<!-- llms-explorer concept facts · https://llms-explorer.com/tree/analytics-node-tier-prorating/ · pack 2026-10-24 · ~915 tokens -->

# Analytics node tier prorating

> Source-anchored research for Analytics node tier prorating. Run status: COMPLETED.

Parent: [Atlas Cluster Topology and Tier Economics](https://llms-explorer.com/tree/atlas-cluster-topology-and-tier-economics/) · 4 facets · 10 facts · page: https://llms-explorer.com/tree/analytics-node-tier-prorating/

## Definitions

- Analytics nodes add to the bill; prorating is not a promise of lower total spend. Studio 3T identifies workload-isolated analytics nodes as optional costs, and MongoDB's announcement describes them as additional read-only nodes. (medium confidence) — [source](https://studio3t.com/knowledge-base/3t-build/desktop-ide/mongodb-atlas/)

## How it works

- A substantially lower analytics tier can cause replication lag and loss of oplog coverage. Node-level savings therefore require checking replication feasibility. (low confidence) (verified as of 2026-10-23) — [source](https://www.mongodb.com/docs/atlas/cluster-config/multi-cloud-distribution.md)
- Patrick McLain's September 28, 2022 incident account reports that secondaryPreferred with analytics tags fell back to the primary when analytics reads failed, degrading operations; changing to secondary restored isolation. Paying for an analytics tier alone did not deliver isolation in that incident. (low confidence) — [source](https://pmclain.com/mongodb/2022/09/28/fun-with-mongodb-atlas-analytics-nodes.html)

## Comparisons and alternatives

- PointFive's FinOps guidance identifies pricing the entire operational cluster for occasional analytics as waste and recommends separate analytics tiers; this echoes MongoDB's stated cost rationale, rather than independently measuring prorating. (medium confidence) — [source](https://www.pointfive.co/efficiency-hub/inefficiencies/overprovisioned-mongodb-atlas-cluster-tier)

## Facts and statements

- Atlas prices a selected analytics tier like the same base tier, then adjusts a differing analytics tier on a prorated per-node basis; the create/edit UI displays the price. (low confidence) (verified as of 2026-10-23) — [source](https://www.mongodb.com/docs/atlas/cluster-config/multi-cloud-distribution.md)
- Dedicated M10+ clusters can use an analytics tier above or below the operational tier, so analytics compute can change without resizing every operational node. (high confidence) (verified as of 2026-10-23) — [source](https://www.mongodb.com/docs/atlas/cluster-config/multi-cloud-distribution.md)
- MongoDB announced independent analytics tiers on August 3, 2022, updated August 23; previously analytics nodes used the same tier as other cluster nodes. (low confidence) — [source](https://www.mongodb.com/company/blog/product-release-announcements/introducing-ability-independently-scale-atlas-analytics-node-tiers)
- On Atlas Core, disk size and IOPS must match across node types; a lower analytics compute tier does not imply separately reduced storage or IOPS. (low confidence) (verified as of 2026-10-23) — [source](https://www.mongodb.com/docs/atlas/cluster-config/multi-cloud-distribution.md)
- Atlas Core Local NVMe SSD requires matching base and analytics tier levels, limiting the use of a cheaper analytics tier. (low confidence) (verified as of 2026-10-23) — [source](https://www.mongodb.com/docs/atlas/cluster-config/multi-cloud-distribution.md)
- Atlas Operator 2.13.1 release notes removed its same-instance-size restriction for analytics nodes. Operator users should distinguish an integration's validation restriction from Atlas's independently selectable analytics tier capability. (medium confidence) — [source](https://github.com/mongodb/mongodb-atlas-kubernetes/releases)