Disclaimer: This is the tentative plan. It will almost certainly change. I do not post this as a promise. Instead, I do it to offer you the flavor and direction of the course. Changes will be announced in class and this will occasionally be updated.

Week 1

• Procedural stuff and Intro
• Networks, overview
• Middleware: RPC and marshalling

Week 2

• Distributed Objects: Java's RMI,
• Distributed Object and interoperability: CORBA Architecture
• Distributed objects and interoperability: XML, XML-RPC, SOAP, WSDL and related technologies Project #1

Week 3

• Physical time and synchronization thereof, NTP
• Logical time, Lamport's algorithm, causality and vector time
• Multicast (FIFO, causally, and total ordered), Matrix time and garbage collection

Week 4

• Exam #1 (Through week 3)
• Concurrency control: intro, central, lamport time, voting, voting districts
• Concurrency control: ring, tree

Week 5

• Concurrency control: path compression; Coordinator election: intro
• Coordinator election: Ring, bully, invitation,
• Project discussion: Project #2

Week 6

• Transactions and distributed transactions, 2PC, 3PC
• Transactions and schedules, serial vs serializable, concurrency control by prevention: 2PL
• Transactions and concurrency control by avoidance (timestamp mechanism) and by validation

Week 7

• Impossibility of agreement: 2 Armys Byzantine Generals
• Replication: Conflict, static quorums, and ghosts
• Replication: Best-effort (epidemic and version vectors)

Week 8

• Review/Slack
• Exam #2 (Through replication)
• No class: Midsemester break

Spring Break

• No class

Week 9

• Recovery and checkpointing: coordinated and uncoordinated, recovery lines, Project #3
• Recovery and logging: sync vs. async, sender vs. receiver, adaptive, recovery algorithms
• Tasks, Threads, and Processes: a distributed view fault tolerance, and recovery

Week 10

• Processor allocation: N or 1/N*N, verview, central, hierarchical
• Processor allocation: IPC and, microeconomic approaches
• Process migration: Overview, VM, commc hannels, kernel state, idle worstations, weighing the costs and benefits

Week 11

• DSM: Overview, unit, protection, migration, &c
• DSM: Weaker techniques
• Traditional File Systems

Week 12

• DFS: Overview, issues, elements
• DFS: NFS vs AFS, Coda, GFS
• Clouds, Hadoop, MapReduce, Project #4

Week 13

• Hadoop file system
• More about Hadoop, MapReduce, &c

Week 14

• Distributed hashing, Peer-to-Peer systems
• Ubiquitous computing, overview
• Ad hoc and sensor networks

Week 15

• Security protocols
• "The World's Most Successful Distributed System"...DNS
• Exam #3

Week of May 8, 2007

• Final exams

Week of May 15, 2007

• Final exams