NetBox Labs has launched Replica Cache in public preview for NetBox Cloud Premium tier customers. It is a read-only API that serves NetBox data from an operationally independent cache, delivering sub-50ms P95 query latency with zero load on the primary database. Designed for large-scale automation teams — AI datacenter operators, hyperscalers, telco teams — it handles high-frequency reads from Ansible inventories, Terraform plans, and monitoring integrations without competing with write operations. The cache replicates changes in near-real-time (typically under 10 seconds), supports server-side filtering, cursor-based pagination, and field selection. It complements the existing TurboBulk feature: TurboBulk handles bulk writes and consistent exports, while Replica Cache offloads high-frequency reads.
Table of contents
The read problem at scaleHow it works — a separate read pathThe platform performance storyHow to try it outSort: