MongoDB Time Series
Build and run data-intensive analytical applications by combining the flexibility of the document model with time series collections.
One platform for all your data
Seamlessly manage the entire time series data lifecycle—ingest, storage, analysis, visualization, and archive. There's no need to worry about performance or scalability since columnar storage and compression are optimized for query speed and cost efficiency, even as data grows over time.
.svg)
Reduce complexity and cost
Eliminate costly, specialized databases that lead to complex data silos, data movement, and operational overhead. Instead, efficiently and securely manage both time series and operational data within a single versatile, modern data platform.
Automatically store time series data in a specialized columnar format optimized for high storage efficiency, reduced I/O, and low latency queries.
Chief Technology Officer, Picap
Chief Technology Officer, Picap
VP of SCALAR, ZF Group
Chief Technology Officer, Ceto
CEO, Digitread Connect
Learning hub
Quick and easy resources to get started with MongoDB Time Series.FAQ
A MongoDB Time Series collection is a specialized collection type designed to store sequences of measurements over a period of time. Unlike standard collections, it automatically organizes data into an internal, optimized storage format (bucketing) that reduces disk usage and improves query performance for time-based patterns.
Use time series collections when your data includes a timestamp for each measurement, metadata labels identifying the source (e.g., sensor ID, stock symbol), and metrics that change over time (e.g., temperature, price). If your workload is insert-heavy and requires range-based queries or real-time analytics, time series collections offer significantly better compression and speed than standard collections.
MongoDB uses a bucketing pattern. It groups documents with the same metadata from similar time ranges into a single internal "bucket." Starting in MongoDB 5.2, it also uses columnar compression, which can reduce the storage footprint by over 90% and minimize I/O during queries.
Yes. Since MongoDB 6.0+, you can create secondary indexes on any field, including measurements and metadata. This allows you to perform fast lookups on non-time-based criteria while still benefiting from the optimized storage of a time series collection.
Yes. While time series data is typically append-only, MongoDB supports updates and deletes. However, updates are generally restricted to the metaField for performance reasons. If you need to automatically expire old data, you can configure automatic removal using the expireAfterSeconds parameter.
Yes. You can shard time series collections to scale horizontally. For the best performance, it is recommended to use the metaField (the source identifier) as part of your shard key to ensure data from the same source is co-located.
Start building with time series collections
- Native time series collections
- Optimized storage and performance
- Unified query and analytics
- Available globally