CSE2520: Big Data Processing

General description

The term “Big Data” describes datasets that are either too big or change too fast or both to be processed on a single computer.

Big Data Processing provides an introduction to systems used to process Big Data. The main focus of the course is understanding the underpinnings of, programming and engineering big data systems; initially, the course explores general programming primitives that span across big data systems and touches upon distributed systems. Then, the course examines in detail the implementation of data analysis algorithms in Spark, in the context of batch processing applications, and Flink, in the context of streaming applications.

Learning objectives

After the end of the course, all students should be able to:

Explain the different dimensions of big data problems
Understand why classical algorithms fail on many big data problems
Understand, explain and apply data processing operations on the Unix shell
Understand, explain and apply data processing operations (filtering, folding, projecting etc)
Understand and explain basic techniques (vector clocks, consensus) in distributed systems
Understand and explain basic data management techniques in distributed databases
Understand and explain the major components of batch data processing frameworks
Create Spark-based algorithms for novel (unseen) practical problems
Describe in which scenaria streaming algorithms are most applicable
Apply basic streaming algorithms in practical problems, using Flink

Course Organization

5 ECTS: This means that you need to devote at least 140 hours of study for this course.
Online meetings: The course consists of 12 2-hour meetings. You are not required, but you are strongly encouraged, to attend.
Homework: In the homework assignments, you will have to write code or reply to open questions. You will always work in pairs.
Groups: The students are responsible to form pairs and communicate them to the course TAs, by registering them to CPM.
Labs: 4 hours per week, designed to help you work together and get support from teaching assistants.
Teaching Assistants: Teaching assistants will be available during lab hours to provide your with feedback on your assignments. Do be active in asking questions, but don’t expect them to provide you with solutions.

Week	Date	Topic	Teacher	Assignment (Deadline)
1	2/9	Course introduction, Big and Fast data, Intro to course PLs	GG
1	4/9	The Unix programming environment, Diomidis’s slides	DS	Unix (16/09)
2	9/9	Programming for Big Data (1)	GG
2	11/9	Programming for Big Data (2)	GG	Functional programming (30/09)
3	16/9	Distributed Systems	GG
3	18/9	Distributed Databases	GG
4	23/9	Distributed Filesystems	GG	~~Distributed Systems (07/10)~~
4	25/9	~~Distributed Systems 4~~ Lecture cancelled		MongoDB (Optional)
5	30/9	Spark: RDDs and Pair RDDs	GG
5	2/10	Spark Internals	GG
6	7/10	Spark SQL, Spark use cases: Synonyms with Word2Vec, Recommending bands, Predicting pull request merges	GG	Spark (21/10)
6	9/10	Graph Processing	GG
7	14/10	Stream processing	GG
7	16/10	Stream processing systems	GG	Flink (01/11)
8	21/10	Data engineering on the cloud	GG
8	23/10	No lecture	GG
9	28/10	Recap, Answers to recap questions (Quintin van Leersum and Mikhail Epifanov)

Teachers

The lab work is supervised by Thomas Overklift.

Online Resources

Portions of this course have been converted to online educational material by other TU Delft teachers. Please take a look at the following EdX MOOCs / ProfEds:

Unix Tools: Data, Software and Production Engineering, by Diomidis Spinellis
Introduction to Functional Programming for Big Data Processing, by Jan Rellermeyer
Taming Big Data Streams: Real-time Data Processing at Scale, by Asterios Katsifodimos

Use them at your discretion to improve your skills.

(TU Delft only): You can find the Collegerama recordings from 2019 here. Please note that the course contents have sligthly changed this year, so do not base your exam studying on the old lectures.

Assignments

You can find the course assignments on Brightspace and linked through this page. There will be 4 assignments instead of 5 due to circumstances; the assignment about distributed systems has been dropped.

All assignments are mandatory.

For submission, we will use CPM. The course name is CSE2520: Big Data Processing

You need to signup to enroll and also declare your pairs
To submit, hit the overview button and select the appropriate assignment
All the assignments have deadlines
Feedback and grading is automatic: the results are available on CPM.
Technical support: ask the Mattermost channel
- If no feedback after 1 hour: DO ASK THE TAs.

The student groups must submit each assignment before 23:59 on the day of the deadline.

Late submission: All submissions must be handed in time, with no exceptions. Any late submission will be discarded and will be graded with 0. In case of provable sickness, please contact the course teacher to arrange a case-specific deadline.

Assessment

Lab assignments (40%): Grade calculated as mean grade for all assignments. There is no minimum grade per individual assignment. If you don’t submit an assignment, or the submission is late, you will get a 0. Each assignment counts for ~~20%~~ 25% of the lab part. The final lab grade has a minimum of 5.
Written Exam (60%): Closed-book exam, multiple choice. Minimum grade: 5

Example exam material

Resit policy

There will be an exam-only resit during Q2/3. You are allowed to transfer your assignment grade to the resit as a whole. This means that you will not be able to re-submit individual assignments. Effectively, you can only resit your written exam.

Course resources

Online lab sessions every Wednesday afternoon using Queue.
- TAs will accepting requests with Jitsi video calls.
- Ask questions about the next assignment
- Ask questions about your grades
You are welcome to join the BDP 2020-2021 Mattermost team
- General questions can be asked here. Questions specific to your solutions should be asked during the lab
- TAs won’t always be active; don’t expect answers outside of lab hours
- You are recommended to help each other, but refrain from sharing solutions (this is considered fraud)
- Be nice :)

The course, by design, touches upon various current technologies; as such, there is no single source of truth. The following is an indicative list of resources where more information can be found. If you were to buy a single book about this course, I would recommend [1].

Bibliography

[1]

M. Kleppmann, Designing data-intensive applications. O’Reilly Media, Inc., 2017.

[2]

J. Laskowski, “Mastering apache spark 2,” 2017. [Online]. Available: https://www.gitbook.com/book/jaceklaskowski/mastering-apache-spark/details.

[3]

S. Ryza, U. Laserson, S. Owen, and J. Wills, Advanced analytics with spark: Patterns for learning from data at scale. O’Reilly Media, Inc., 2015.

[4]

H. Karau, A. Konwinski, P. Wendell, and M. Zaharia, Learning spark: Lightning-fast big data analysis. O’Reilly Media, Inc., 2015.

[5]

H. Karau and R. Warren, High performance spark. O’Reilly Media, Inc., 2017.

[6]

B. Chambers and M. Zaharia, Spark: The definitive guide. O’Reilly Media, Inc., 2017.

[7]

T. Akidau, S. Chernyak, and R. Lax, Streaming systems: The what, where, when, and how of large-scale data processing. O’Reilly, 2018.

[8]

C. Martella, R. Shaposhnik, D. Logothetis, and S. Harenberg, Practical graph analytics with apache Giraph. Springer, 2015.

[9]

I. Robinson, J. Webber, and E. Eifrem, Graph databases: New opportunities for connected data. Springer, 2015.