Continuing the recent jam-packed content for the Bay Area Scala Enthusiast meetings, this month we once again bring you a double-headliner. Mark Hamstra will talk about Spark: The Scala-centric part is the Spark cluster computing system that I am working with (http://www.spark-project.org/). The actual data mining algorithm that I've re-implemented in Scala/Spark is Classification based on Predictive Association Rules (CPAR -- http://www.siam.org/proceedings/datamining/2003/dm03_40YinX.pdf). Spark is in many ways similar to MapReduce/Hadoop, and leverages much of the same infrastructure. The talk will cover the entire Hadoop ecosystem from a high level, how Spark fits in, and what the significant differences are -- particularly the Mesos cluster manager, Spark's distributed data sets and their support for iterative algorithms, and more. And then James Earl Douglas will talk about a declarative UI framework he has put together using continuations and functional reactive programming. It's a neat way to hide the noise of Java's ActionListener API and to keep UI behavior cleanly defined in one place. Since Dick will be in Belgium for Devoxx, Dave Briccetti has kindly volunteered to step in and run the meeting. Thanks Dave. We will also have lightning talks and pizza. The details: Monday 11/14/2011 7pm-9pm in the Unite Conference Room 2025 Stierlin Ct Mountain View, CA 94043 See you there for a fun evening. Please try to arrive a little before 7pm to ensure you can get in easily. |
|
|
BASE Meeting #43 - Mon Nov 14th 7pm at LinkedIn, Mountain Viewposted Nov 8, 2011 12:30 PM by Domain Admin |