Plans for the Elasticsearch Add-On for Heroku differ based on:
- How much memory and disk space are available
- How many data centers your cluster is replicated across to achieve high availability
Available memory is an important factor for performance when sizing your Elasticsearch cluster, and replicating across multiple data centers is important for the resilience of production applications.
To learn more about what plans are available for Heroku users, see the Elasticsearch add-on in the Elements Marketplace.
You should time the migration to a new plan to ensure proper application function during the migration process. A cluster that is already overwhelmed with requests will take much longer to migrate to a larger capacity; if your workload warrants a plan change to increase capacity, migrate to a larger plan early.
To migrate to a new plan, use the
heroku addons:upgrade command and include one of the available plans:
foundelasticsearch:dachs-standard foundelasticsearch:beagle-standard foundelasticsearch:dachs-ha foundelasticsearch:boxer-standard foundelasticsearch:beagle-ha foundelasticsearch:labrador-standard foundelasticsearch:boxer-ha foundelasticsearch:husky-standard foundelasticsearch:labrador-ha foundelasticsearch:husky-ha
For example: Migrate from the smallest, default
dachs-standard plan to the larger
beagle-ha plan that includes high availability for MY_APP:
heroku addons:upgrade foundelasticsearch:beagle-ha --app MY_APP Changing foundelasticsearch-defined-nnnnn on MY_APP from foundelasticsearch:dachs-standard to foundelasticsearch:beagle-ha... done, $201/month
Upgrading to a new plan may involve extending the existing cluster with new nodes and migrating data from the old nodes to the new ones. When the migration is finished, the old nodes are shut down and removed from the cluster. For HA clusters, you can continue to search and index documents while this plan change is happening.