Object Regions
By default, Tigris automatically and intelligently distributes your data close to the users. However, there might be instances where you wish to manage data locality, either to enhance access speed or to comply with legal regulations by limiting data access to specific regions.
To achieve this, simply include the X-Tigris-Regions header in your PUT
requests. Tigris will then write data only to the regions specified in this
header, respecting your preferences and legal requirements.
The list of regions is available here.
Tigris Region Header
This header may include a single or comma-separated list of multiple Tigris regions mentioned here.
The table below illustrates examples of how you can manage data regions for different use cases.
| Use-Case | Header |
|---|---|
| Restrict to Europe | "X-Tigris-Regions:fra" OR "X-Tigris-Regions:lhr" |
| Restrict to Europe(multiple copies) | "X-Tigris-Regions:fra,lhr" |
| Restrict to USA | "X-Tigris-Regions:sjc" OR "X-Tigris-Regions:iad" |
| Restrict to USA(multiple copies) | "X-Tigris-Regions:sjc,iad" |
| Multiple Copies Globally | "X-Tigris-Regions:sjc,fra,sin" |
Global copies may be intended to boost access speed by dispersing copies across different regions spanning various countries or continents and you want copies to always available.
When Tigris stores multiple copies of data, the storage cost is calculated as the number of copies multiplied by the storage size. By default, this count is one. However, if you control the distibution of data and specify for example, two regions in the request header, the count will be considered as two.
Next steps
- Check out the Example usage for more details on how to use them in your application.