Data-driven businesses that use data analysis to direct their decisions, are collecting data at ever-increasing rates. At the heart of these businesses is the database used to store and analyze that data. As the size and speed of the database comes under stress, the quality and timeliness of analysis begins to suffer because compromises are made to accommodate the flood of information. For many applications, the bottleneck can be the database’s bottommost layer, the storage engine. As the database grows, as insertion rates increase, and as range queries become more complex, many storage engines are not keeping up with users’ demands for timely analysis on fresh data.
MySQL’s pluggable storage engine architecture affords a choice among storage engines that are suited to a variety of tasks. This session will explore the performance issues that are encountered by storage engines as they insert, organize, and query data in large data warehouse applications, and the limitations of various techniques.
A revolutionary new storage engine technology, fractal trees, will be discussed. Fractal trees overcome the performance challenges of data warehouse applications, and do so without sacrificing the benefits of a generalized MySQL database. Because fractal trees dramatically speed the response times of both data insertions and range queries, data warehouse applications are no longer limited by storage engine performance, and businesses can once again do detailed analysis on fresh data.
Prof. Farach-Colton is an expert in algorithmics and information retrieval. He was an early employee at Google and is a Professor of Computer Science at Rutgers University. Martin Farach-Colton co-founded Tokutek in 2006 in order to commercialize his research on storage systems.
View a complete list of MySQL contacts.