15-440/640 Distributed Systems

Spring 2021

Readings  Projects  Problem Sets 

Overview

15-440 is an introductory course in distributed systems. The emphasis will be on the techniques for creating functional, usable, and scalable distributed systems. To make the issues more concrete, the class includes several multi-week projects requiring significant design and implementation

The goals of this course are twofold. First, for students to gain an understanding of the principles and techniques behind the design of distributed systems, such as locking, concurrency, caching, prefetching, scheduling, and communication across the network. Second, for students to gain practical experience designing, implementing, and debugging real distributed systems.

The major themes this course will teach include:

Grading

All coursework is done individually. There are no teams or project partners. For in-person offerings of this course (pre-COVID and, hopefully, post-COVID) the evaluation was based on projects (45%), problem sets (20%), mid-term exam (15%), and final exam (20%). For the Spring 2021 offering of the course, because of COVID constraints, the exams are replaced by cumulative problem sets. These are like open-book exams, but without the time pressure. As their name implies, they include all the material covered in class up to that point. We refer to the other problem sets as topical problem sets since they focus on topics. So the Spring 2021 evaluation will be as follows (all weights approximate, within a 5% range):

Learning Outcomes

We expect students to gain a deep understanding, fluency in reasoning, and hands-on implementation skills of the following core systems concepts in distributed systems:

  1. Communication and remote procedure call; control semantics and language limitations; exactly-once, at-most-once, at-least-once; serialization and de-serialization; end-to-end argument and its application to real systems; integration with threading; concurrency of operations

  2. Data caching and one-copy semantics; cache consistency protocols and implementation tradeoffs; origins of temporaland spatial locality; cache quality metrics; application-specific consistency protocols; prefetching: benefits and risks; extraction of hints; buffer bloat

  3. Failures in distributed systems: origins and empirical studies; fail fast and Byzantine failures; fundamental limits of failure resilience

  4. Fault tolerance: atomic transactions; ACID property; implementation challenges; shadowing, intentions lists and write-ahead logging; tradeoffs in physical logging and operation logging; nested transactions; distributed transactions

  5. Consensus and blockchain: unanimity (two-phase commit); majority (leader election, Paxos); Byzantine (single-shot and Dolev-Strong); state machine replication and Streamlet; Bitcoin

  6. Common programming paradigms such as Map-Reduce, MPI and GraphLab

  7. (Only if time permits): Achieving high availability: voting-based preservation of one-copy semantics; taxonomy of replication strategies: pessimistic and optimistic approaches; read-write and write-write conflicts; server-client and peer-to-peer strategies; caching and disconnected operation; resolving conflicts; exploiting low bandwidth to improve availability

Course Logistics

Professors

Name Office Hours Office Phone Andrew email
Mahadev Satyanarayanan (Satya)   Tue 1:00 - 3:00 pm GHC-9123   x8-3743  satya
Padmanabhan Pillai (Babu) Wed 11:00 am - 1:00 pm   GHC-9232    pspillai
Runting Shi (Elaine) Mon 4:00 - 6:00 pm CIC-2217   rshi
         

Teaching Assistants

Name Office Hours Andrew email  
Nathan Ang Thu 6:00-8:00 pm nathanan
Junwon Chang (Joseph)   Sat 9:00-11:00 am junwonc
Wenxin Ding (Freda) Fri 10:00-12:00 wenxind
Timothy Ganger Sat 4:00-6:00 pm tganger
Ziying He Fri 5:00-7:00 pm ziyingh
Roger Iyengar Wed 1:00-3:00 pm raiyenga
Ishaan Jaffer Thu 2:00-4:00 pm ijaffer
Ibnul Jahan Tue 4:00-6:00 pm iej
Chen Jin (Crystal) Fri 8:00-10:00 chenj
Yajin Li Mon 10:00-12:00 pm  yajinl
Diego San Miguel Fri 2:00-4:00 pm dsanmigu
Riccardo Santoni Mon 8:00–10:00 pm rsantoni
Yiwen Song (Victor) Thu 9:00-11:00 pm yiwenson
Haithem Turki Wed 3:00-5:00 pm hturki
Clarissa Xu Tue 6:30-8:30 pm csxu
     

Lectures

Recitations

Course Notes

Will be placed in the course AFS area at: /afs/andrew/course/15/440/classnotes after each class. These notes are for your personal use only. Please do not distribute them.

