Work on the big data integration group, try to integrate spark framework to company’s platform, and analyze the performance of a developing product by reproducing Amplab’s benchmark (including Hive, Impala, Spark, Cascading) in local hadoop cluster.