Advanced Computer Networks

Syllabus

  1. Instructor: Wenfei Wu
  2. Class Time: 9:00-11:59 AM on Monday
  3. Textbook: Computer Networks (Sixth Edition), by Andrew S. Tanenbaum, Nick Feamster, and David Wetherall
  4. Office Hour: 2:00-4:30 PM on Monday
  5. Grading: [30%] Quiz * 3; [30%] In-class Presentation; [40%] Research Project
  6. Slides&Videos: link

Calendar

The instructor would fill in the topic into the calendar as the class proceeds.

Week Date Topic
1 9/11 Overview
2 9/18 Physical Layer
3 9/25 Data Link Layer
4 10/2 No Class [National Day]
5 10/9 Network Layer (Quiz 1)
6 10/16 Transport Layer
7 10/23 Application Layer
8 10/30 Network Security (Quiz 2)
9 11/6 Multimedia
10 11/13 Data Center Networks (Quiz 3)
11 11/20 In-Network Computing
12 11/27 Research Topics
13 12/4 Research Topics
14 12/11 Research Topics
15 12/18 Research Topics
16 12/25 Project Presentation

Topics and Reading

0. Guidance

  1. [Paper Reading] How to read a paper, SIGCOMM 2007. Link
  2. [Graduate Study] How to Have a Bad Career in Research/Academia. Link
  3. [Research] How to Do Great Research. Link

1. Principles

  1. [End-to-End] End-To-End Argument in System Design, ACM 1984. Link
  2. [Layering]

2. Physical Layer

  1. [Coding] XORs in the air: practical wireless network coding, SIGCOMM06. Link

4. Network Layer

  1. [Switch] Forwarding metamorphosis: fast programmable match-action processing in hardware for SDN, SIGCOMM13. Link, Presented by Yuchen Xu
  2. [Software Switch] The Design and Implementation of Open vSwitch, NSDI15. Link, Presented by Weize Gao
  3. [Routing] Routing on Multiple Optimality Criteria, SIGCOMM20. Link, Video
  4. [IXP] An Industrial-Scale Software Defined Internet Exchange Point, NSDI16. Link, Audio
  5. [BGP] An analysis of BGP convergence properties, SIGCOMM99. Link
  6. [Middlebox] mOS: A Reusable Networking Stack for Flow Monitoring Middleboxes, NSDI17. Link
  7. [Internet] Internet Inter-Domain Traffic, SIGCOMM10. Link
  8. [Internet] Seven years in the life of Hypergiants’ off-nets, SIGCOMM21. Link, Video
  9. [IPv6]
  10. [SDN]

5. Transport Layer

  1. [RPC] Datacenter RPCs can be General and Fast, NSDI19. Link
  2. [Socket] Socksdirect: datacenter sockets can be fast and compatible, SIGCOMM19. Link, Video
  3. [Congestion Control] Data-Center TCP (DCTCP), SIGCOMM10. Link
  4. [User-Level Stack] mTCP: a Highly Scalable User-level TCP Stack for Multicore Systems, NSDI14. Link, Video
  5. [QUIC]

6. Application Layer

  1. [DHT] Chord: A scalable peer-to-peer lookup service for internet applications, SIGCOMM01. Link
  2. [HTTP]
  3. [DNS]

7. Network Security

  1. [Cipher]
  2. [Authentication]
  3. [Public Key]
  4. [PKI]
  5. [IPSec/TLS/HTTPS]

8. Multimedia

  1. [QoS] Multi-Resource Fair Queueing for Packet Processing, SIGCOMM12. Link
  2. [Rate Limiter] Scalable On-Switch Rate Limiters for the Cloud, INFOCOM21. Link
  3. [Video] Learning in situ: a randomized experiment in video streaming, NSDI20. Link
  4. [Scheduling]

9. Network Management

  1. [Architecture] A clean slate 4D approach to network control and management, SIGCOMM 2005. Link
  2. [Monitor]
  3. [Planning] Understanding Lifecycle Management Complexity of Datacenter Topologies, NSDI19. Link
  4. [Development] NFD: Using Behavior Models to Develop Cross-Platform Network Functions, INFOCOM21. Link
  5. [Diagnosis] Virtual network diagnosis as a service, SoCC13. Link
  6. [Diagnosis] CTF: Anomaly Detection in High-Dimensional Time Series with Coarse-to-Fine Model Transfer, INFOCOM21. Link

10. Data Center Networks

  1. [Architecture] VL2: a scalable and flexible data center network, SIGCOMM09. Link
  2. [Architecture] A Scalable, Commodity Data Center Network Architecture, SIGCOMM08. Link
  3. [Architecture] Re-architecting datacenter networks and stacks for low latency and high performance, SIGCOMM17. Link
  4. [Transport] Swift: Delay is Simple and Effective for Congestion Control in the Datacenter, SIGCOMM20. Link
  5. [Traffic Engineering] SecondNet: a data center network virtualization architecture with bandwidth guarantees, CoNEXT10. Link
  6. [Load Balancing] CONGA: distributed congestion-aware load balancing for datacenters, SIGCOMM14. Link
  7. [Fault Tolerance] F10: A Fault-Tolerant Engineered Network, NSDI13, Link
  8. [Routing]
  9. [Security]
  10. [Application]

11. Software-Defined Networks

  1. [Switch] Using trio: juniper networks’ programmable chipset - for emerging in-network applications, SIGCOMM22. Link
  2. [Architecture] Ethane: Taking control of the Enterprise, SIGCOMM 2007. Link
  3. [Load Balancer] A High-Speed Load-Balancer Design with Guaranteed Per-Connection-Consistency, NSDI20. Link
  4. [QoS] Twenty Years After: Hierarchical Core-Stateless Fair Queueing, NSDI21. Link
  5. [RPC] R2P2: Making RPCs first-class datacenter citizens, ATC19. Link
  6. [RPC] NetRPC: Enabling In-Network Computation in Remote Procedure Calls, NSDI23. Link
  7. [Application] ATP: In-network Aggregation for Multi-tenant Learning, NSDI21. Link
  8. [Application] A Generic Service to Provide In-network Aggregation for Key-value Streams, ASPLOS23. Link
  9. [Management] FAst in-network GraY failure detection for ISPs, SIGCOMM22. Link

12. Network Verification and Repair (Optional)

  1. [Route Verification]
  2. [Middlebox Verification]
  3. [Repair]

Comments