Shark: SQL and Rich Analytics at Scale

TL;DR

Shark is a scalable data analysis system that combines SQL query processing with complex analytics, achieving significant speedups over Hadoop and Hive while maintaining fault tolerance and flexibility.

Contribution

It introduces a unified engine with a novel distributed memory abstraction enabling fast SQL and analytics at scale, combining MPP performance with MapReduce-like fault tolerance.

Findings

SQL queries up to 100x faster than Hive

Machine learning programs up to 100x faster than Hadoop

Achieves speedups similar to MPP databases with fault tolerance

Abstract

Shark is a new data analysis system that marries query processing with complex analytics on large clusters. It leverages a novel distributed memory abstraction to provide a unified engine that can run SQL queries and sophisticated analytics functions (e.g., iterative machine learning) at scale, and efficiently recovers from failures mid-query. This allows Shark to run SQL queries up to 100x faster than Apache Hive, and machine learning programs up to 100x faster than Hadoop. Unlike previous systems, Shark shows that it is possible to achieve these speedups while retaining a MapReduce-like execution engine, and the fine-grained fault tolerance properties that such engines provide. It extends such an engine in several ways, including column-oriented in-memory storage and dynamic mid-query replanning, to effectively execute SQL. The result is a system that matches the speedups reported for MPP analytic databases over MapReduce, while offering fault tolerance properties and complex analytics capabilities that they lack.