Presented By O'Reilly and Cloudera
Make Data Work
5–7 May, 2015 • London, UK

Apache Spark: What's new; what's coming

Patrick Wendell (Databricks)
11:45–12:25 Wednesday, 6/05/2015
Hadoop & Beyond
Location: Buckingham Room - Palace Suite
Average rating: ****.
(4.73, 15 ratings)
Slides:   1-PDF 

Prerequisite Knowledge

This talk assumes only basic knowledge of Apache Spark. Users with more advanced understanding will get more from this talk, but everyone is welcome to attend.


The last year has seen significant growth in the Spark community, with several major releases (Spark 1.0, 1.1, and 1.2), new standard libraries (Spark MLlib and Spark SQL), and an ecosystem of community projects based on Spark.

This talk will provide an overview of Apache Spark and its current feature set, adoption, and use cases. It will then cover recent feature additions to Apache Spark such as elastic scaling support, new algorithms in MLlib, and the Spark SQL datasources API. It will also outline the Spark roadmap for upcoming months. Since this talk is not until May, the specific roadmap details will likely be determined close to the talk itself.

This talk is being submitted by Patrick Wendell, release manager of Spark 1.0, 1.1, and 1.2.

Photo of Patrick Wendell

Patrick Wendell


Patrick Wendell is a cofounder of Databricks and committer and PMC member of Apache Spark. He is the release manager of Spark’s 1.0, 1.1, and 1.2 releases. Before helping start Databricks, Patrick was a Ph.D student working in the U.C. Berkeley AMPLab, focusing on large scale data-intensive computing and advised by Ion Stoica.

Comments on this page are now closed.


Picture of Fouad Bendris
Fouad Bendris
10/05/2015 19:28 BST

Nice insights around Spark ! very smart prez & useful tricks ;()