Cloud Computing EN.601.419/EN.601.619

Administrivia

  • Instructor: Soudeh Ghorbani
  • TAs/CAs: Bo Guan, Manish Sharma, Matthew Francis-Landau, Zhengyuan (Harley) Wang
  • Lecture time: Mondays and Wednesdays, 12-1:15pm
  • Location: Homewood Campus, Hodson 213
  • Office hours: Thursdays 4:30-5:30pm, Malone 223
  • Piazza discussion board

Course Description

Clouds host a wide range of the applications that we rely on today. In this course, we study common cloud applications, traffic patterns that they generate, critical networking infrastructures that support them, and core networking and distributed systems concepts, algorithms, and technologies used inside clouds. We will also study how today's application demand is influencing the network's design, explore current practice, and how we can build future's networked infrastructure to better enable both efficient transfer of big data and low-latency requirements of real-time applications. The format of this course will be a mix of lectures, discussions, assignments, an exam, and a project designed to help students practice and apply the theories and techniques covered in the course.
Prerequisites: EN.601.226 or permission. Recommended: a course in operating systems, networks or systems programming. Students can only receive credit for one of 601.419/619.

Acknowledgment: This course is influenced by and uses materials from Hopkins Cloud Computing Security, Google Cloud Platform Specialization, Amazon AWS Cloud Specialization, UIUC Cloud Specialization, UW Distributed Systems, and Stanford CS244. We particularly thank Joel Coffman and acknowledge the support from Google through a Google Platform grant.

Topics


We will cover the following topics:
  • Datacenters
  • Software-defined networks (SDN)
  • Programmable networks
  • Verification
  • Big data systems
  • Cloud storage
  • Virtualization

Grading Policy

The class is graded as follows:
  • Project (45%)
    • Proposal (5%)
    • Checkpoints (10%)
    • Midterm presentation (10%)
    • Poster presentation (10%)
    • Final paper (10%)
  • Paper reviews (30%)
  • Final exam (10%)
  • Two assignments (10%)
  • Participation (5%)

Project

The project is a main component of this course. The goal is to eihter build a functional system or conduct high-quality novel research related to cloud computing that, by the end of the semester, would be publishable as a paper in a top-quality workshop like HotNets or HotCloud, and when expanded to a full paper would be publishable in a top-quality conference. You may work in groups of 2 to 4 people. The steps in the project are as follows:
  • Project proposal: During the first month of the class, you should think about the topic you want to work on and find partners. You are welcome (and encouraged) to explore your own ideas. However, you can also talk with the instructor who will suggest some topics (you need to set an appointment). Submit a project proposal as a private note on Piazza. The proposal should be at most one page and inlcude each of the following:
    • the problem you plan to address
    • what will be your first steps to attack the problem
    • what is the most closely related work, with either (a) citations of at least 2 similar systems (if you are building a system), or (b) at least 3 academic paper citations (if you are working on a research project). You should explain why your proposed problem is different than those or why your proposed solution is better. You should actively search for related work, not just cite systems and papers that the instructor mentions.
    • Who the people on your team are and how you plan to partition the work among the team
  • The proposal can be short. It should simply demonstrate that you have a plausible project and know how to attack it. The instructor will give a grade for the proposal, and either approve the project or ask for a revision.
  • Checkpoints: Each project should have a website. We will have a lab to demonstrate how to build this. Every two weeks from the date that you submit your proposals, each team should post a status report on its website describing the progress since the last checkpoint and each member's contribution.
  • Midterm presentation: Give a presentation in class describing what problem you are solving, why existing approaches will not solve your problem, your solution approach, and your progress in your solution. You must demonstrate progress in your solution.
  • Final paper: This is a short paper suitable for submission to a workshop. It should clearly state the system being built or the problem being solved, its importance, related work, your approach, evaluation, and results, conclusion, discussion of limitations, and future work. The paper should be at most 8 pages. But you will be judged on results, not pagecount!
  • Poster presentation: At the end of the course, during the final exam period, we will have a poster session. This will be an opportunity for the instructor to ask questions about your project, and also for other students and faculty in the department to see the cool work that you've done.
Dates for the above steps will be announced on Piazza. In general, you are encouraged to meet with the instructor and seek advice on the project as often as you like.
Can a project be shared with another course's project or independent research? It is OK, and often a good idea, to work on a class project that complements your other ongoing projects and has a related topic. However, you should identify the piece of the larger project that you are working on for this course, with separate pieces for other courses. Check with your other instructors as well.

