Image: ESA - C.Carreau (SEMPDN9OY2F)
Overview | Assignments | Policies

NETS 2120: Scalable and Cloud Computing (Fall 2023)

What is the "cloud"? How do we build software systems and components that scale to millions of users and petabytes of data, and are "always available"?

In the modern Internet, virtually all large Web services run atop multiple geographically distributed data centers: Google, Yahoo, Facebook, iTunes, Amazon, eBay, Bing, etc. Services must scale across thousands of machines, tolerate faults, and support thousands of concurrent requests. Increasingly, the major providers (including Amazon, Google, Microsoft, HP, and IBM) are looking at "hosting" third-party applications in their data centers - forming so-called "cloud computing" services. A significant number of these services also process "streaming" data: geocoding information from cell phones, tweets, streaming video, etc.

This course, aimed at a sophomore with exposure to basic programming within the context of a single machine, focuses on the issues and programming models related to such cloud and distributed data processing technologies: data partitioning, storage schemes, stream processing, and "mostly shared-nothing" parallel algorithms.

NETS2120 is a required course for the NETS program and a core requirement for the Data Science Minor. It also counts as a project elective for CSCI and ASCS, and as an Information Systems Elective for SSE.

Instructor

Andreas Haeberlen
Office hours: TBA (Levine 560)

Teaching assistants

Yuying Fan (Head TA) yuyingf@seas.upenn.edu OH: TBA
Arnav Chopra arnavc@wharton.upenn.edu OH: TBA
Ryoma Harris ryomah@seas.upenn.edu OH: TBA
Jordan Hochman jhawkman@seas.upenn.edu OH: TBA
Serena Huang serhuang@sas.upenn.edu OH: TBA
Scott Klein sklein24@wharton.upenn.edu OH: TBA
Kaily Liu liukaily@seas.upenn.edu OH: TBA
Dhatri Medarametla dhatrim@seas.upenn.edu OH: TBA
Rohan Moniz rmoniz@seas.upenn.edu OH: TBA
Rebecca Peng rpeng@seas.upenn.edu OH: TBA
Eric Wang etnw@seas.upenn.edu OH: TBA
Ellen Yan eyan2024@seas.upenn.edu OH: TBA
Jeremy Zein zeinje@seas.upenn.edu OH: TBA
Chi Zhang chizhg@seas.upenn.edu OH: TBA
Kunli Zhang kunliz@seas.upenn.edu OH: TBA

Format

The format will be two 1.5-hour lectures per week, plus assigned readings. There will be regular homework assignments, two midterms, and a term project. We will use an online forum for course-related discussions.

Time and location

Mondays/Wednesdays 10:15-11:45am (Location TBA)

Prerequisites

CIS 1200, Programming Languages and Techniques
CIS 1600, Mathematical Foundations of Computer Science

Textbooks

Spark: The Definitive Guide, by Bill Chambers and Matei Zaharia (O'Reilly)
ISBN 9781491912218; read online for free, or buy for approx. $54.

Data-Intensive Text Processing with MapReduce, by Jimmy Lin and Chris Dyer (Morgan & Claypool)
ISBN 978-1608453429; read online for free, or buy for approx. $40.

Additional materials will be provided as handouts or in the form of light technical papers.

Grading

Homework 30%, Term project 30%, Exams 35%, Participation/quizzes 5%

Policies

You can find a list of key course policies here.

Assignments

Homework assignments will be available for download; solutions should be submitted via GradeScope.

