[DOC] ML Extensibility Phase1: ML Inference Endpoint Connectors

[sean-zheng-amazon]

What do you want to do?

• Request a change to existing documentation
• Add new documentation
• Report a technical problem with the documentation
• Other

Tell us about your request. Provide a summary of the request and all versions that are affected.
see feature details in: opensearch-project/ml-commons#746, targeting 2.8

What other resources are available? Provide links to related issues, POCs, steps for testing, etc.

[hdhalter]

@Naarcha-AWS - I updated the name of the doc issue to match the feature name. I also added this issue to the github issue in the project. This is the first phase for a framework that enables no-code integrations through web services to run ML workloads outside of an OpenSearch cluster.

[ylwu-amzn]

@Naarcha-AWS From PenTest team , we should highlight that all variables in "parameters" block can be overridden in predict request. If user want to keep the variable value immutable, they should hard code the value in body template, rather than put into parameter block.

[ylwu-amzn]

From PenTest, we should create encryption key automatically when cluster starts. So ml-commons will create a random encryption key when cluster starts and store it to a new system index .plugins-ml-config. User don't need to configure master key manually.