Schedule

CS677 Distributed Systems
Spring 2006

Schedule

Schedule

The following schedule is tentative. Please check frequently and refresh your browser.

Week	Tuesday	Thursday	Assignment	Readings & Notes
1/31, 2/2	Introduction	2 Generals Problem		Week 0 Notes Why building distributed systems using unreliable components is hard.
2/7, 2/9	Time, Ordering and Global Predicate Detection	Lamport Clocks, Snapshot Protocol		Week 1 Notes How to represent and reason about time in the absence of a globally synchronized clock. O. Babaoglu and K. Marzullo Consistent Global States of Distributed Systems K. M. Chandy and L. Lamport Distributed Snapshots: Determining Global States of Distributed Systems
2/14, 2/16	Causal Order Vector Clocks	Detecting non-stable properties	Homework 1	Week 2 Notes How to detect causal relationships between events.
2/23		Causality and predicate detection wrap-up		Week 3 Notes Homework for no grade: understand how to compute "Possibly" and "Definitely" using vector clocks.
2/28, 3/2	Consistency	3-Phase Commit	Homework 1 due before class on 2/28	Week 4 Notes How to preserve consistency correctness. Bernstein, Goodman and Hadzilacos Distributed Recovery D. Skeen Non-blocking Commit Protocols D. Skeen Determining the Last Process to Fail
3/7, 3/9	Atomic Commit contd.	Intro to Fault-Tolerance	Project Part I Homework 2	Week 5 Notes I Week 5 Notes II How to preserve atomic semantics amidst faults. E. Elnozahy, L. Alvisi, Y.M.Wang, and D.B. Johnson A Survey of Rollback Recovery Protocols in Message Passing Systems L. Alvisi, K. Bhatia, and K. Marzullo Tracking Causality in Causal Message Logging Protocols
3/14, 3/16	Fault-tolerance contd.	The State Machine Approach	Homework 2 due 3/19	Week 6 Notes Consistency Notes A state machine model to emulate a single node system. F.B. Schneider Implementing Fault-Tolerant Services Using The State Machine Approach: A Tutorial T. C. Bressoud, F. B. Schneider Hypervisor-based Fault Tolerance
3/28	State machine wrap-up Consistency models	Consensus Midterm		How to agree.

4/4,4/6	Terminating Reliable Broadcast (TRB)	TRB		Week 7/8 How to agree given a leader. L. Lamport, R. Shostak, and M. Pease The Byzantine Generals Problem S. Srikanth and S. Toueg Simulating Authenticated Broadcasts to Derive Simple Fault-Tolerant Algorithms V. Hadzilacos and S. Toueg A Modular Approach to Fault-Tolerant Broadcasts and Related Problems
4/11,4/13	Arbitrary Failures with message authentication	AF without MA, *cast		Week 9 Notes Fisher, Lynch, Paterson Impossibility of Consensus PODC 2001 Influential paper award announcement
4/18,4/20	Impossibility of consensus (FLP)	Paxos		Week 10 Notes L. Lamport: The Part-Time Parliament L. Lamport: Paxos Made Simple T. Chandra and S. Toueg Unreliable Failure Detectors for Reliable Distributed Systems
4/25,4/27	Practical Byzantine Fault Tolerance (PBFT)	PBFT	Project Part II	Week 11, BASE, Separating M. Castro and B. Liskov: Practical Byzantine Fault-tolerance M. Castro and B. Liskov, Proactive Recovery in a Byzantine Fault-Tolerant System R. Rodriguez, M. Castro and B. Liskov, BASE: Using abstraction to improve fault-tolerance J. Yin, J.P. Martin, A. Venkataramani, L. Alvisi and M.Dahlin. Separating Agreement from Execution in Byzantine Fault-Tolerant Systems.
5/2,5/4	Replication / Distributed File Systems (DFS)	DFS overview, PRACTI replication	Homework 3 (due 5/15)	Week 12 Karin Petersen , Mike J. Spreitzer , Douglas B. Terry, Marvin M. Theimer , and Alan J. Demers, Flexible Update Propagation for Weakly Consistent Replication Xerox Parc's Bayou webpage N. Belaramani, M. Dahlin, L. Gao, A. Nayate, A. Venkataramani, P. Yalagandula, J. Zheng, PRACTI Replication
5/9,5/11	Guest lecture by Emery Berger: Replication to tolerate memory errors	Student Presentations Yun: Information retrieval in delay-tolerant networks Zongfang: A Logic of Authentication		Week 13 DieHard: Probabilistic Memory Safety for Unsafe Languages, Emery D. Berger and Benjamin G. Zorn Burrows, Abadi, Needham, A Logic of Authentication
5/16	Student Presentations Alodeep: Mesh-based content dissemination Vitaly: Fault-scalable byzantine fault-tolerant services (sosp 2005)			Week 14 SplitStream: High-Bandwidth Multicast in Cooperative Environments, M. Castro, P. Druschel, A. Kermarrec, A. Nandi, A. Rowstron, A. Singh Bullet: High Bandwidth Data Dissemination Using an Overlay Mesh , D. Kostic, A. Rodriguez, J. Albrecht, A. Vahdat Fault-scalable Byzantine Fault-Tolerant Services, Michael Abd-El-Malek, G. Ganger, G. Goodson, M. Reiter, J. Wylie
5/18			Take-home Final
5/23, 5/25	Final due	Project report due

13	Wait-free Synchronization	Wait-free Synchronization	Implement a wait-free data structure	Week 11 Notes M. Herlihy Wait-free Synchronization M. Herlihy A Methodology for Implementing Highly Concurrent Data Objects P. Jayanti On the robustness of Herlihy's hierarchy
14	Randomized consensus	Quorum Systems		Week 12 Notes M. Ben-Or Another advantage of free choice: Completely Asynchronous Agreement Protocols D. Gifford Weighted Voting for Replicated Data H. Garcia-Molina and D. Barbara How to Assign Votes in a Distributed System
15	Epidemic protocols	Thanksgiving	Homework 3	Week 13 Notes A. Demers, D. Greene, C. Hauser, W. Irish, J. Larson, S. Shenker, H. Sturgis, D. Swinehart, and D. Terry. Epidemic algorithms for replicated database maintenance. K.P. Birman, M. Hayden, O. Ozkasap, Z. Xiao, M. Budiu, and Y. Minsky. Bimodal Multicast. P. Th. Eugster, R. Guerraoui, S.B. Handurukande, P. Kouznetsov, A.-M. Kermarrec. Lightweight Probabilistic Broadcast.
16	Byzantine Quorum Systems	Byzantine Quorum Systems		Week 14 Notes D. Malkhi and M. Reiter Byzantine Quorum System J.P. Martin, L. Alvisi, and M. Dahlin Minimal Byzantine Storage

Acknowledgement: The material for this class liberally borrows from material for a similar class taught by Lorenzo Alvisi at UT Austin. Other sources include Prashant Shenoy and Ken Birman.