Reference architecture GenAI - final review

deploy-manage/reference-architectures.md

@@ -31,8 +31,6 @@
 
 | Architecture | When to use |
 | --- | --- |
-| [*Hot/Frozen - High Availability*](/deploy-manage/reference-architectures/hotfrozen-high-availability.md)<br>A high availability architecture that is cost optimized for large time-series datasets. | * Have a requirement for cost effective long term data storage (many months or years).<br>* Provide insights and alerts using logs, metrics, traces, or various event types to ensure optimal performance and quick issue resolution for applications.<br>* Apply Machine Learning and Search AI to assist in dealing with the large amount of data.<br>* Deploy an architecture model that allows for maximum flexibility between storage cost and performance.<br> |
+| [*Hot/Frozen - High Availability*](/deploy-manage/reference-architectures/hotfrozen-high-availability.md)<br>A high availability architecture that is cost optimized for large time-series datasets. | • Have a requirement for cost effective long term data storage (many months or years).<br>• Provide insights and alerts using logs, metrics, traces, or various event types to ensure optimal performance and quick issue resolution for applications.<br>• Apply {{ml-app}} and Search AI to assist in dealing with the large amount of data.<br>• Deploy an architecture model that allows for maximum flexibility between storage cost and performance.<br> |
+| [*GenAI search - High Availability*](/deploy-manage/reference-architectures/genai-search-high-availability.md)<br>A high availability architecture that is performance-optimized for low-latency information retrieval using lexical, temporal, geospatial, and/or vector/semantic search, or a hybrid combination of each. | • Lexical, vector/semantic, temporal, geospatial, hybrid, and multi-modal search (text, audio, image, video).<br>• Generative AI applications such as assistants, agents, and agentic workflows using Retrieval Augmented Generation (RAG) and/or Model Context Protocol (MCP).<br>• Integrating with foundation models (Azure OpenAI, Anthropic, Amazon Bedrock, etc.) and re-ranking techniques for optimal relevance.<br>• Applications requiring low-latency time series analytics and {{nlp}}.<br>• Integration with common development frameworks for Large Language Models, including LlamaIndex, LangChain, and LangSmith.<br> |
 | Additional architectures are on the way.<br>Stay tuned for updates. |  |
-
-
-