Justin Kestelyn is Director of Developer Community Outreach for Cloudera. Previously, he spent nine years at Oracle, building its developer and technical end-user communities. You can follow him on Twitter at @kestelyn.
Some people like to read books or enroll in classes to enrich their knowledge. Personally, I prefer a good conference. So fortunately for me, the Intel Developer Forum is just around the corner.
For someone like me, a Cloudera employee interested in the Apache Hadoop ecosystem, this conference is an alluring one. Intel, of course, will be a primary influencer on the Total Cost of Ownership of commodity Hadoop clusters over time - based on approaches to hardware optimization and energy efficiency, just for starters. The tools it gives hardware developers for turning those dials are important, and it is in a unique position to help Apache Hadoop cluster admins count on as much performance as possible.
For those reasons, there are a number of tasty technical sessions on the menu. Here are the ones that got my attention (times, dates, and rooms subject to change), listed in chrono order, and why:
"The Future of Energy Efficient Computing"
Weds. 10:30am, Room 2004
The surge in data volumes, and the growing prevalence of highly scalable approaches such as Apache Hadoop to processing it, will make energy-efficient computing the next frontier for infrastructure designers. Those clusters are getting more and more hungry for power.
"Big Data and Analytics in Healthcare and Life Sciences"
Weds. 12:45pm, Room 2009
There is no target-richer environment for big data analytics, and hence for Apache Hadoop, than Life Sciences. The Human Genome alone offers enough application opportunities for a lifetime.
"Optimal Systems for Big Data Performance and Scalability"
Weds. 2PM, Room 2002
With choices come trade-offs. What is the most efficient approach to squeezing maximum performance out of a cluster? Where should Apache Hadoop end, and the relational database engine begin (and vice versa)? These are questions that every implementor needs answered.
"Big Data Meets High Performance Computing"
Weds. 3:30pm, Room 2002
The convergence between Big Data, HPC, and Apache Hadoop is growing; managers of large scientific workloads with complex job scheduling requirements are taking a close look for new applications.
"Big Data Analytics on a Performance-optimized Hadoop Infrastructure"
Thurs. 2:30pm, Room 2008
Presented by Intel's open source strategist Vin Sharma. I'm interested to learn more about how Intel is collaborating with the Apache Hadoop community on benchmarks, tools, and utilities.
For me, these sessions alone are worth the price of admission. But even if you have a different agenda, make a point to stop by the Cloudera exhibit and say hello.