Course: Big Data Analysis

Fall 2013

This schedule is tentative and subject to change

Make sure to check my.poly.edu for course announcements

News

• Assignment on Mapreduce and Pig, due on Dec 1st. Please see http://my.poly.edu

• Nov 7th: New quizzes have been assigned. Please see http://www.newgradiance.com/services/servlet/COTC

The deadline is Nov 15th. Please make sure that you have your correct name and Poly ID in your Gradiance account.

• Dr. C Mohan's presentation is now available at http://bit.ly/CMnMDS

For frequently asked questions about the course and homework assignments, please check our BigDataAnalysisFAQ.

Week 1: Monday Sept. 9th - Course Overview

• Course overview and introduction to Big Data Analysis
• Lecture notes: http://vgc.poly.edu/~juliana/courses/cs9223/Lectures/intro.pdf
• Student survey -- to be filled out today!

Required Reading

• Mining of Massive Datasets, Chapter 1
• Data-Intensive Text Processing with MapReduce, Chapter1

Additional References

• Dilbert's BigData
• New York Time's "How BigData Became so Big"
• World Economic Forum: Big Data, Big Impact
• The Analytics Journey
• BigData Analytics Usecases

Week 2: Monday Sept. 16th - Map-Reduce/Hadoop

• Introduction to Map-Reduce and high-level data processing languages
• Lecture notes: http://vgc.poly.edu/~juliana/courses/cs9223/Lectures/hadoop.pdf
• Hand out AWS tokens. Notes on using AWS.
• Apache Hadoop
• The Map-Reduce ecosystem: Pig, Hive, Mahout

Assignment

• cs9223 Mapreduce Assignment
• This is an individual assignment. You may not collude with any other individual, or plagiarise their work.

For more details see http://cis.poly.edu/policies.

• You assignment is due on Sun Sept 29th. Make sure you can login and access my.poly.edu!
• If you have questions about the assignment, we will hold office hours on Sept 23, 2013 from 2:30-3:30pm at 2 Metrotech, room 10.018

Required Reading

• Mining of Massive Datasets, Chapter 2
• Data-Intensive Text Processing with MapReduce, Chapter 2 and Chapter 3
• original google map-reduce paper

Week 3: Monday Sept. 23rd - Data Management for Big Data

• Databases and Big Data: Persistence, Querying, Indexing, Transactions
• Lecture notes: http://vgc.poly.edu/~juliana/courses/cs9223/Lectures/paralleldb-vs-hadoop.pdf

Related Topics

• BigTables and NoSQL stores. Tuple store vs. column stores: HBase, MongoDB, Cassandra
• HBase book HBase: The Definitive Guide. Random Access to Your Planet-Size Data: http://shop.oreilly.com/product/0636920014348.do
• HBase book. Chapter 8 Architecture for information about transactional processing, WriteAhead Log notably, and how consistency is being maintained.
• Transactions in NoSQL stores. Google's percolator, [1].
• "NewSQL" stores: more on Hive, VoltDB, HadoopDB,
• Beyond MapReduce: Berkeley's Spark, UC Irvine's Asterix, Google's Dremel

Required Reading

• PDMBS vs. MapReduce
• http://cacm.acm.org/magazines/2010/1/55744-mapreduce-a-flexible-data-processing-tool/fulltext
• Parallel data processing with MapReduce: a survey. Lee et al, SIGMOD Record 2011
• Benchmark DBMS vs MapReduce (2009)

Additional References

• http://www.computerworld.com/s/article/9224180/What_s_the_big_deal_about_Hadoop_
• Bigtable: A Distributed Storage System for Structured Data
• HadoopDB: An Architectural Hybrid of MapReduce and DBMS Technologies for Analytical Workloads
• Low Overhead Concurrency Control for Partitioned Main Memory Databases
• ASTERIX: Towards a Scalable, Semistructured Data Platform for Evolving-World Models.
• Dremel: Interactive Analysis of Web-Scale Datasets
• Large-scale Incremental Processing Using Distributed Transactions and Notifications

Week 4: Monday Sept 30th - Invited lecture by Dr. C. Mohan (IBM)

• Note that we will meet at a different location: NYU CUSP, 1 Metrotech Center, 19th floor

• Tutorial: An In-Depth Look at Modern Database Systems: http://bit.ly/CMnMDS

• Abstract: This tutorial is targeted at a broad set of database systems and applications people. It is intended to let the attendees better appreciate what is really behind the covers of many of the modern database systems (e.g., NoSQL and NewSQL systems), going beyond the hype associated with these open source and commercial systems. The capabilities and limitations of such systems will be addressed. Modern extensions to decades old relational DBMSs will also be described. Some application case studies will also be presented. An outline of problems for which no adequate solutions exist will be included. Such problems could be fertile grounds for new research work.

• Presenter: Dr. C. Mohan, IBM Fellow, IBM Almaden Research Center, San Jose, CA 95120, USA.

