Apache Celeborn

Remote shuffle data service

Visit Apache Celeborn official site

What it does

Celeborn moves shuffle data into a separate service, providing a remote data-exchange layer that distributed processing engines can use during execution.

When to use it

In environments with changing compute nodes, examine Spark and Flink integrations to understand how shuffle storage is separated from execution.

Apache AGE

Graph queries within PostgreSQL

  • development
  • documentation

Research