Using the JDBC Vector Store connector

Overview

JDBC vector store is a Java-specific feature, available only for Java applications.

Overview

JDBC vector store is a Java-specific feature, available only for Java applications.

Overview

The JDBC Vector Store connector can be used to access and manage data in SQL databases. The connector has the following characteristics.

Limitations

PostgreSQL leverages pgvector for vector indexing and search, uniquely offering approximate search capabilities. Other providers lack support for indexing, providing only exhaustive vector search.

Getting started

Include the latest version of the Semantic Kernel JDBC connector in your Maven project by adding the following dependency to your pom.xml:

<dependency>
    <groupId>com.microsoft.semantic-kernel</groupId>
    <artifactId>semantickernel-data-jdbc</artifactId>
    <version>[LATEST]</version>
</dependency>

You can then create a vector store instance using the JDBCVectorStore class, having the data source as a parameter.

import com.microsoft.semantickernel.data.jdbc.JDBCVectorStore;
import com.microsoft.semantickernel.data.jdbc.JDBCVectorStoreOptions;
import com.microsoft.semantickernel.data.jdbc.JDBCVectorStoreRecordCollection;
import com.microsoft.semantickernel.data.jdbc.JDBCVectorStoreRecordCollectionOptions;
import com.microsoft.semantickernel.data.jdbc.mysql.MySQLVectorStoreQueryProvider;
import org.postgresql.ds.PGSimpleDataSource;

public class Main {
    public static void main(String[] args) {
        // Configure the data source
        PGSimpleDataSource dataSource = new PGSimpleDataSource();
        dataSource.setUrl("jdbc:postgresql://localhost:5432/sk");
        dataSource.setUser("postgres");
        dataSource.setPassword("root");

        // Build a query provider
        // Other available query providers are PostgreSQLVectorStoreQueryProvider, SQLiteVectorStoreQueryProvider
        // and HSQDBVectorStoreQueryProvider
        var queryProvider = MySQLVectorStoreQueryProvider.builder()
                .withDataSource(dataSource)
                .build();

        // Build a vector store
        var vectorStore = JDBCVectorStore.builder()
                .withDataSource(dataSource)
                .withOptions(JDBCVectorStoreOptions.builder()
                        .withQueryProvider(queryProvider)
                        .build())
                .build();
    }
}

You can also create a collection directly.

var collection = new JDBCVectorStoreRecordCollection<>(
    dataSource,
    "skhotels",
    JDBCVectorStoreRecordCollectionOptions.<Hotel>builder()
        .withRecordClass(Hotel.class)
        .build()
);