• Bio: Dr. C. Mohan has been an IBM researcher for 31 years in the information management area, impacting numerous IBM and non-IBM products, the research community and standards, especially with his invention of the ARIES family of locking and recovery algorithms, and the Presumed Abort commit protocol. This IBM, ACM and IEEE Fellow has also served as the IBM India Chief Scientist. In addition to receiving the ACM SIGMOD Innovation Award, the VLDB 10 Year Best Paper Award and numerous IBM awards, he has been elected to the US and Indian National Academies of Engineering, and has been named an IBM Master Inventor. This distinguished alumnus of IIT Madras received his PhD at the University of Texas at Austin. He is an inventor of 38 patents. He serves on the advisory board of IEEE Spectrum and on the IBM Software Group Architecture Board’s Council. More information can be found in his home page at http://bit.ly/CMohan

Week 5: Monday Oct. 7th - Query Processing on Mapreduce and High-level Languages

• Pig Latin and Query Processing:
  • Relational processing over MapReduce
  • Queries over MapReduce
• In-class assignment

Required Reading

• Pig Latin: A Not-So-Foreign Language for Data Processing

Additional References

• Jaql: A Scripting Language for Large Scale Semistructured Data Analysis
• Hive - A Warehousing Solution Over a Map-Reduce Framework

Week 6: Wed Oct. 16th - Fall Break - Make-up class

• Reproducibility and Data Exploration: http://vgc.poly.edu/~juliana/courses/cs9223/Lectures/reproducibility.pdf
• Large-scale information integration: http://vgc.poly.edu/~juliana/courses/cs9223/Lectures/web-information-integration.pdf

Week 7: Monday Oct. 21st - Invited Speaker: Alberto Lerner

• Inside MongoDB

Week 8: Monday Oct 28th- Statistics is easy - Invited Speaker: Dennis Shasha

• Guest lecture by Dennis Shasha: Statistics is Easy
• Introduction to Provenance

Required Reading

• http://www.morganclaypool.com/doi/abs/10.2200/S00142ED1V01Y200807MAS001 -- book is available for free for NYU students
• Second edition of the book: http://www.morganclaypool.com/doi/pdf/10.2200/S00295ED1V01Y201009MAS008

• We will cover the material planned for "Week 10: Monday Nov. 11th": Finding Similar Items

Week 9: Monday Nov. 4th - Finding Similar Items, Information Integration

• Similarity: Applications, Measures and Efficiency considerations
  • Lecture notes: http://vgc.poly.edu/~juliana/courses/cs9223/Lectures/similarity.pdf
• Similarity application: Information integration on the Web:
  • Lecture notes: http://vgc.poly.edu/~juliana/courses/cs9223/Lectures/web-info-integration.pdf
• Homework presentation and demo

Required Reading

• Mining of Massive Datasets, chapter 3; information integration; entity resolution

Homework Assignment

Due Nov 15th, 2013 Your assignment is in http://www.newgradiance.com/services. Please see http://vgc.poly.edu/~juliana/courses/cs9223 for instructions on how to access this service.

Week 10: Monday Nov. 11th - MapReduce Algorithm Design

• Lecture notes: http://vgc.poly.edu/~juliana/courses/cs9223/Lectures/mapreduce-indexing-graph.pdf

Required Reading

• Chapters 3 and 4 in textbook: Data-Intensive Text Processing with MapReduce, by Lin and Dyer

Homework Assignment

Due Nov 15th, 2013 Your assignment is in http://www.newgradiance.com/services. Please see http://vgc.poly.edu/~juliana/courses/cs9223 for instructions on how to access this service.

Week 11: Monday Nov 18th- MapReduce Algorithm Design and Graph Processing

• Lecture notes: http://vgc.poly.edu/~juliana/courses/cs9223/Lectures/mapreduce-indexing-graph.pdf

Homework Assignment

Your Mapreduce/Pig assignment is available from Blackboard. It is Due December 1st.

Required Reading

• Data-Intensive Text Processing with MapReduce, Chapter 4 (Inverted Indexing for Text Retrieval) and 5(Graph Algorithms)

Additional Reading

• 1998 PageRank Paper
• Mining of Massive Datasets, Chapter 5 (Link Analysis)
• Pregel: A System for Large-Scale Graph Processing. Google. [2]

Week 12: Monday Nov. 25th - Large-Scale Visualization

• Invited lectures by:
  • Dr. Lauro Lins (AT&T Research)
  • Dr. Huy Vo (NYU Center for Urban Science and Progress)

• Lecture notes:
  • https://www.dropbox.com/s/7t2vqryj5zgs44n/intro-to-visualization.pdf
  • https://www.dropbox.com/s/btb3ocupkmpgefi/nanocubes.pdf

Required Reading

The Value of Visualization, Jarke Van Wijk http://www.win.tue.nl/~vanwijk/vov.pdf

Tamara Munzner's Book draft 2 available online http://www.cs.ubc.ca/~tmm/courses/533/book/

Nanocubes Paper http://nanocubes.net http://nanocubes.net/assets/pdf/nanocubes_paper_preprint.pdf

Additional Reading

imMens Paper (to contrast with nanocubes) http://vis.stanford.edu/papers/immens

Week 13: Monday Dec. 2nd - Frequent Itemsets

• Lecture notes: http://vgc.poly.edu/~juliana/courses/cs9223/Lectures/association-rules.pdf

Additional Reading

• Mining association rules between sets of items in large databases. Agrawal et al., SIGMOD 1993. http://www.almaden.ibm.com/cs/quest/papers/sigmod93.pdf
• Fast algorithms for mining association rules. Agrawal and Srikant, VLDB 1994. https://www.seas.upenn.edu/~jstoy/cis650/papers/Apriori.pdf
• An effective hash-based algorithm for mining association rules. Park et al., SIGMOD 1995. http://www.dmi.unict.it/~apulvirenti/agd/PCY95.pdf

Optional Quiz

Due Dec 9th

Week 14: Monday Dec. 9th - - EM and exam review

• Lecture notes: http://vgc.poly.edu/~juliana/courses/cs9223/Lectures/hmm-em-mapreduce.pdf

Readings

Data-Intensive Text Processing with MapReduce, Chapter 6 (EM Algorithms for Text Processing)