Paper Reviews

For each class, we will have one assigned paper that you should read prior to class and be ready to discuss during the class. You should submit one paper review for each class on Pizza publicly (so that it is visible to the instructor and other students) by 5:00pm the day before the lecture for which the paper was assigned. This review should be relatively short (about one paragraph). It should not summarize the paper or repeat it; we have all read the paper already. Instead your review should inlcude at least two comments on the paper that supply information not in the paper itself. For example, a comment might be:
  • a suggestion to build on or extend the paper's ideas in future work
  • a criticism of the paper
  • an advantage of the paper (not discussed in the paper)
  • an alternative solution for the solutions discussed in the paper
  • a response to another student's comment
You are encouraged to read and comment on the other students' reviews. However, please write down notes on your own thoughts independently prior to reading other students' reviews. Collaborating with other students to write reviews is not permitted. Your reviews should ideally include original ideas that do not appear in the other students' reviews. However, If you independently make similar points, that is acceptable. Each review will be given a score in the [0,2] range. We will ignore the 2 lowest paper review grades which means that you may skip any 2 paper reviews without affecting your grade. You will receive a deduction of one letter grade for missing more than 2 reviews.

Assignments

There will be 2 assignments during the semester which includes introducing several tools for experimental research in cloud computing and written questions covering roughly the first 1/3 and 2/3 of the course. You may work alone or in groups of 2 for assignments.

Late policy

For reviews, assignemnts, and project milestons: up to 24 hours, 30% deduction, with no submissions accepted past 24 hours late. We accommodate special medical circumstances, such as death in the family or hospitalization, with appropriate documentation. We cannot accommodate excuses such as "My laptop died".

Participation

You are expected to attend all sessions of the class. The general policy is that a student will automatically receive a deduction of one letter grade for missing more than 2 lectures. Class sessions combine lectures, discussions of reading, and presentations by students. In all cases, the class is focused around discussion. Please comment, question, and interact! I ask that you do not use laptops during class. This way, we will all be maximally engaged.

Final exam

There will be a final exam covering all the topics we discuss in the course as well as the papers that you review.

Hopkins Disability Accommodations

We are committed to providing access for all qualified students, including those with disabilities. For more information, you are encouraged to review the guidelines here and to contact the Office of Institutional Equity.

Academic Honesty and Cheating

The author of all writing, ideas, and other work must be clearly credited. For example, if your presentation of a past paper uses some slides from the author, you must credit the author. The standard penalty for a first instance of cheating is a grade of zero on the assignment in question, plus a reduction of one full letter grade in your final course grade. For details, please see the departmental honor code.

Project Ideas

How can you pick a good research project topic? Your taste for projects will evolve over years, but to get started, here are a few places to look.

Workshops and Conferences

Browse programs at top conferences to see current research topics. Workshops often contain early work on "hot" new directions, raising more questions than answers. These are good conferences and workshops to check out when looking for papers to present on a certain topic, or to see current areas of research when looking for project inspiration:

Survey Papers

Readings

The required papers are listed on the schedule. There is no required textbook. If you need a refresher for networks and distributed systems, you might take a look at standard books such as
  • Tanenbaum and Steen, Distributed Systems: Principles and Paradigms
  • Peterson and Davie, Computer Networks
  • Kurose and Ross, Computer Networking: A Top-Down Approach
Date
Topic
Readings, Assignments, and Notes
Fundamental Architectural Principles and Key Protocols
September 4 Introduction
Internet & IP
Layering pinciple

No reviews due

Syllabus

Optional:The Design Philosophy of the DARPA Internet Protocols (D. Clark, SIGCOMM 1988)

September 6

Transport layer
End-to-end principle

No reviews due

Where to look for project ideas

End-to-End Arguments in System Design (J.H. Saltzer, D.P. Reed and D. Clark, ACM Trans. on Computer Systems, Vol. 2, No. 4, Nov 1984, pp. 277-288)

Presentation topic preferences due (5pm)

September 11 Congestion control

Congestion Avoidance and Control (Jacobson, SIGCOMM 1988)

September 13 Global connectivity

BGP Routing Policies in ISP Networks (Caesar and Rexford, IEEE Network Magazine, Nov/Dec 2005).

