DSE Tiered Storage

Documentation about the features that automate the smart movement of data across different types of storage media.

To improve performance and reduce manual processes, DSE Tiered Storage automates the smart movement of data across different types of storage media. DSE Tiered Storage improves efficiency of faster and more expensive media, and mitigates the performance impact that slower storage media has on your most common queries. With DSE Tiered Storage, data is matched to the most suitable drive type, according to the performance and cost characteristics it requires.

DSE Tiered Storage is beneficial for applications like social media applications that have a lot of time series data, where recent data is accessed more frequently than older data. DSE Tiered Storage is appropriate when the data most often read is most frequently written. Common use cases appropriate for DSE Tiered Storage are described in this DataStax blog post.

DSE Tiered Storage is not recommended:
  • When entire data sets are accessed at approximately the same frequency.
  • When the data access frequency is not correlated to data age.
  • For use with DSE Search.


Increases productivity Automates the movement of data between storage media. Eliminates manually moving data.
Improves performance Data is stored by age, so that frequently accessed data is stored on solid state drives (SSDs) for fastest performance.
Transparent data access Access to the data in different storage tiers is transparent to users and developers.
Lowers storage costs Improves data center cost efficiency. Automatically stores less frequently accessed historical data on slower, less expensive storage media, such as spinning disks.
Flexible configuration options Different server configurations are easy to support and configure. Disk layout is configured per node, so you can test adjustments on single nodes before deploying cluster wide.
Compaction strategies Uses DateTieredCompactionStrategy (DTCS). DTCS compacts based on partition age to automate moving data by row between storage media.