# CS595 - DISC - Spark **Lecturer**: [Boris Glavic](http://www.cs.iit.edu/~glavic/) **Semester**: fall 2021

# 3. DISC ## DISC - Spark

## Overview - Like MapReduce, Spark is a DISC system - Computations are done in memory (when possible) - Different fault tolerance mechanism - less I/O intensive - More higher-order functions (not just map & reduce) - Written in Scala (Object-oriented + functional JVM language)

## Resilient Distributed Datasets (RDDs)

## Lineage-based Fault Tolerance -

## Spark's API and SparkSQL - for some more examples see: [https://github.com/IITDBGroup/CS595-repository/blob/master/spark.org](here)