Tentative schedule

DateTopicDetailsReadingRemarks
Aug 30 Introduction Course introduction
"What is the Cloud, and why is it interesting?"
Data-centric computing
Course goals
Logistics
Policies
Overview of topics
Sep 6 The Cloud What is the Cloud?
Cloud hardware
Problems with classical scaling
Utility computing
Kinds of clouds
Virtualization
Cloud challenges
Armbrust: A view of cloud computing HW0 released
Sep 11 Concurrency Scalability and parallelization; Amdahl's law
Synchronization/concurrency/consistency
Mutual exclusing and locking
"NUMA, shared-nothing"
"Frontend/backend, sharding"
Vogels: Eventually consistent
TBALast day to add
Sep 13 The Internet The Internet; packet switching
Path properties; TCP
HW1 overview
MDN: A re-introduction to JavaScript HW0 due; HW1 released
Sep 18 Faults and Failures Fault models
Examples of non-crash faults
Replication; durability and availability
Primary-backup replication
Quorum replication
Network partitions; CAP theorem
Tseitlin: The antifragile organization
Sep 20 Cloud basics History of cloud computing
Interacting with the cloud
EC2 basics
EBS basics
Overview of some other AWS services
"Cloud computing features, issues, and challenges: a big picture" HW1MS1 due
Sep 25 Cloud storage Key-value stores
KVS and concurrency
KVS and the Cloud
Case study: S3
Case study: DynamoDB
Cooper et al.: PNUTS to Sherpa - Lessons from Yahoo!'s Cloud Database
Sep 27 Spark Introduction to scalable analytics
MapReduce
The Streams API
Apache Spark
Lambdas and serialization
"Spark textbook, Chapter 2 and 3" HW1MS2 due
Oct 2First midterm exam
TBALast day to drop
Oct 4 Programming in Spark Spark jobs
Working with files
Spark transformations
Spark actions
The Structured API
Distributed shared variables
"Spark textbook, Chapters 4-8" HW2 released
Oct 9 Understanding Spark Origins of Spark
The HDFS file system
Using HDFS
Apache Livy
Zaharia et al.: Cluster Computing with Working Sets
Oct 11 Graph algorithms Distributed graph algorithms
Distributed graphs
Graph algorithms in Spark
Single-source shortest path
K-Means clustering
Naive Bayes learning
"Lin & Dyer, Chapter 5" HW2MS1 due
Oct 12-15Fall Break
Oct 16 Random-walk algorithms Random-surfer model
Naive PageRank
Full PageRank
Adsorption / label propagation
Baluja et al.: Video Suggestion and Discovery for YouTube HW2MS2 due; HW3 released
Oct 18 Iterative processing Iterative processing
Bulk synchronous parallelism
Pregel and graph processing
Overview of deep neural nets
MXnet
Malewicz et al.: 'Pregel - A System for Large-Scale Graph Processing'
Oct 23 Web programming Web overview
HTML and CSS
Client/server model
The Domain Name System
"Cloudflare: HTTP/3: The past, the present, and the future"
Oct 25 Web programming (continued) HTTP and HTTPS
HTTP/2 and HTTP/3
Server design
HW3 due
TBALast day to designate course as pass/fail
Oct 30 Node.js Motivation: CGI and servlets
Node.js; basic operation
Hello world with Node
Accessing data
Cookies and sessions
"Node at LinkedIn: the pursuit of thinner, lighter, faster" HW4 and project handout released
Nov 1 Dynamic content Project overview
Project advice
The Document Object Model
XMLHttpRequest
React: Facebook's Functional Turn on Writing JavaScript
TBALast day to withdraw
Nov 6 AJAX AJAX overview
AJAX with jQuery
socket.io and async
Working with APIs
HW4MS1 due; team formation deadline; project begins
Nov 8 Web services and XML Web services
Data interchange; challenges
Data formats
XML
Working with XML
DTDs
XML Schema
XML DOM
First project check-in
Nov 13 Security Cryptography; RSA
Digital signatures
Attacks and Defenses (Part 1)
Attacks and Defenses (Part 2)
Current OWASP Top 10 HW4MS2 due
Nov 15 Databases Motivations for databases and data management [Video]
"Relational model, data streams" [Video]
SQL basics; declarative approach; query optimization [Video]
Transactions; ACID [Video]
F1: A Distributed SQL Database That Scales HW4MS3 due; second project check-in
Nov 20 Peer-to-peer Decentralization
Partly centralized systems; BitTorrent
Unstructured overlays; epidemic protocols
Structured overlays; consistent hashing; KBR
Case study: Pastry
Security challenges
Rodrigues and Druschel: P2P systems Third project check-in
Nov 22No class (Friday schedule)
Nov 27 Case study: Bitcoin Distributed ledgers
Bitcoin and Proof-of-Work
Bitcoin Script
Challenges in Bitcoin
Nakamoto: Bitcoin
Nov 29 Case study: Facebook Facebook's TAO
Scalability in TAO
Fault handing in TAO
Facebook's Haystack
Haystack design
Bronson et al.: TAO: Facebook's Distributed Data Store for the Social Graph Fourth project check-in
Dec 4 Special topics Differential privacy
Federated analytics
Differential Privacy: The Pursuit of Protections by Default
Dec 6 TBA TBA
Dec 11Second midterm exam
Dec 14-21Project demos; written reports due