Course introduction

Class stuff and office hour

Prerequisites

  • Familiar with OS & networking
  • System-level programming experience
  • Comfortable with concurrency and threading

Polling of PhD and Master

Course readings

  • Lectures are based on research papers
  • Check webpage for schedules
  • Lectures assume you have read the assigned papers
  • No textbook

Important website

  • class schedule
  • piazza

How are you evaluated?

  • Labs 70 points
    • You must work alone on all assignments.
    • A single deadline: Nov 1
    • A bonus lab that can pump your score up to 80 (B+)
      • revision: 80 is now A-
    • Late policy: 10 off per day, at most 60 (after 6 days)
  • A bonus project 30 points
    • optional only if you finish at least lab3 by Oct 1
      • revision: you can only get a bonus score if you finish lab 1/2/3 before ddl (Nov 1) and get at least 60 points
    • you can group or you can do it individually, but the bonus points will be distributed evenly among you
    • I will release a few candidate projects which you can choose from
    • You can also propose your own, but it has to be related to distributed systems research and has to go through me first.
      • warning: a randomly pitched idea (e.g., let me show you what I did in my intern / past work / another class) will not be accepted.
    • You will likely receive 0, 10, 20, 30 bonus points depending on your implementation and presentation.
  • Grading standard
    • A: achieve > 90 in score
    • A-: achieve > 80 in score, or ranking 10%
      • revision: A-: achieve >= 80 in score, or ranking 10%
    • B+: score > 70, or ranking 30%
    • B: score > 60, or ranking 50%
    • B-: score > 50, or ranking 70%
    • C+: score > 40, or ranking 90%
    • C: score > 20, or ranking 95%
    • F: score <= 20 and last 5%
  • Other bonus
    • reporting a technical error I made gives you 2%, up to 20%
    • reporting a non-technical error I made gives you 0.5%, up to 5%
      • including grammar errors in any written text.
      • excluding grammar errors in any verbal communication unless it is a specific and repeated error.
    • Note: this will be added to your score after it is multiplied by the base score.
      • e.g., if you get 80 on your labs+project, and you have 3% bonus from correcting errors, your final score will be 82.4

Integrity policies

  • The work that you turn in must be yours.
  • You must acknowledge your influences.
  • You must not look at, or use, solutions from prior years or the Web, or seek assistance from the Internet.
  • You must take reasonable steps to protect your work.
  • You must not publish your solutions.
  • If there are inexplicable discrepancies between exam and lab performance, we will over-weight the exam and interview you.

Penalty

  • Violate policy -> Incomplete, report to the department
  • Attempt to negotiate on grading -> 10% off per each attempt

What are distributed systems?

  • Machines communicate to provide some service for applications
  • Multiple hosts
  • A local or wide area network

Why distributed systems?

  • ease-of-use (web, NFS)
  • availability/reliability (hardware/software failures)
  • scalable capacity (CPU, memory, storage)
  • modular functionality (authentication service)

Downside

  • “A distributed system is one in which the failure of a computer you didn’t even know existed can render your own computer unusable.” – Leslie Lamport

Main challenges/topics in distributed systems

  • Abstraction/Interface
    • different system requirements: file system/database/disk
    • simple, flexible, implementation-friendly
  • System architecture
    • data center / wide area
    • client-server / peer-to-peer
  • Fault Tolerance
    • backup/replication
    • backup fail-over
  • Consistency
    • keep replicas identical
    • keep replicas non-identical
  • Performance
    • throughput (parallelism/divide load)
    • latency (queuing, minimize critical path)