We are looking for a very strong Hadoop Administrator to join us on a proof of concept initiative where a small group will be setting up a Hadoop environment from scratch.
While the project is research oriented, its fruits will be applied to a mission-critical system. Engineers will enjoy the fun, experimental, and fast-paced nature of a proof of concept program, with the satisfaction of building technology that will be utilized. This position is for someone who combines serious technical skills, a thirst for challenge, and enough humility to be self-reflective about their technical decisions.
True, solid, hands-on Hadoop administrator experience! (We love architects, systems engineers, systems administrators, and developers with Hadoop experience - and we would be happy to talk with them about other programs we work on - but this role requires actual hands-on Hadoop administration experience.)
Firm knowledge of HDFS and MapReduce
Experience setting up clusters, preferably from the ground up, preferred
Experience maintaining and monitoring clusters
Ability to create patches, script, and do some software development
A deep understanding of Hadoop design principles and factors that affect distributed system performance, including hardware and network considerations/configurations
Ability to load data into the cluster from dynamically-generated files using Flume and from relational database management systems using Sqoop, preferred
A firm handle on what system administration issues exist within the Hadoop ecosystem and associated components such as Hive, Pig, and HBase