Geek Night is a regular event to promote the sharing of technical knowledge and increase collaboration among the geeks in the National Capital Region (Delhi). It is organized by a passionate group of programmers and sponsored by Thoughtworks.
We love feedback! If you have any suggestions or cribs, feel free to fill out our feedback form. Don't worry, your feedback will remain completely anonymous.Geek Night Volunteers
Real-time API delivering Data @ Scale
You're pretty familiar with real-time api's delivering data in matter of seconds. But what happens when each request needs to process data in GB and then return data in hundreds of MBs. Join us as we walk you through such an architecture and the key design decisions that went into building this real-time API. We examine the hybrid approach that combined batch processing on Hadoop with real-time processing on the Java application backed by relational store.
By: Akash Mishra
A date with Olga: Generating sales/lead opportunities using the power of Hadoop
We talk about our experience of building our first full-fledged production system on the Hadoop platform. We examine the business problem and why Hadoop turned out to be the ideal platform for architecting this solution. We'll look at some of the implementation concerns that ought to be factored, and how they can impact the overall solution.
By: Rahul Joshi & Vaibhav Khunger
Let the Sparks fly
Spark is a general-purpose, powerful engine processing large datasets. Applications built over Spark run 10-100x faster than MapReduce. Spark is compatible with HDFS, HBase, Cassandra, or any Hadoop data source. It also supports complex analytics through stream processing, machine learning and graphing algorithms. In this talk, you'll get a quick introduction of the Spark ecosystem. We'll also look some of the advantages of the Spark platform, and how easy it is to build applications and models using Spark APIs.
By: Varun Nidhi
ML riding high on Spark
We look at the different machine-learning algorithms and their practical applications. We see how Hadoop/Spark platform has provided a lot more opportunities for us to experiment with these algorithms, and how research can be brought into mainstream production use.
By: Gautam Punhani
Networking & Dinner
(served at our premises)