100 likes | 185 Vues
Learn how P.QL ensures performance in web applications and tackles complexity in querying ever-growing data sets, presenting a solution independent of scale with SQL-like syntax. Explore the state of big data analytics from Watson/IBM to Berkeley Data Analysis System (BDAS).
E N D
Guaranteed Performance While Querying Ever-Growing Data P QL Michael Armbrust BEARS Conference – February 2012
Force developers to use simple operations (get/put) • Makes complexity obvious
Force developers to use simple operations (get/put) • No optimization or data independence
PIQLSolution • Performance Insightful Query Language • SQL-like • Builds on existing scalable storage • Guaranteed performance independent of scale P QL
Problem with Cost Based Optimization Plan Choices Sequential Scan Random Lookups
BIG DATA: The State of the Art Watson/IBM Algorithms search Machines People
AMP: A Holistic Approach search Algorithms Watson/IBM Machines People
BDAS: Berkeley Data Analysis System A Top-to-Bottom Rethinking of the big data analytics stack integrating Algorithms, Machines, and People Data Collector Algo/Tools Data Analyst Infra. Builder Data Source Selector Result Control Center Visualization Analytics Libraries, Data Integration Poster Session @amplab 465 SODA Hall Higher Query Languages / Processing Frameworks Monitoring/Debugging Quality Control Resource Management Crowd Interface Storage Data Collector