework submission — it is a piece of writing with an audience. The format encourages you to think more carefully about what you write, and the running nature of the blog lets you draw &lt;strong&gt;connections&lt;/strong&gt; between papers as the quarter progresses. By Week 9, you will have built a personal map of the research landscape.&lt;/p&gt;
&lt;h3 id="setup"&gt;Setup&lt;/h3&gt;
&lt;p&gt;We provide a GitHub Pages template you can fork and have running in five minutes:&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;→ &lt;a href="https://github.com/NU-AquaLab/cs440-reading-blog" target="_blank" rel="noopener"&gt;cs440-reading-blog template&lt;/a&gt;&lt;/strong&gt;&lt;/p&gt;
&lt;ol&gt;
&lt;li&gt;Fork the template repository (or click &amp;ldquo;Use this template&amp;rdquo;).&lt;/li&gt;
&lt;li&gt;Rename your copy to &lt;code&gt;cs440-blog&lt;/code&gt; (or any name you like).&lt;/li&gt;
&lt;li&gt;Go to &lt;strong&gt;Settings → Pages&lt;/strong&gt; and set the source to the &lt;code&gt;main&lt;/code&gt; branch.&lt;/li&gt;
&lt;li&gt;Edit &lt;code&gt;_config.yml&lt;/code&gt; — change &lt;code&gt;author&lt;/code&gt; to your name.&lt;/li&gt;
&lt;li&gt;&lt;strong&gt;Add the instructor as a collaborator:&lt;/strong&gt; Settings → Collaborators → add &lt;code&gt;fabianeb&lt;/code&gt;.&lt;/li&gt;
&lt;li&gt;Your blog will be live at &lt;code&gt;https://&amp;lt;your-username&amp;gt;.github.io/cs440-blog/&lt;/code&gt;.&lt;/li&gt;
&lt;/ol&gt;
&lt;p&gt;Your repo may be &lt;strong&gt;public or private&lt;/strong&gt; — that is up to you. GitHub Pages works with private repos on free student accounts (via &lt;a href="https://education.github.com/" target="_blank" rel="noopener"&gt;GitHub Education&lt;/a&gt;). The instructor will see your posts as a collaborator either way.&lt;/p&gt;
&lt;h3 id="post-structure"&gt;Post structure&lt;/h3&gt;
&lt;p&gt;Each post should address three things:&lt;/p&gt;
&lt;ol&gt;
&lt;li&gt;&lt;strong&gt;Key idea.&lt;/strong&gt; What problem is the paper solving and what is the core contribution? (2–3 sentences)&lt;/li&gt;
&lt;li&gt;&lt;strong&gt;Critique.&lt;/strong&gt; What is convincing, what is not, what is missing? (2–3 sentences)&lt;/li&gt;
&lt;li&gt;&lt;strong&gt;Connections.&lt;/strong&gt; How does this relate to earlier readings? Link to your previous posts where relevant. (1–2 sentences)&lt;/li&gt;
&lt;/ol&gt;
&lt;p&gt;That&amp;rsquo;s it — roughly 150–200 words. The template repository includes an example post on Misa et al., &lt;em&gt;&amp;ldquo;There is More to Internet Invariants Than Meets the Eye&amp;rdquo;&lt;/em&gt; (NINeS 2026).&lt;/p&gt;
&lt;h3 id="logistics"&gt;Logistics&lt;/h3&gt;
&lt;ul&gt;
&lt;li&gt;&lt;strong&gt;Due by 9:00 PM the evening before class.&lt;/strong&gt;&lt;/li&gt;
&lt;li&gt;One skip allowed per quarter without penalty.&lt;/li&gt;
&lt;li&gt;No blog post is required for background papers — only the research paper marked &lt;strong&gt;[R]&lt;/strong&gt; on the calendar.&lt;/li&gt;
&lt;/ul&gt;
&lt;hr&gt;
&lt;h2 id="presenting"&gt;PRESENTING&lt;/h2&gt;
&lt;p&gt;Each student presents one or two research papers during the quarter. A good presentation motivates the problem, explains the technical approach, presents the evaluation, and leads a discussion.&lt;/p&gt;
&lt;p&gt;Your presentation should include:&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;&lt;strong&gt;Introduction:&lt;/strong&gt; Context and motivation for the work. Any background the audience needs.&lt;/li&gt;
&lt;li&gt;&lt;strong&gt;Methods:&lt;/strong&gt; The technical approach — how did they do it?&lt;/li&gt;
&lt;li&gt;&lt;strong&gt;Results:&lt;/strong&gt; The main contributions and findings.&lt;/li&gt;
&lt;li&gt;&lt;strong&gt;Discussion:&lt;/strong&gt; An interpretation of the results — what does this mean for the field?&lt;/li&gt;
&lt;li&gt;&lt;strong&gt;Critique and class discussion:&lt;/strong&gt; End with two to three questions or issues worthy of further discussion. Your job as presenter is not only to explain the paper but to lead the class conversation.&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;Today, many papers have a video of the author&amp;rsquo;s conference presentation. Take advantage of this and watch it before attempting to read the paper.&lt;/p&gt;
