Move Less, Move Faster: Speeding Up Citus Cluster Scaling | POSETTE: An Event for Postgres 2026
New Citus version adds parallel rebalancing to speed up cluster scaling when adding worker nodes.
“The data was not a bottleneck. The machinery that moves the data was the bottleneck.”
A Citus engineering talk explains two new enhancements ('move less, move faster') that parallelize the rebalancer when scaling out a Postgres/Citus cluster, motivated by a customer whose 18-20TB rebalance took 3-4 days due to serial shard and reference-table moves. It's a useful database engineering insight but has minimal AI-industry signal.