Data Retention Feature

Data Retention Feature

The Data Retention feature automatically removes old data from the bucket after the period you define here. The goal is to help customers to avoid unnecessary costs with data that don’t need to be kept for a long period of time.

The Data Retention is available only for the Optimized Device Data (Immutable).

To select the data Retention, you must select a
Period and a Retention. These fields will be combined to create the data retention for your bucket.


The Period field allows you to select a chunk period to retain data. In combination with the Retention field, it will define how long the data should be kept in your bucket.

Each period (chunk) is limited to 1 million registers.
For example, setting this field to Monthly means that you can store up to 1 Million data registers per month, while settings this field to Daily means that you can store only 1 Million data registers per day. Make sure you take this into consideration, as you cannot change this Period field later on. 


The Retention field is used to define how many Periods to retain data for. You can always change the Retention later, and the maximum value for this field depends on the selected Period. Below you will find all the possible values for the retention:

Retention range

The retention feature will keep data for the selected retention plus the data in the current period. 
If you select a value of 0 for this field, it means that the data will be retained only for the current Period. 
For example, if you select a Period of Weekly and a Retention of '0' , your data will be removed the next Monday at 00:00 UTC, regardless if you created your bucket, say, on Friday. If you select '1' instead, you will always have data from the whole last week plus the data from this current week - data length can vary from 7 days (minimum) to 13 days (maximum).

How it Works

The Period does NOT take into account when you created your bucket, it will check if your data should be deleted at certain times regardless of when your bucket was created.

Data starts being removed at
DailyStarts removing old data every day at 00:00 UTC
WeeklyStarts removing old data every Monday at 00:00 UTC
MonthlyStarts removing old data on the first day of every Month at 00:00 UTC
QuarterlyStarts removing old data every 3 months at 00:00 UTC, starting from January 1st. 

Data is removed by a task that starts in the times defined above. Therefore, data in your bucket can still be presented for several hours after the period crosses the 0:00h in your local time. 


You can also access the data retention parameters using our SDK.

    • Related Articles

    • Simulator (Data Stream)

      If you don't have a real sensor (hardware) to send data through the Internet in hand, you can still try out all of TagoIO's features by selecting some device simulators that will feed the data buckets in real-time. It's like you were adding a virtual ...
    • Deleting Data

      DEL - Headers: Authorization: Your device token. Optional Query Strings KEY TYPE DESCRIPTION query string pre-defined by Tago qty string Maximum number of data to be returned start_date string Start date end_date string end ...
    • Getting Data

      HEADERS: Authorization: Your-Device-Token GET - QUERYSTRINGS: KEY TYPE DESCRIPTION variable string || array Get variables query string pre-defined by Tago qty string Maximum number of data to be returned.  start_date string ...
    • Data Export

      You can download data from your buckets using one of the available formats. Go to the list of Buckets, and click on the 'Export Data' icon located on the top right. Select the format of the file you want to download: CSV XLS JSON XML You can select ...
    • Data Storage

      Data storage is related to the variables stored in the data buckets that are fed by the devices. Each register is composed of a variable and all metadata associated with it. The data storage limit will define the maximum storage capacity for the ...
        Take a look at our updated roadmap here!
          What would you like to see on TagoIO?

            Future and past Webinars