Presto Newsletter – December 2018

Welcome to the Presto Newsletter, a monthly round up of news and events relevant to the Presto community.

Events

Presto – Fast SQL on Anything (Video & Slides)
Starburst CTO, Kamil Bajda-Pawlikowski presents the Presto architecture, how it works, best use cases, recent enhancements to the project, and the roadmap going forward.

• View the video & slides

Alluxio & Presto: Architecture for Fast SQL in the Cloud (Meetup slides)
Deploying Presto on the Alluxio open source distributed storage system for fast analytics in the cloud.

• View the slides

Presto News & Knowledge

HyperLogLog in Presto: A significantly faster way to handle cardinality estimation
Facebook describes the implementation of the Presto HyperLogLog data types for fast estimation of distinct values in large, high-cardinality data sets.

User Defined Function (UDF) Portability across Data Engines at LinkedIn
LinkedIn describes their implementation of a portable UDF API that works across Presto, Spark, and Hive.

5 ways Facebook improved compression at scale with Zstandard
Zstd improvements increase ORC file compression ratios and speed by double-digit percentages.

Presto Releases & New Features 

Starburst Presto 213e is ready for download. Security audit logging, new ANALYZE SQL command, plus many other new performance, security, and ease of use enhancements, .

• Download Presto 213e
• Read Presto 213e Release Notes

Engineers Corner 

Elasticsearch connector for Presto
The latest release of a new Presto – Elasticsearch connector for easier analysis and integration of data stored in Elasticsearch.