COS518: Advanced Computer Systems (Spring 2018)

Schedule

Please check the following syllabus frequently as we will be updating assigned readings throughout the course. Signup for readings on Piazza.

Lectures and related materials are licensed for use under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License, attributed to Michael Freedman, Princeton University.

Student presentations should be 10-12 minutes and follow this format.

Date	Lecture topic	Reading topic	Readings (everyone)	Readings (sign-up)
Fundamentals
M 02/05	Introduction [pdf] [pptx]	Introduction
W 02/07	Naming and weak consistency [pdf] [pptx]	Naming	Hints for Computer System Design End-to-End Arguments in System Design Reading/presenting papers
F 02/09	Team selections due
Storage
M 02/12	Strong consistency [pdf] [pptx]	Weak consistency	Managing Update Conflicts in Bayou, a Weakly Connected Replicated Storage System	Dynamo: Amazon's Highly Available Key-value store PNUTS: Yahoo!'s Hosted Data Serving Platform TAO: Facebook’s Distributed Data Store for the Social Graph
W 02/14	Consensus [pdf] [pptx]	Strong consistency		Object Storage on CRAQ Don't Settle for Eventual: Scalable Causal Consistency for Wide-Area Storage with COPS Making Geo-Replicated Systems Fast as Possible, Consistent When Necessary
M 02/19	Transactions [pdf] [pptx]	Consensus	In Search of an Understandable Consensus Algorithm (Raft) Viewstamped Replication Revisited (optional)	There Is More Consensus in Egalitarian Parliaments Just say NO to Paxos Overhead: Replacing Consensus with Network Ordering
W 02/21	Transactions 2 [pdf] [pptx]	Transactions	Spanner: Google's Globally-Distributed Database	Extracting More Concurrency from Distributed Transactions (ROCOCO) Eris: Coordination-Free Consistent Transactions Using In-Network Concurrency Control
M 02/26	Project proposals due		Private piazza post
M 02/26	Column stores [pdf] [pptx]	K-V stores	LevelDB	SILT: A Memory-Efficient, High-Performance Key-Value Store Cache Craftiness for Fast Multicore Key-Value Storage (Masstree)
W 02/28	Flash [pdf] [pptx]	Column stores		Bigtable: A Distributed Storage System for Structured Data C-Store: A Column-oriented DBMS Replex: A Scalable, Highly Available Multi-Index Data Store
M 03/05	Caching [pdf] [pptx]	Flash		RIPQ: Advanced Photo Caching on Flash for Facebook CORFU: A Shared Log Design for Flash Clusters Write Once, Get 50% Free: Saving SSD Erase Costs Using WOM Codes
Big Data
W 03/07	Batch processing [pdf] [pptx]	Caching		Fast Crash Recovery in RAMCloud Scaling Memcache at Facebook
F 03/09	Finalize project proposal		Private piazza post
M 03/12	Stream processing [pdf] [pptx]	Batch processing	MapReduce : Simplified Data Processing on Large Clusters	Resilient Distriuted Datasets: A Fault-Tolerant Abstraction for In-Memory Cluster Computing Improving MapReduce Performance in Heterogeneous Environments
W 03/14	Graph processing [pdf] [pptx]	Stream processing	The world beyond batch: Streaming 101 The world beyond batch: Streaming 102	Discretized Streams: Fault-Tolerant Streaming Computation at Scale (Spark Streaming) MillWheel: Fault-Tolerant Stream Processing at Internet Scale
M 03/19	Spring Break
W 03/21	Spring Break
M 03/26	Machine learning [pdf] [pptx]	Graph processing	GraphLab: A New Framework For Parallel Machine Learning	Distributed GraphLab: A Framework for Machine Learning and Data Mining in the Cloud Pregel: A System for Large-Scale Graph Processing
W 03/28	Interim project presentations
M 04/02	Scheduling [pdf] [pptx]	Machine Learning	Machine Learning: The High-Interest Credit Card of Technical Debt	TensorFlow: A System for Large-Scale Machine Learning Scaling Distributed Machine Learning with the Parameter Server
Applications
W 04/04	Security and secure systems [pdf] [pptx]	Scheduling	Dominant Resource Fairness: Fair Allocation of Multiple Resource Types	Sparrow: Distributed, Low Latency Scheduling SLAQ: Quality-Driven Scheduling for Distributed Machine Learning
M 04/09	Distributed hash tables [pdf] [pptx]	Security and secure systems	Secure Untrusted Data Repository (SUNDR)	SPORC: Group Collaboration using Untrusted Cloud Resources CryptDB: Protecting Confidentiality with Encrypted Query Processing
W 04/11	Content-delivery systems (designs) [pdf] [pptx]	Distributed hash tables	Chord: A Scalable Peer-to-peer Lookup Service for Internet Applications A Scalable Content-Addressable Network (optional)	Kademlia: A Peer-to-peer Information System Based on the XOR Metric Beehive: O(1) Lookup Performance for Power-Law Query Distributions in Peer-to-Peer Overlays
M 04/16	Content delivery systems (experiences) [pdf] [pptx]	Content-delivery systems (designs)		The Effectiveness of Request Redirection on CDN Robustness BitTyrant: Do incentives build robustness in BitTorrent?
W 04/18	Coding in systems (guest lecturer Wyatt Lloyd) [pdf] [pptx]	Content delivery systems (experiences)		The Akamai Network: A Platform for High-Performance Internet Applications Experiences with CoralCDN: A Five-Year Operational View
M 04/23	Bitcoin and blockchains [pdf] [pptx]	Coding in systems	f4: Facebook’s Warm BLOB Storage System	Erasure Coding in Windows Azure Storage EC-Cache: Load-balanced, Low-Latency Cluster Caching with Online Erasure Coding
W 04/25	Ethereum and smart contracts (guest lecturer Arvind Narayanan) [pdf] [pptx]	Bitcoin and blockchains	Bitcoin: A Peer-to-Peer Electronic Cash System	Blockstack: A Global Naming and Storage System Secured by Blockchains Bitcoin-NG: A Scalable Blockchain Protocol
M 04/30	Anonymity systems [pdf] [pptx]	Ethereum and smart contracts	Ethereum: A Next-Generation Smart Contract and Decentralized Application Platform	Demystifying Incentives in the Consensus Computer On Scaling Decentralized Blockchains
W 05/02	TimescaleDB	Anonymity systems		Tor: The Second-Generation Onion Router Freenet: A Distributed Anonymous Information Storage and Retrieval System
M 05/14		Final project presentation (9-11:30am)
Tu 05/15		Final project report due (Dean's date)