OpenSearch scaling manager is used to elastically scale a cluster to ensure optimum cluster performance and expenses involved. Scaling Manager can automatically scale up or scale down an OpenSearch node based on the effect of load on metric in cluster.
Scaling Manager can be used to automate manual scale up, scale down and reduce the manual effort to achieve the same. Scale up, Scale down can happen whenever it meets the criteria which is mentioned by the user. In addition to this there is event based scaling where as scale up, scale down happens at specific time.
We encourage you to explore the and contribute to Scaling Manager.
To get started find the guidelines for contributing OpenSearch Scaling Manager
To learn more about Scaling Manager, take a look at the following resources:
-
Design Architecture - Design Architecture of Scaling Manager
Open Search Simulator is a python sub project that attempts to mimic to behavior of an AWS on which OpenSearch is deployed. It exposes a set of APIs that let's user get and set cluster paramaters like cpu usage statistics, number of nodes of cluster, etc.
Find more about Trouble Shooting.