| 1 |
Tue 03/24 |
Introduction & Motivation
Slides & Discussion |
Suggested Reading:
DDIA Ch. 1
DS4 Ch. 1
Optional Readings:
Dean & Barroso — The Tail at Scale |
HW1 Out |
|
Thu 03/27 |
Processes, Threads, and RPC
Slides & Discussion |
Required Reading:
DS4 3.1, DDIA Ch. 4
Graduate Reading:
Waldo et al. — A Note on Distributed Computing
Optional Readings:
DS4 Ch. 3.2-3.6
Birrell & Nelson — Implementing Remote Procedure Calls |
|
| 2 |
Tue 03/31 |
Logical Time and Coordination
Slides & Discussion |
Required Reading:
DDIA Ch. 8 §"Unreliable Clocks"
DS4 Ch. 5.2
Graduate Reading:
Lamport — Time, Clocks, and the Ordering of Events
Optional Readings:
DDIA Ch. 9 §"Ordering Guarantees"
Mattern — Virtual Time and Global States |
|
| 2 |
Thu 04/02 |
Failures and Fault Models
Slides & Discussion |
Required Reading:
DDIA Ch. 8
Graduate Reading:
Chandra & Toueg — Unreliable Failure Detectors
Optional Readings:
Fischer, Lynch, Paterson — FLP Impossibility
Lamport et al. - The Byzantine Generals Problem
Hayashibara et al. - The ϕ Accrual Failure Detector |
|
| 2 |
Fri 04/03 |
|
|
HW1 Due |
| 3 |
Tue 04/07 |
Replication
Slides & Discussion |
Required Reading:
DDIA Ch. 5
Graduate Reading:
van Renesse & Schneider — Chain Replication
Optional Readings:
Oki & Liskov — Viewstamped Replication
Terrace & Freedman — Object Storage on CRAQ |
|
| 3 |
Thu 04/09 |
Partitioning
Slides & Discussion |
Required Reading:
DDIA Ch. 6
Graduate Reading:
DeCandia et al. — Dynamo: Amazon's Highly Available Key-Value Store
Optional Readings:
Chang et al. — Bigtable |
HW2 Out |
| 4 |
Tue 04/14 |
Consistency Models
Slides & Discussion |
Required Reading:
DDIA Ch. 9 p321–352
Graduate Reading:
Herlihy & Wing — Linearizability (§1-3 Only, Proofs Optional)
Optional Readings:
Gilbert & Lynch — Brewer's Conjecture and the CAP Theorem
Terry et al. — Session Guarantees for Weakly Consistent Replicated Data
Vogels - Eventually Consistent |
|
| 4 |
Thu 04/16 |
Consensus I (Paxos)
Slides & Discussion |
Required Reading:
DS4 §8.2.4
Graduate Reading:
Lamport — Paxos Made Simple
Optional Readings:
Lamport — The Part-Time Parliament
Chandra et al. — Paxos Made Live |
|
| 4 |
Sun 04/19 |
|
|
HW2 Due |
| 5 |
Tue 04/21 |
Consensus II (Raft)
Slides & Discussion |
Required Reading:
DDIA Ch. 9 §"Fault-Tolerant Consensus" (p364-369)
Graduate Reading:
Ongaro & Ousterhout — In Search of an Understandable Consensus Algorithm (Raft)
Optional Readings:
Howard et al. — Flexible Paxos: Quorum Intersection Revisited |
|
| 5 |
Thu 04/23 |
Distributed Transactions
Slides & Discussion |
Required Reading:
DDIA Ch. 7 §"The Slippery Concept of a Transaction" (p221-228)
DDIA Ch. 9 §"Distributed Transactions and Consensus (upto "Fault Tolerant Consensus") (p352-360)
Graduate Reading:
Gray & Lamport — Consensus on Transaction Commit (§1-5, proofs optional)
Optional Readings:
Helland — Life Beyond Distributed Transactions: An Apostate's Opinion |
HW3 Out |
| 6 |
Tue 04/28 |
Distributed File Systems |
Required Reading:
DDIA Ch. 10 §"MapReduce and Distributed Filesystems"
Graduate Reading:
Ghemawat et al. — The Google File System
Optional Readings:
Shvachko et al. — The Hadoop Distributed File System
Weil et al. — Ceph: A Scalable, High-Performance Distributed File System |
|
| 6 |
Thu 04/30 |
Coordination Services |
Required Reading:
DS4 §5.3.6
DDIA Ch. 6 §"Request Routing"
Graduate Reading:
Hunt et al. — ZooKeeper: Wait-free Coordination for Internet-scale Systems
Optional Readings:
Burrows — The Chubby Lock Service |
|
| 7 |
Tue 05/05 |
Global Distributed Databases |
Required Reading:
DDIA Ch 7. §"Snapshot Isolation and Repeatable Read" (p237-239) Ch. 8 §"Synchronized Clocks for Global Snapshots (p294)"
Graduate Reading:
Corbett et al. — Spanner: Google's Globally Distributed Database
Optional Readings:
Bacon et al. — Spanner: Becoming a SQL System
Kulkarni et al. — Logical Physical Clocks and Consistent Snapshots in Globally Distributed Databases |
|
| 7 |
Thu 05/07 |
Cluster Management and Orchestration |
Required Reading
DS4 §3.2.2-3.2.3
Graduate Reading:
Verma et al. — Large-scale Cluster Management at Google with Borg
Optional Readings:
Burns et al. — Borg, Omega, and Kubernetes
Hindman et al. — Mesos: A Platform for Fine-Grained Resource Sharing in the Data Center |
|
| 7 |
Fri 05/08 |
|
|
HW3 Due |
| 8 |
Tue 05/12 |
Distributed Computation Frameworks |
Required Reading DDIA Ch.10
Graduate Reading:
Dean & Ghemawat — MapReduce: Simplified Data Processing on Large Clusters
Optional Readings:
Zaharia et al. — Resilient Distributed Datasets: A Fault-Tolerant Abstraction for In-Memory Cluster Computing
Abadi et al. — TensorFlow: A System for Large-Scale Machine Learning |
HW 4 Out |
| 8 |
Thu 05/14 |
Stream Processing |
Required Reading:
DDIA Ch. 11
Graduate Reading:
Chandy & Lamport — Distributed Snapshots: Determining Global States of Distributed Systems
Optional Readings:
Akidau et al. — The Dataflow Model
Kreps et al. — Kafka: A Distributed Messaging System for Log Processing
Carbone et al. — Apache Flink: Stream and Batch Processing in a Single Engine |
|
| 9 |
Tue 05/19 |
Distributed AI Infrastructure |
Required and Graduate Reading:
Kwon et al. — Efficient Memory Management for Large Language Model Serving with PagedAttention
Optional Readings:
Dean et al. — Large Scale Distributed Deep Networks Shoeybi et al. — Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism |
|
| 9 |
Thu 05/21 |
Wrap-up & Review |
|
|
| 10 |
Fri 05/22 |
|
|
HW4 Due |
| Finals Week |
Fri 05/29 |
Final Exam |
|
|