Elasticsearch 1.x to 8.x Migration for Logistics

The challenge

The client’s logistics platform — moving more than a million shipments a year — was still running Elasticsearch 1.x, a version old enough to create real scalability, cost, and performance problems of its own.

More than 7TB of data sat in that cluster, and the business-critical nature of the operation meant the upgrade path to 8.x and Elastic Cloud had to protect uptime, not just modernize the version number.

What we delivered

SquareShift deployed Elastic Cloud and ran the version jump as a phased, parallel migration rather than a single cutover, moving data and traffic across in stages instead of all at once.

Frozen-tier storage took over for cold, rarely-accessed data, cutting storage cost without touching what the team could still query day to day. Kibana dashboards were rebuilt on the new stack, and the client’s own team was trained end-to-end on operating the Elastic Cloud environment.

The payoff

The full 1.x-to-8.x migration, across a 7TB+ cluster, cost the client just 3 hours of downtime — not the days a big-bang cutover of this size would normally risk.

Frozen-tier storage cut storage cost by 60%, and the client’s team now runs and monitors the platform themselves, on dashboards SquareShift rebuilt in Kibana.