Optional: Nation-State Hegemony in Internet Routing (Edmundson et al., COMPASS 2018), Espresso (Yap et al., SIGCOMM 2017), Internet at the Speed of Light (Singla et al., HotNets 2014)

Software Defined Networks (SDN)
September 18 SDN

OpenFlow: Enabling Innovation in Campus Networks (McKeown wt al., CCR 2008)

Optional: The Road to SDN: An Intellectual History of Programmable Networks (Feamster et al., CCR 2014), Google's Move to SDN (Communications of the ACM, March 2016, Vol. 59 No. 3)

September 20 Software-defined routing [Michael]

B4: Experience with a Globally-Deployed Software Defined WAN (Jain et al., SIGCOMM 2013)

Optional: B4 and After: Managing Hierarchy, Partitioning, and Asymmetry for Availability and Scale in Google's Software-Defined WAN (Hong et al., SIGCOMM 2018)

Datacenters
September 25 Datacenter network architecture

Scalable, Commodity DC Net Arch (Al Fares et al., SIGCOMM 2008)

Optional: F10 (Liu et al., NSDI 2013), Jellyfish (Singla et al., NSDI 2012)

Project proposals due (11:59pm)

September 27 No class -- Soudeh traveling
October 2 Datacenters today [Robin]

Google Datacenters (Singh et al., SIGCOMM 2015)

Optional: Facebook Datacenters (Roy et al., SIGCOMM 2015)

October 4 Network virtualization

VMware NVP (Koponen et al., NSDI 2014)

Optional: Google Andromeda (Dalton et al., NSDI 2018)

Middleboxes
October 9 Middleboxes [Steven]

An Untold Story of Middleboxes in Cellular Networks (Wang et al., SIGCOMM 2011)

Optional: Making Middleboxes Someone Else's Problem (Sherry et al., SIGCOMM 2012)

Network Verification
October 11 Static verifiers

HSA (Kazemian et al., NSDI 2012)

Optional: VeriFlow (Khurshid et al., NSDI 2013)

October 16 Dynamic verifiers [Justin]

Minesweeper (Beckett et al., SIGCOMM 2017)

Optional: A Formally Verified NAT (Zaostrovnykh et al., SIGCOMM 2017)

Reliability and Security
October 18 Failures and routing convergence [Steven]

Evolve or Die (Govindan et al, SIGCOMM 2016)

Optional: Delayed Internet Routing Convergence (Labovitz et al., SIGCOMM 2000), Internet Routing Instability (Labovitz et al., Transactions on Networking, Vol. 6, No. 5, 1998)

October 23 Reliable routing

Fibbing (Vissicchio et al., SIGCOMM 2015)

Optional: Consensus Routing (John et al., NSDI 2008)

October 25 No class -- Soudeh traveling
Work on the midterm presentations
October 30 Secure routing

How Secure are Secure Interdomain Routing Protocols? (Goldberg, SIGCOMM 2010)

Optional: Survey of BGP security (Butler, Proc IEEE, 2010)

Congestion Control
November 1 Congestion control in datacenters

DCTCP (Alizadeh et al., SIGCOMM 2010)

Optional: MPTCP (Wischik et al., NSDI 2011), pFabric (Alizadeh et al., SIGCOMM 2013)

November 6 Modern congestion control [Michael]

PCC (Dong et al., NSDI 2015)

Optional: Remy (Winstein et al., SIGCOMM 2013), PCC Vivace (Dong et al., NSDI 2018)

Forwarding
November 8 Forwarding

50-Gb/s IP Router (Partridge et al., ToN 1998)

Optional: PacketShader (Han et al., SIGCOMM 2010)

November 13 Project midterm presentations

Project midterm presentations due: November 8, 11:59pm

November 15 Programmable forwarding

RMT (Bosshart et al., SIGCOMM 2013)

Optional: Packet Transactions (Sivaraman et al., SIGCOMM 2016)

Thanksgiving Vacation
Special Topics
November 27 Internet Exchange Points (IXPs)

Anatomy of a Large European IXP (Ager et al., SIGCOMM 2012)

Optional: ISP Interconnectivity in Africa (Gupta et at., PAM 2014)

November 29 Censorship [Justin]

Towards a Comprehensive Picture of the Great Firewall's DNS Censorship (Anonymous, FOCI 2014)

Optional: Global Measurement of DNS Manipulation (Pearce et al., USENIX Security Symposium 2017), Networks Should Run Themselves (Feamster and Rexford, 2018)

December 4 Content delivery networks (CDNs) [Robin]

The Serving Infrastructure of a Large CDN (Wohlfart et al., SIGCOMM 2018)

Optional: A Case for a Coordinated Internet Video Control Plane (Liu et al, SIGCOMM 2012)

December 6 Final Project Presentations

Optional: Networks Should Run Themselves (Feamster and Rexford, 2018)

TBD Project papers due