{"id":3919,"date":"2024-12-06T11:32:15","date_gmt":"2024-12-06T19:32:15","guid":{"rendered":"https:\/\/devblogs.microsoft.com\/semantic-kernel\/?p=3919"},"modified":"2024-12-10T10:49:15","modified_gmt":"2024-12-10T18:49:15","slug":"customer-case-study-how-to-use-elasticsearch-vector-store-connector-for-microsoft-semantic-kernel-for-ai-agent-development","status":"publish","type":"post","link":"https:\/\/devblogs.microsoft.com\/agent-framework\/customer-case-study-how-to-use-elasticsearch-vector-store-connector-for-microsoft-semantic-kernel-for-ai-agent-development\/","title":{"rendered":"Customer Case Study: How to use Elasticsearch Vector Store Connector for Microsoft Semantic Kernel for AI Agent development"},"content":{"rendered":"

How to use Elasticsearch Vector Store Connector for Microsoft Semantic Kernel for AI Agent development<\/strong><\/span><\/h2>\n

Today we’re excited to feature the Elastic team to share more about their Semantic Kernel Elasticsearch Vector Store connector for Microsoft Semantic Kernel. Read the entire announcement here<\/a>. I’ll turn it over to Srikanth Manvi and Florian Bernd to dive into it.<\/p>\n

In collaboration with the\u00a0Microsoft Semantic Kernel<\/a>\u00a0team, we are announcing the availability of\u00a0Semantic Kernel Elasticsearch Vector Store Connector<\/a>, for\u00a0Microsoft Semantic Kernel<\/a>\u00a0(.NET) users. Semantic Kernel simplifies building enterprise-grade AI agents, including the capability to enhance large language models (LLMs) with more relevant, data-driven responses from a Vector Store. Semantic Kernel provides a seamless abstraction layer for interacting with Vector Stores like Elasticsearch, offering essential features such as creating, listing, and deleting collections of records and uploading, retrieving, deleting individual records.<\/p>\n

The\u00a0out-of-the-box Semantic Kernel Elasticsearch Vector Store Connector<\/a>\u00a0supports the Semantic Kernel\u00a0vector store abstractions<\/a>\u00a0which make it very easy for developers to plugin Elasticsearch as a vector store while building AI agents.<\/p>\n

Elasticsearch has a strong foundation in the open-source community and recently adopted the\u00a0AGPL license<\/a>. Combined with the open-source Microsoft Semantic Kernel, these tools offer a powerful, enterprise-ready solution. You can get started locally by spinning up Elasticsearch in a few minutes by running this command\u00a0curl -fsSL https:\/\/elastic.co\/start-local | sh\u00a0<\/code>(refer\u00a0start-local<\/a>\u00a0for details) and move to\u00a0cloud-hosted<\/a>\u00a0or\u00a0self-hosted<\/a>\u00a0versions while productionizing your AI agents.<\/p>\n

In this blog we look at how to use\u00a0Semantic Kernel Elasticsearch Vector Store Connector<\/a>\u00a0when using Semantic Kernel. A Python version of the connector will be made available in the future.<\/p>\n

High-level scenario<\/strong><\/h3>\n

In the following section we go through an example. At a high-level we are building a RAG (Retrieval Augmented Generation) application which takes a user’s question as input and returns an answer. We will use Azure OpenAI (local LLM<\/a>\u00a0can be used as well) as the LLM, Elasticsearch as the vector store and Semantic Kernel (.net) as the framework to tie all components together.<\/p>\n

If you are not familiar with RAG architectures, you can have a quick introduction with this article:\u00a0https:\/\/www.elastic.co\/search-labs\/blog\/retrieval-augmented-generation-rag<\/a>.<\/p>\n

The answer is generated by the LLM which is fed with context, relevant to the question, retrieved from Elasticsearch vectorstore. The response also includes the source that was used as the context by the LLM.<\/p>\n

RAG Example<\/span><\/strong><\/h3>\n

In this specific example, we build an application that allows users to ask questions about hotels stored in an internal hotel database. The user could e.g. search for a specific hotel, based on different criteria, or ask for a list of hotels.<\/p>\n

For the example database, we generated a\u00a0list of hotels<\/a>\u00a0containing 100 entries. The sample size is intentionally small to allow you to try out the connector demo as easily as possible. In a real-world application, the Elasticsearch connector would show its advantages over other options, such as the `InMemory` vector store implementation, especially when working with extremely large amounts of data.<\/p>\n

The complete demo application can be found in the Elasticsearch vector store connector\u00a0repository<\/a>.<\/p>\n

Let\u2019s start with adding the required NuGet packages and using directives to our project:<\/p>\n

dotnet add package \"Elastic.Clients.Elasticsearch\" -v 8.16.2\r\ndotnet add package \"Elastic.SemanticKernel.Connectors.Elasticsearch\" -v 0.1.2\r\ndotnet add package \"Microsoft.Extensions.Hosting\" -v 9.0.0\r\ndotnet add package \"Microsoft.SemanticKernel.Connectors.AzureOpenAI\" -v 1.30.0\r\ndotnet add package \"Microsoft.SemanticKernel.PromptTemplates.Handlebars\" -v 1.30.0<\/code><\/pre>\n
using System;\r\nusing System.IO;\r\nusing System.Linq;\r\nusing System.Threading.Tasks;\r\n\r\nusing Elastic.Clients.Elasticsearch;\r\nusing Elastic.Transport;\r\n\r\nusing Microsoft.Extensions.DependencyInjection;\r\nusing Microsoft.Extensions.Hosting;\r\nusing Microsoft.Extensions.VectorData;\r\nusing Microsoft.SemanticKernel;\r\nusing Microsoft.SemanticKernel.Data;\r\nusing Microsoft.SemanticKernel.Embeddings;\r\nusing Microsoft.SemanticKernel.PromptTemplates.Handlebars;<\/code><\/pre>\n
We can now create our data model and provide it with Semantic Kernel specific attributes to define the storage model schema and some hints for the text search:<\/p>\n
\/\/\/ <summary>\r\n\/\/\/ Data model for storing a \"hotel\" with a name, a description, a  description embedding and an optional reference link.\r\n\/\/\/ <\/summary>\r\npublic sealed record Hotel\r\n{\r\n\t[VectorStoreRecordKey]\r\n\tpublic required string HotelId { get; set; }\r\n\r\n\t[TextSearchResultName]\r\n\t[VectorStoreRecordData(IsFilterable = true)]\r\n\tpublic required string HotelName { get; set; }\r\n\r\n\t[TextSearchResultValue]\r\n\t[VectorStoreRecordData(IsFullTextSearchable = true)]\r\n\tpublic required string Description { get; set; }\r\n\r\n\t[VectorStoreRecordVector(Dimensions: 1536, DistanceFunction.CosineSimilarity, IndexKind.Hnsw)]\r\n\tpublic ReadOnlyMemory<float>? DescriptionEmbedding { get; set; }\r\n\r\n\t[TextSearchResultLink]\r\n\t[VectorStoreRecordData]\r\n\tpublic string? ReferenceLink { get; set; }\r\n}<\/code><\/pre>\n
The Storage Model Schema attributes (`VectorStore*`) are most relevant for the actual use of the Elasticsearch Vector Store Connector, namely:<\/p>\n