Limiting the Number of Editions in a Dataset

By default, there is no limit on the number of editions that can be created and preserved in a dataset. Each time an unstructured pipeline or Export Step is run, a new edition is generated and the TTL for that edition is written to disk. You can modify a dataset, however, to set a limit on the number of editions to archive. When a limit is set, Anzo to ages off and removes older editions from disk. Follow the steps below to limit the number of editions for a dataset.

1. In the Anzo application, expand the Blend menu and click Datasets. Anzo displays the Datasets screen, which lists the catalog of datasets. For example:

   

2. Click the dataset for which you want to configure the maximum number of editions. The Explore screen is displayed.
3. Click the Overview tab. Then click Advanced to expand the screen and show the advanced settings. The image below shows the settings.

   

4. Click in the Maximum Number of Archives field to make it editable. Then type the number of archived editions that you want to keep for the dataset. For example:

   

5. Click the checkmark icon () to save the change.

The dataset is now configured to limit the number of editions that are retained. When the maximum number is reached, Anzo ages off and removes the oldest edition. For unstructured datasets, Anzo also removes the corresponding Elasticsearch Index JSON backups and pipeline runs.