English

If sufficient demand

This course will teach the doctoral students how to analyze and design programs for big data. It will provide knowledge on big data architectures, languages, and ecosystems with a focus on Spark. The techniques presented in the course are expected to have high impacts in a variety of fields such as data analysis, customer recommendation, trend prediction, pattern recognition, etc.

The course consists of four full-day sessions that will address:
1/ Cloud architectures, Spark concepts, and Spark programming.
2/ Intermediate and advanced Spark.
3/ Supervised machine-learning with Spark: MLlib and MLlib programming.
4/ Unsupervised machine learning.

show a knowledge of the architectures for big data processing.
show understanding of the structure of big data programming models and ecosystems

show her/his capability to operate big data architectures and design and write programs using Spark.

show the ability to select and assess architectures and algorithms for big data problems.

Lectures

Laboratory exercises

Exercises

Project

Written assignments

The assessment will consist of programs and reports to hand in

Failed, pass

Good programming skills in Java, Scala, or Python. Knowledge of statistics

Karau, A., Wendell, P. & Zaharia, M.: Learning Spark. O'Reilly Media, Inc, 2015. ISBN 9781449358624.
Karau, H., Konwinski, A., Wendell, P. & Zaharia, M.: Learning Spark. 2015. ISBN 9781449358624.

EDA025F

 -09-08

FN1/Anders Gustafsson