In this work, we will explore Apache Hadoop, Yarn and Spark installed on a Raspberry Pi Cluster and the challenges that come with it. We will dive into the Spark Machine Learning Library to make use of distributed in memory computations and detect the centres of crime in Washington DC with the help of K-Means, see which products are often bought together on Amazon with the help of Frequent Pattern Growth and classify Pictures with Support Vector Machines.
Conception and analysis of a Raspberry Cluster with Apache Spark
Universität Ulm Universität Ulm
MA Abschlussvortrag, Nicolas Kuhaupt, Ort: O27/545, Datum: 29.03.2017, Zeit: 10:30 Uhr