Old things that used to be going on.
The class will require you to
read papers
and
do a final project.
You will also be expected to
participate in discussion.
Finally, for at least one paper, you will be expected to
take notes and write them up carefully.
Hence, the basic idea is to read, discuss, and learn about distributed systems. Then, once we all have a handle on some of the concepts and systems, you will apply what you have learned and produce a set of terrific final projects. We will get going on the projects relatively early so as to better enable you to do an excellent job.
1/21 First day Class overview 1/28 Communication U-net , Linda , IVY 2/4 General systems Locus , Sprite 2/11 Services TACC , Porcupine , DDS 2/18 Storage Systems xFS , Petal , Frangipani 2/25 Memory Management Cooperative Caching , GMS 3/3 Scheduling LARD , Migration (I) , Migration (II) 3/10 Peer-to-peer Pangaea , Ivy 3/17 Project interlude Work on projects 3/24 Project interlude Work on projects 3/31 Project interlude Work on projects 4/7 Time Time and Clocks , Snapshots 4/14 Replication Consistency , Optimistic 4/21 Project Updates 15 minute status updates 4/28 Failures Fail-stop Processors , Byzantine Generals 5/5 Consensus Paxos , Simplified , Disk
Read the assigned papers BEFORE class. Without doing so, discussion is a little more difficult.
Form a discussion group. Discussion groups are useful in developing the ability to critically evaluate research, but are not mandatory in this course.