Join us for out March Meeting when Joe McTee will open with The Genealogy of Troy (Client-side Cassandra) followed by Danial Glauser presenting Machine Learning with Apache Mahout.
Sponsorship and Door Prize
We are also thankful for TekSystems' sponsorship's of the pizza and drinks.
We continue to meet at the Wolf Law Library:
Wolf Law Building, Room #207
2450 Kittredge Loop Road
Boulder, CO 80309
Directions can be found here.
6:00-7:00: The Genealogy of Troy (Client-side Cassandra)
Cassandra stands out amongst the big data products in its ability to handle optimized writes of large amounts of data while providing configurable fault tolerance and data integrity. Two popular libraries that allow the JVM developer to leverage these capabilities are Hector and the recently open sourced Astyanax. In this talk, Joe will present examples of storing time series data in a Cassandra data store using both of these libraries. There will be code! As an added bonus, a mechanism to unit test using an embedded Cassandra client will be presented.
Slides can be found here.
About Joe McTee
Joe is a Principal Engineer at Tendril, developing products that bring consumers, utilities, and consumer product manufacturers together in a partnership to save energy while maintaining quality lifestyle. In his 4 years at Tendril, he has worked on load control algorithms, smart outlets, smart thermostats, and is currently working on consumer focused energy reports that can be delivered both electronically and via paper. He is passionate about energy conservation. Ask him about solar energy if you have an hour! In his spare time, Joe is the current Boulder JUG coordinator.
7:00-7:30: Food, Soda and Networking
We are grateful to TekSystems for their continued sponsorship of the Food and Soda!
7:30-9:00: Machine Learning with Apache Mahout
Have you wondered why you were being asked a certain question when the system should already know the answer? Well, if it can't know the answer then at least it can make a good suggestion, right? As a programmer how would you write code to handle this? Could you find patterns in the data that are not obvious even with lots of domain knowledge? Machine Learning to the rescue. We'll explore how with a little math and a large data set you can quickly construct a recommendation engine. There are numerous algorithms to choose from and we will spend time reviewing their strengths and weaknesses. Data sets can quickly grow too large for a single system so we will also explore how things change when you distribute the work. All examples will be in Clojure using the Apache Mahout library.
Slides can be found here.
About Daniel Glauser
Daniel Glauser has spoken on for audiences in Denver, Boulder, Colorado Springs and Trivandrum, India. Daniel organizes the Denver Clojure Meetup where he focuses on working with strong members of the Clojure community to teach everything from beginning functional programming to advanced concurrency patterns. Daniel is a software architect with over thirteen years of experience working for companies like Comcast, NBC-Universal and BellSouth. Currently Daniel works as a software architect for VMware in Colorado Springs where he is working on large scale cloud management systems. Daniel's interests include functional programming, big data, distributed systems, logic systems, and enterprise architecture. In Colorado Daniel has spoken at DJUG, DOSUG, BJUG and CSOSUG.