Setting the file. One moment.
Chapter 45 · Amazon Opensearch Service
Subchapter 45.19
references/provisioning-domain-deploy-search.mdMarkdown2 KBView on GitHub
Deploy index configuration, ML models, and pipelines to a provisioned domain.
Create the index with mappings from local setup:
PUT <domain-endpoint>/<index-name>
{
"settings": { ... },
"mappings": { ... }
}Configure replicas (1-2) for high availability.
POST <domain-endpoint>/_plugins/_ml/models/_register?deploy=true
{
"name": "huggingface/sentence-transformers/all-MiniLM-L12-v2",
"version": "1.0.1",
"model_format": "TORCH_SCRIPT"
}See provisioning-agentic-setup.md Steps 1-2 for IAM role and connector setup pattern.
Test inference:
POST <domain-endpoint>/_plugins/_ml/models/<model-id>/_predict
{ "parameters": { "inputText": "hello world" } }PUT <domain-endpoint>/_ingest/pipeline/<pipeline-name>
{
"description": "Embedding pipeline",
"processors": [{
"text_embedding": {
"model_id": "<model_id>",
"field_map": { "<text-field>": "<vector-field>" }
}
}]
}Attach to index:
PUT <domain-endpoint>/<index-name>/_settings
{ "index.default_pipeline": "<pipeline-name>" }PUT <domain-endpoint>/_search/pipeline/<search-pipeline-name>
{
"phase_results_processors": [{
"normalization-processor": {
"normalization": { "technique": "min_max" },
"combination": { "technique": "arithmetic_mean", "parameters": { "weights": [0.3, 0.7] } }
}
}]
}Index test documents and verify pipeline processing with appropriate search queries.