<!-- llms-explorer concept facts · https://llms-explorer.com/tree/index-storage-cost/ · pack 2026-10-15 · ~857 tokens -->

# Index Storage Cost

> Depth-first rabbithole dossier for Index Storage Cost; source-anchored research pack.

Parent: [MongoDB Atlas Cost Optimization](https://llms-explorer.com/tree/mongodb-atlas-cost-optimization/) · 3 facets · 8 facts · page: https://llms-explorer.com/tree/index-storage-cost/

## Measurements and reference values

- [low] In one case study, the largest collection held about 300M documents using 130GB, with indexes accounting for another 30GB; after a data retention policy it fell to about 70M documents using less than 40GB plus 21GB of indexes. (verified as of 2026-10-15) Sources: https://dev.to/dlondero/saving-70-000-a-year-on-atlas-mongodb-534m — source: `~/.global-ai-hub/research/.pipeline/mongodb-2026-10-14/frb/index-storage-cost-267559520b/rabbithole-synthesis.md#core`

## Problems, failure modes and limitations

- [low] Unused indexes in MongoDB consume storage and slow down writes without delivering value. Sources: https://www.mongodb.com/docs/atlas/architecture/current/cost-saving-config.md — source: `~/.global-ai-hub/research/.pipeline/mongodb-2026-10-14/frb/index-storage-cost-267559520b/rabbithole-synthesis.md#patterns`
- [low] An index can reach the two billion document limit. Sources: https://www.mongodb.com/docs/atlas/architecture/current/cost-saving-config.md — source: `~/.global-ai-hub/research/.pipeline/mongodb-2026-10-14/frb/index-storage-cost-267559520b/rabbithole-synthesis.md#troubleshooting`

## Facts and statements

- [low] The MongoDB Performance Advisor suggests dropping unused, redundant, and hidden indexes to improve write performance and increase storage space. Sources: https://www.mongodb.com/docs/atlas/performance-advisor/drop-indexes.md — source: `~/.global-ai-hub/research/.pipeline/mongodb-2026-10-14/frb/index-storage-cost-267559520b/rabbithole-synthesis.md#core`
- [low] MongoDB Atlas does not use hidden indexes to support queries, but they still impact write performance and consume storage space. Sources: https://www.mongodb.com/docs/atlas/performance-advisor/drop-indexes.md — source: `~/.global-ai-hub/research/.pipeline/mongodb-2026-10-14/frb/index-storage-cost-267559520b/rabbithole-synthesis.md#core`
- [low] The Performance Advisor flags an index as unused if it has not supported a query in 7 or more days after creation or server restart, inspecting only the 20 most active collections. Sources: https://www.mongodb.com/docs/atlas/performance-advisor/drop-indexes.md — source: `~/.global-ai-hub/research/.pipeline/mongodb-2026-10-14/frb/index-storage-cost-267559520b/rabbithole-synthesis.md#core`
- [low] The Performance Advisor flags an index as redundant if it matches the prefix of another index in the same collection. Sources: https://www.mongodb.com/docs/atlas/performance-advisor/drop-indexes.md — source: `~/.global-ai-hub/research/.pipeline/mongodb-2026-10-14/frb/index-storage-cost-267559520b/rabbithole-synthesis.md#core`
- [low] The Performance Advisor always recommends dropping hidden indexes, which are displayed on the Atlas Performance Advisor page from the top 20 most active namespaces at page load. Sources: https://www.mongodb.com/docs/atlas/performance-advisor/drop-indexes.md — source: `~/.global-ai-hub/research/.pipeline/mongodb-2026-10-14/frb/index-storage-cost-267559520b/rabbithole-synthesis.md#core`

## Related concepts

- Index — is a part of Index Storage Cost
- Storage — is a part of Index Storage Cost
