Dremel: Interactive Analysis of Web-Scale Datasets


Dremel is a scalable, interactive ad-hoc query system for analysis of read-only nested data. By combining multi-level execution trees and columnar data layout, it is capable of running aggregation queries over trillion-row tables in seconds. The system scales to thousands of CPUs and petabytes of data, and has thousands of users at Google. I'll describe the architecture and implementation of Dremel, and explain how it complements MapReduce-based computing. I'll present a novel columnar storage representation for nested records and discuss experiments on few-thousand node instances of the system.


Sergey Melnik is a software engineer at Google, where he develops infrastructure for managing web-scale data. Before that he was a researcher in the Database Group at Microsoft Research, a visiting researcher at Stanford University, and served as an invited expert at the World-Wide Web Consortium. He got an M.S. and a Ph.D. in Computer Science from the University of Leipzig, Germany. He received a best dissertation award from the German C.S. Society and best paper awards at SIGMOD and ICDE conferences.