Follow Datanami:
February 20, 2020

Alluxio Looks to Bring Data Closer to Presto Engine

Alluxio, the cloud data orchestration specialist, is the latest vendor to join the Presto Foundation, which is shepherding the popular open source distributed SQL query engine.

Presto was launched by Facebook (NASDAQ: FB) engineers in 2012 as a follow-on to Apache Hive. The SQL query engine is touted as running interactive queries against data sources up to the petabyte range. The resulting analytics run at speeds comparable to commercial data warehouses.

Alluxio, San Mateo, Calif., was spun off from an open source project at the University of California at Berkeley’s AMPLab. The startup said its open source technology coupled with Presto would provide a new framework for cloud-based analytics.

The company’s software stack “brings data closer to Presto, dramatically simplifying hybrid cloud environments without creating any additional copies of data,” said Alluxio CEO Steven Mih.

Besides Facebook, Alluxio joins Presto Foundation founding members Alibaba (NYSE: BABA), Twitter (NYSE: TWTR) and Uber (NYSE: UBER).  The project joined the Linux Foundation last September.

Alluxio’s framework tackles data engineering hurdles common to interactive SQL workloads, including latency associated with remote data and object storage. Along with network latency and the inability of public cloud storage platforms to adequately handle analytics workloads, the company said its version of Presto separates the computing and storage stacks. The approach is designed to accelerate interactive analytics across files and object stores.

The framework makes high-priority data local, eliminating the need to generate and manage copies while connecting to cloud storage to query data regardless of storage location.

The startup released the latest version of its platform last summer in hopes of helping customers cope with data silos. A data management feature lands data in the appropriate system according to age: hot data may reside in RAM, warm data may in SSDs or hard disk drives, the coldest data in less expensive repositories offered by like Amazon Web Services S3 or Google Cloud.

“With a system like this you can easily orchestrate your data and move the useful data to the right place,” Haoyuan “H.Y.” Li, Alluxio’s founder and CTO, told Datanami last summer.

Alluxio said this week it will sponsor and exhibit at the inaugural PrestoCon on March 24.

Recent items:

Alluxio Bolsters Data Orchestration for Hybrid Cloud World

Presto Backers Bolster its Open Source Origins