I'm currently working on AOSS (AWS OpenSearch Serverless) after working a while on Opensearch (OS)/ Elasticsearch (ES), and I would like to understand how AOSS manage indices shards to be able to prevent errors, behaviours and else..
So, does someone have an idea about it ?
On OS/ES we can handle shards by ourself, but as I understood on AOSS, it will be managed / modified on the fly.
Is there any way to know at what time it will be modified ? Is it related to the ingestion volume, duration, or is based on average usage ?
Thank you in advance for your answers.