This is what Dianhui (a.k.a Dennis) reported.
Dianhui (Dennis) Zhu presented his work on “Genomic Data Analysis with Hadoop”. Some of the Hadoopers who didn’t attend his first presentation were very interested in his work and requested this 2nd presentation.
The presentation was very successful. It started with a short description of the challenges that the researchers are facing in the field of genomic data analysis. It was then followed by a use case study on the use of Hadoop framework to detect genomic sequence patterns.
Dianhui actually walked Houston Hadoopers through the code and showed how to test it.. In the last 20 minutes, Dianhui demonstrated how to setup a 4-node Hadoop cluster from scratch and then did a live demo on the cluster. A performance comparison was conducted based on a real dataset. The attendants saw that the speed on the 4-node cluster was roughly three times faster than on a single node.
The meetup was held for the first time at the SHMsoft's new office location. Hadoopers were energized by pizza (thanks to the hospitality of SHMsoft). The presentation was accompanied by a lively interactive session with excellent questions and answers. It was finished with a perfect live demo.
See y'all there next time!
Art: Vassily Kandinsky - Untitled First Abstract Watercolor