Analyzing Data at Scale with the Berkeley Data Analytics Stack

• Activity Number: 271
• Type: Invited
• Date/Time: Tuesday, August 5, 2014 : 8:30 AM to 10:20 AM
• Sponsor: IMS
• Author(s): Michael Franklin*+
• Companies: University of California, Berkeley
• Keywords: Data Science ; Big Data ; Data Analytics ; Software ; Healthcare

Abstract

Advances in data collection, processing and analysis are transforming organizations and enterprises of all types. The opportunities to exploit Big Data in Healthcare and Medicine are particularly vast. At the Berkeley Algorithms, Machines and People Laboratory (AMPLab) a unique collaboration of researchers in data management, computing systems, and machine learning, a building a new, unified software stack for large-scale data analytics. This stack called the Berkeley Data Analytics Stack (BDAS), combines several modalities of data analytics including SQL queries, graph processing, approximate query answering and declarative machine learning. BDAS enables analysts, scientists, and researchers to build data science pipelines combining these various modalities. In this presentation, I will first survey the rapidly changing landscape of large-scale data analytics. I will then present an overview of the BDAS system, and describe its applicability to medicine and health care, including cancer genomics, clinical data analysis, patient outcome, and healthy living applications.