# Data migration Memgraph supports multiple data sources for importing data into a running instance. Whether your data is structured in files, relational databases, or other graph databases, Memgraph provides the flexibility to integrate and analyze your data efficiently. Memgraph supports file system imports like Parquet, CSV and JSONL files, offering efficient and structured data ingestion. **However, if you want to migrate directly from another data source, you can use the [`migrate` module](https://memgraph.com/docs/advanced-algorithms/available-algorithms/migrate)** from Memgraph MAGE for a direct transition without the need to convert your data into intermediate files. > **Note** > > In order to learn all the pre-requisites for importing data into Memgraph, check > [import best practices](https://memgraph.com/docs/data-migration/best-practices). > **Info** > > If you have a SQL data model and > want to migrate to Memgraph, you can try out our [Agent](https://memgraph.com/docs/ai-ecosystem/agents) > that leverages the LLM to automate the process of modeling and migration. ## File types ### Parquet files Parquet files can be imported efficiently from the local disk and from s3:// using the [LOAD PARQUET clause](https://memgraph.com/docs/querying/clauses/load-parquet). ### CSV files CSV files provide a simple and efficient way to import tabular data into Memgraph using the [LOAD CSV clause](https://memgraph.com/docs/querying/clauses/load-csv). Using the [LOAD CSV import guide](https://memgraph.com/docs/data-migration/csv), you can implement your own CSV import through Memgraph LAB, or a simple database driver. ### JSON files Memgraph supports importing data from JSON files, allowing structured and semi-structured data to be efficiently loaded, using the [`json_util` module](https://memgraph.com/docs/advanced-algorithms/available-algorithms/json_util) and [`import_util` module](https://memgraph.com/docs/advanced-algorithms/available-algorithms/import_util). Memgraph also support JSONL files in which every line is formatted as a separate JSON document. Such JSONL files can be efficiently imported using the [LOAD JSONL clause](https://memgraph.com/docs/querying/clauses/load-jsonl). Check out the [JSON import guide](https://memgraph.com/docs/data-migration/json). ### Cypherl file Cypherl file is a simple format which defines a cypher command per line. It is usually divided in 2 parts: - `CREATE` statements for nodes, followed by - `MATCH`, `MATCH` + `CREATE` statements for relationships An example for node creation in Cypherl would be similar to this query: ```cypher CREATE (:Person {id: 1, name: "John", age: 18}); CREATE (:Person {id: 2, name: "Mark", age: 25}); ``` While the example for relationship creation in Cypherl would be similar to this query: ```cypher MATCH (p1:Person {id: 1}) MATCH (p2:Person {id: 2}) CREATE (p1)-[:KNOWS]->(p2); ``` If you have existing Cypherl files defining your graph structure, Memgraph can execute them directly to populate your database. See the [Cypherl import guide](https://memgraph.com/docs/data-migration/cypherl). ## Database management systems ### DuckDB Memgraph is fully compatible with all data source systems supported by DuckDB. Using our [DuckDB migration module](https://memgraph.com/docs/advanced-algorithms/available-algorithms/migrate#duckdb), you can load data from any DuckDB-supported source, run analytical queries directly on it, and import the results into Memgraph. - [List of DuckDB supported data sources](https://duckdb.org/docs/stable/data/data_sources.html) ### MySQL Memgraph can integrate with MySQL databases, enabling seamless migration and real-time synchronization. Learn more about [MySQL data import](https://memgraph.com/docs/advanced-algorithms/available-algorithms/migrate#mysql). ### PostgreSQL Memgraph can integrate with PostgreSQL databases using the [PostgreSQL migration module](https://memgraph.com/docs/advanced-algorithms/available-algorithms/migrate#postgresql). ### SQL Server For enterprises using Microsoft SQL Server, Memgraph supports data import using ETL processes or direct queries. Read the [SQL Server import documentation](https://memgraph.com/docs/advanced-algorithms/available-algorithms/migrate#sql_server). ### StarRocks Memgraph supports migration from **StarRocks** through its compatibility with the [MySQL migration module](https://memgraph.com/docs/advanced-algorithms/available-algorithms/migrate#mysql). Since StarRocks implements a MySQL-compatible protocol, you can use the same module to connect and migrate data into Memgraph. ### OracleDB Memgraph can extract and import data from OracleDB, leveraging SQL queries or external connectors. Find more details in the [OracleDB import guide](https://memgraph.com/docs/advanced-algorithms/available-algorithms/migrate#oracle_db). ## Graph databases ### Neo4j Memgraph offers a dedicated [Neo4j migration module](https://memgraph.com/docs/advanced-algorithms/available-algorithms/migrate#neo4j) that allows users to easily transition from Neo4j. By connecting to an existing Neo4j instance, you can migrate your entire graph to Memgraph using a single Cypher query. - [Step-by-step guide to migrate from Neo4j using a single Cypher query](https://memgraph.com/docs/data-migration/migrate-from-neo4j-using-single-cypher-query) ### Memgraph