&lt;p&gt;Recommended resource: S. Peyton Jones et al., &lt;a href="https://simon.peytonjones.org/great-research-talk/" target="_blank" rel="noopener"&gt;&lt;em&gt;How to Give a Great Research Talk&lt;/em&gt;&lt;/a&gt;, SIGPLAN Notices, 1993.&lt;/p&gt;</description></item><item><title>Reproduction Project</title><link>https://aqualab.cs.northwestern.edu/course/cs440-f26/project/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://aqualab.cs.northwestern.edu/course/cs440-f26/project/</guid><description>&lt;div class="course-tabs"&gt;&lt;a href="https://aqualab.cs.northwestern.edu/course/cs440-f26/"
class="course-tab"&gt;
Syllabus
&lt;/a&gt;&lt;a href="https://aqualab.cs.northwestern.edu/course/cs440-f26/calendar/"
class="course-tab"&gt;
Calendar
&lt;/a&gt;&lt;a href="https://aqualab.cs.northwestern.edu/course/cs440-f26/reading/"
class="course-tab"&gt;
Reading Blog &amp;amp; Presentations
&lt;/a&gt;&lt;a href="https://aqualab.cs.northwestern.edu/course/cs440-f26/project/"
class="course-tab course-tab-active"&gt;
Reproduction Project
&lt;/a&gt;&lt;/div&gt;
&lt;h2 id="reproduction-project-50"&gt;REPRODUCTION PROJECT (50%)&lt;/h2&gt;
&lt;p&gt;Working in teams of two, you will reproduce a key result from a published networking systems paper and write a structured critical assessment informed by that experience.&lt;/p&gt;
&lt;p&gt;Reproducing research results lets you gain a detailed, in-depth understanding of an important paper, the research area and underlying concepts, and its key results. You concentrate your energy on the most rewarding part of the research effort — building a prototype or experimental framework and carrying out the experiments — knowing that the problem is interesting and can deliver.&lt;/p&gt;
&lt;p&gt;You will start by selecting a paper you are interested in learning more about. After choosing a paper and sketching the idea of your project (what part of the work you want to reproduce and how you may go about it), you are strongly encouraged to contact the original authors, informing them of your intended project and asking for advice and suggestions. We will be happy to help you with this.&lt;/p&gt;
&lt;p&gt;&lt;em&gt;Inspired by Stanford CS244 and the &lt;a href="https://reproducingnetworkresearch.wordpress.com/" target="_blank" rel="noopener"&gt;Reproducing Network Research blog&lt;/a&gt;.&lt;/em&gt;&lt;/p&gt;
&lt;h3 id="the-critique-section"&gt;THE CRITIQUE SECTION&lt;/h3&gt;
&lt;p&gt;The final report must include a dedicated &lt;strong&gt;critique section&lt;/strong&gt; (~2 pages) grounded in what you found during reproduction. Address:&lt;/p&gt;
&lt;ol&gt;
&lt;li&gt;&lt;strong&gt;Claim fidelity.&lt;/strong&gt; Which claims held up? Which could not be verified, and why?&lt;/li&gt;
&lt;li&gt;&lt;strong&gt;Hidden conditions.&lt;/strong&gt; What must be true for the result to hold that the paper does not state explicitly?&lt;/li&gt;
&lt;li&gt;&lt;strong&gt;Underspecification.&lt;/strong&gt; What did the authors leave ambiguous, and how did you resolve it?&lt;/li&gt;
&lt;li&gt;&lt;strong&gt;Generalizability.&lt;/strong&gt; Does the result extend beyond the paper&amp;rsquo;s specific setup?&lt;/li&gt;
&lt;li&gt;&lt;strong&gt;Reviewer&amp;rsquo;s assessment.&lt;/strong&gt; Knowing what you now know, what would you ask the authors to change?&lt;/li&gt;
&lt;/ol&gt;
&lt;p&gt;A reproduction that succeeds but reveals underspecified conditions is just as valuable as one where the result fails to replicate.&lt;/p&gt;
&lt;h3 id="deliverables-and-timelines"&gt;DELIVERABLES AND TIMELINES&lt;/h3&gt;
&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Deliverable&lt;/th&gt;
&lt;th&gt;Due&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;Team formation&lt;/td&gt;
&lt;td&gt;Week 1 Thu&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Paper selection + project proposal&lt;/td&gt;
&lt;td&gt;Week 3 Tue&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;In-class proposal pitch (5 min)&lt;/td&gt;
&lt;td&gt;Week 4 Thu&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Midpoint report&lt;/td&gt;
&lt;td&gt;Week 6 Tue&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Final report (incl. critique section)&lt;/td&gt;
&lt;td&gt;Week 9 Thu&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Project presentations&lt;/td&gt;
&lt;td&gt;Finals Week&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p&gt;&lt;strong&gt;Format:&lt;/strong&gt; 6 pages max, HotNets format, appendices allowed.&lt;/p&gt;</description></item></channel></rss>