Textbooks and Optional Readings

There are no required textbooks. Here are three good references to use for optional reading:

In addition, the “Readings” link at the top of this web page has some specifc optional readings that we will mention at different points in the lectures. The pdf’s of these optional readings are available on this course website.

Course Policies

Prerequisites

Because this course has a big project component, you must be proficient in C and Java programming on UNIX systems. It is required that you have taken 15-213 and gotten a “C” or higher since many of the programming skills you will need are taught in that course. If you received a C in 15-213, you must meet with your academic advisor to discuss your background before taking 15-440, perhaps taking an additional course to sharpen your systems skills.

Policy on Academic Integrity

The Carnegie Mellon University Policy on Academic Integrity applies to this course. All students are expected to carefully review this policy and to adhere to it for all aspects of this course.

Guidance on Collaboration:

Guidance on Sharing: You may not supply work that you complete during 15-440 to other students in future instances of this course or for use in future instances of this course (just as you may not use work completed by students who’ve taken the course previously).

Recordings

For Spring 2021, Zoom sessions of each lecture and recitation will be recorded by CMU Computing Services and posted to the Canvas page for this course. All other recordings are prohibited.

Limit on Using TA time

To be fair to everyone, especially when there is a long line of students waiting for a TA’s attention, there will be a limit of 10 minutes on all consultations. If a student is not done at the end of 10 minutes, he/she goes back to the end of the line before getting more time with the TA. Be prepared before you meet with a TA. If you need help finding a bug, narrow and simplify the problem down in advance of meeting with the TA.

Piazza Policy

This course uses the Piazza website for answering questions: here is the Piazza page for this course.

Think of piazza as raising your hand in class and asking a question. No question is too stupid to ask, so don’t be afraid. For every person who asks a question, there are probably many others to whom the same question has already arisen or will arise soon. The answer to your question may benefit them as well. In addition, there may be some people to whom your question did not occur. By asking the question, you are helping them see a subtlety that they may have not seen before. Direct email to the instructors will not be answered.

At all times, we expect you to use your good judgement in your Piazza posts (questions as well as answers to the questions of fellow students). Part of the learning process is struggling with the material until you arrive at the right insight for you to understand it. Posting too much detail in response to a request for assistance can impair learning. On the other hand, sometimes it’s great to be nudged in the right direction when you’re not able to get out of a rut. And, of course, misunderstandings of the assignment or tools available should be helped rapidly. Please use your best judgement when posting to the Piazza site, as if you were collaborating with your friends in person. A few rough guidelines:

Examples of good things to post and answer:

Examples of bad things to post or answer:

We expect you to have made reasonable effort to think for yourself before posting a piazza question. This is especially true with respect to debugging your code. Did you try the man pages? Did you do a Google search for possibly relevant resources? Did you look at the previous questions people have already asked, and at the answers provided? Did you insert printf’s and try to understand what is going on with your code?

Don’t use autolab as a debugging tool. We expect you to have made reasonable effort to get your code debugged before submitting to autolab. Creating test cases and stress testing your code is part of what a project is all about. Without putting in that effort, you are missing an important part of the learning opportunity in the course. Submission to autolab should be the last step of a process in which you have tested, debugged and improved your code to the furthest extent of your abilities. Sending an autolab dump in a piazza post and saying “please help” is an egregious violation of piazza etiquette.

Private posts on Piazza are not supported. This is a policy decision for this class. Remember, posting on piazza is similar to raising your hand and asking a question. Other students benefit from your asking the question, and seeing the instructors’ reply. We do allow your posts to be anonymous to fellow students, if you so choose. That is already a degree of privacy beyond what is possible when asking a question in class. For the genuinely rare occasions when you need to make a private request that is not related to course content, a special private mailing list has been created.

For requests that truly need to be private: send email to 15-440-student-private@lists.andrew.cmu.edu and one of the instructors will reply. Email to this list involving course content (e.g., clarifications to class material) will be ignored; you should be posting such questions on Piazza.

Policy on Late Submissions

For projects:

For problem sets: No late submissions are accepted, with or without penalty. Make sure you submit on time.

Style Guide for Projects

In addition to testing your code’s functionality, we will also be reserving a portion of each project’s points for its style and readability. The most important thing is a consistent and legible style. We are mostly looking to see that you chose a style that is readable and reasonable, and that you use the same style consistently throughout a project. Use common sense: don’t have 500-character lines of code, don’t name your variables foo (unless that make sense in its context), and avoid mixing case conventions randomly.

We will be looking for the following things:

Here is a Google style guide that may be helpful.

Wellbeing

Here are some tips for wellbeing.