Sean Owen introduces Spark, Scala and random decision forests, and demonstrates the process of analyzing a real-world data set with them.
John Davies shows a Spring work-flow consuming 7.4kB XML messages, binding them to 25kB Java but storing them in just 450 bytes each, 10 million derivative contracts in-memory on a laptop.
Lin Qiao discusses the architecture of Gobblin, LinkedIn’s framework for addressing the need of high quality and high velocity data ingestion.
Eugene Mandel discusses challenges of conforming data sources and compares processing stacks: Hadoop+Redshift vs Spark, showing how the technology drives the way the problem is modeled.
Sharan Kalwani presents the history of HPC and the technologies and trends which have contributed to creating the world of big data, covering applications of HPC resulting in big data technologies.
John Leach explains using HBase co-processors to support a full ANSI SQL RDBMS without modifying the core HBase source, showing how Hadoop/HBase can replace traditional RDBMS solutions.
Roman Shaposhnik discusses more advanced features of HDFS, in addition to how YARN has enabled businesses to massively scale their systems beyond what was previously possible.
Yann Yu discusses how Solr and Hadoop complement each other, and how to use Solr as a real-time, analytical, full-text search front-end to data stored in Hadoop.
Mike Keane presents how Conversant migrated to Flume, managing 1000 agents across 4 data centers, processing over 50B log lines per day with peak hourly averages of over 1.5 million log lines/sec.
Jeremy Stieglitz discusses best practices for a data-centric security , compliance and data governance approach, with a particular focus on two customer use cases.
Dean Wampler argues that Spark/Scala is a better data processing engine than MapReduce/Java because tools inspired by mathematics, such as FP, are ideal tools for working with data.
Matei Zaharia talks about the latest developments in Spark and shows examples of how it can combine processing algorithms to build rich data pipelines in just a few lines of code.