The fastest way to migrate from one Memgraph instance to another is by using [snapshot recovery](https://memgraph.com/docs/database-management/backup-and-restore#restore-data), which leverages Memgraph’s proprietary durability files for efficient data backup and restore. This process can be parallelized during startup for even faster recovery. If you only want to migrate a portion of your data, Memgraph also provides a [Memgraph migration module](https://memgraph.com/docs/advanced-algorithms/available-algorithms/migrate#memgraph) that allows selective migration between Memgraph instances. We also provide guides to migrate from Neo4j using a CSV file or by using a single Cypher query: - [Migrate from Neo4j](https://memgraph.com/docs/data-migration/migrate-from-neo4j) - [Migrate from Neo4j using a single Cypher query](https://memgraph.com/docs/data-migration/migrate-from-neo4j-using-single-cypher-query) ## RPC Protocols ### Arrow Flight Memgraph supports migration from any system that uses **Arrow Flight** as its RPC protocol for fast and efficient data transfer. The [Arrow Flight migration module](https://memgraph.com/docs/advanced-algorithms/available-algorithms/migrate#arrow_flight) enables you to stream data directly into Memgraph. ## Streaming Sources ### Apache Kafka Memgraph offers [native integration with **Apache Kafka**](https://memgraph.com/docs/data-streams#kafka-and-redpanda), enabling real-time data streaming directly into the graph. It includes built-in Kafka consumers for easy setup and processing of incoming data streams. For more advanced setups, users can [integrate with **Kafka Connect**](https://memgraph.com/docs/data-streams/kafka) or build custom consumers using [Memgraph client libraries](https://memgraph.com/docs/client-libraries) available in multiple programming languages. ### Apache Pulsar Memgraph offers [native integration with **Apache Pulsar**](https://memgraph.com/docs/data-streams#pulsar), enabling real-time data streaming directly into the graph. It includes built-in Pulsar consumers for easy setup and processing of incoming data streams. For more advanced setups, users can build custom consumers using [Memgraph client libraries](https://memgraph.com/docs/client-libraries) available in multiple programming languages. ### Apache Spark Memgraph is compatible with the **Neo4j Spark Connector**, which makes it easy to integrate Spark-powered pipelines with Memgraph. You can use this connector to process large-scale data in Spark and ingest the results directly into Memgraph for real-time graph analytics. - [Step-by-step guide to migrate with Apache Spark](https://memgraph.com/docs/data-migration/migrate-with-apache-spark) ### Dremio Memgraph supports migration from the **Dremio** query engine using the [Arrow Flight RPC protocol](https://memgraph.com/docs/advanced-algorithms/available-algorithms/migrate#arrow_flight), which enables high-performance data transfer from data lakes and lakehouses. You can use Dremio to query formats like Apache Iceberg and stream the results directly into Memgraph. - [Guide to migrate Iceberg tables from a data lake using Dremio](https://memgraph.com/docs/data-migration/migrate-iceberg-tables-from-data-lake-using-dremio) ## Data platforms ### ServiceNow Memgraph supports the [migration module from ServiceNow](https://memgraph.com/docs/advanced-algorithms/available-algorithms/migrate#servicenow) through its dedicated migration module. By connecting to the **ServiceNow REST API**, the module enables you to retrieve and ingest data directly into Memgraph for further analysis and visualization. ## Storage services ### Amazon S3 Memgraph supports data migration from **Amazon S3** and other **S3-compatible storage services**. Using the [S3 migration module](https://memgraph.com/docs/advanced-algorithms/available-algorithms/migrate#s3), you can load structured data stored in cloud object storage directly into Memgraph for graph-based processing and analysis. ## Other ### Data from an application or a program Memgraph offers a [wide range of client libraries](https://memgraph.com/docs/client-libraries) that can be used to connect directly to the platform and import data. ### Parquet, ORC or IPC/Feather/Arrow file If you are a Python user you can import Parquet, ORC or IPC/Feather/Arrow file into Memgraph using [GQLAlchemy](https://memgraph.github.io/gqlalchemy/how-to-guides/loaders/import-table-data-to-graph-database/). Parquet files migration are also supported with the [DuckDB migration](#duckdb). ### NetworkX, PyG or DGL graph If you are a Python user you can import NetworkX, PyG or DGL graph into Memgraph using [GQLAlchemy](https://memgraph.github.io/gqlalchemy/how-to-guides/translators/import-python-graphs/). ## Need help with data migration? Schedule a 30 min session with one of our engineers to discuss how Memgraph fits with your architecture. Our engineers are highly experienced in helping companies of all sizes to integrate and get the most out of Memgraph in their projects. Talk to us about data modeling, optimizing queries, defining infrastructure requirements or migrating from your existing graph database. No nonsense or sales pitch, just tech. ![](https://memgraph.com/docs/pages/getting-started/memgraph-office-hours.svg) - [Book a call](https://memgraph.com/office-hours)