Alluxio Data Orchestration Platform

In November Alluxio launched version 2.7 of the Alluxio Data Orchestration Platform, the San Mateo, Calif.-based company’s software for managing large-scale distributed data workloads and connecting data sources scattered across hybrid and multi-cloud IT systems.

Alluxio’s software is a virtual distributed file system that separates compute from storage and makes all data appear local no matter where it’s stored. That unifies access to distributed data, providing a way to link business analytics and data-driven applications to distributed data sources and making management of distributed data more efficient.

The new 2.7 release provides a five-fold improvement in I/O performance efficiency for machine learning training by parallelizing data loading, data pre-processing and training pipelines. The new edition also offers enhanced performance insight and support for open table formats like Apache Hudi and Iceberg, allowing the system to scale up access to data lakes for faster Presto- and Spark-based analytics.