Title: OceanXtremes: Oceanographic Data-Intensive Anomaly Detection and Analysis Portal
Presenting Author: Thomas Huang
Organization: Jet Propulsion Laboratory
Co-Author(s): Brian Wilson, Edward Armstrong, Toshio Chin, George Chang, Nga Quach, Kevin Gill, Frank Greguska

Abstract:
OceanXtremes is a Cloud-based anomaly detection platform and web portal. It is being developed at the NASA Jet Propulsion Laboratory. It is currently in its first year of development. As a big oceanographic data analytic solution, OceanXtremes is taking a brand new approach in analyzing large collection of science observations using on-premise and commercial Cloud solutions. The goal of OceanXtremes is to deliver a rapid data analytic environment for climate research, which high performance geospatial search and data management solutions, distributed and multi-core algorithm execution framework, and scale-out, memory-based analysis platform. This presentation covers the current architecture of OceanXtremes, initial anomaly outputs, and portal design.