This document provides an introduction to Apache Spark presented by Vincent Poncet of IBM. It discusses how Spark is a fast, general-purpose cluster computing system for large-scale data processing. It is faster than MapReduce, supports a wide range of workloads, and is easier to use with APIs in Scala, Python, and Java. The document also provides an overview of Spark's execution model and its core API called resilient distributed datasets (RDDs).