The atmospheric science (AS) community generates model and observational data to simulate and monitor the Earth system. Big data in the AS community has arrived: high volumes (petabytes), at increasing velocity (to AS groups worldwide) and variety (of data formats and resolutions), are need for the veracity of models and observation systems that add value to the policy-making process. As scientists require solutions that allow interaction with these big data, the community is interested in the Map Reduce paradigm and Apache Spark. This talk presents a specific NASA Advanced Information Systems Technology (AIST) project called “SciSpark” that marries Apache Spark with climate science. SciSpark is a scalable system for interactive AS analysis. We will demonstrate SciSpark’s scientific data ingestion, visual interaction and metrics generation using the Spark engine.
