{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "product": {
    "name": "LakeXpress",
    "tagline": "On-prem to cloud, in minutes, not hours",
    "category": "Data Integration",
    "type": "Command-line database-to-Parquet export and lakehouse publishing tool",
    "version": "0.4.10",
    "releaseDate": "2026-09-24",
    "description": "LakeXpress is a command-line tool that exports database tables and views to partitioned Parquet files and publishes them to cloud data platforms. It uses the FastBCP engine to stream data in parallel without exhausting memory. Files are written to local disk or cloud object storage (S3, GCS, Azure, OneLake), and can then be exposed as external tables or loaded as internal tables in Snowflake, Databricks, Microsoft Fabric, BigQuery, Amazon Redshift or MotherDuck, or registered in AWS Glue or DuckLake. Each pipeline is stored as a sync configuration in the LakeXpress DB and run again with one command, fully or incrementally.",
    "shortDescription": "Export database schemas to Parquet on local disk or cloud storage, in parallel, and publish them to Snowflake, Databricks, Fabric, BigQuery, Redshift, MotherDuck, Glue or DuckLake with one command.",
    "architecture": {
      "summary": "LakeXpress connects four components. The source database is read by LakeXpress (FastBCP engine) and written as Parquet to intermediate storage; publishing to a target platform is optional; the LakeXpress DB holds configurations and run state.",
      "components": [
        {
          "name": "Source database",
          "role": "Tables and views to export. LakeXpress only needs read access.",
          "options": ["Oracle", "PostgreSQL", "SQL Server", "MySQL", "MariaDB", "SAP HANA", "Teradata"]
        },
        {
          "name": "Intermediate storage",
          "role": "Where the Parquet files are written. Files can be used as they are, or published.",
          "options": ["Local filesystem", "AWS S3", "S3-compatible (MinIO, OVH...)", "Google Cloud Storage", "Azure Blob Storage", "Azure Data Lake Storage Gen2", "Microsoft Fabric OneLake"]
        },
        {
          "name": "Target platform (optional)",
          "role": "Data platform or catalog where tables are created over, or loaded from, the files.",
          "options": ["Snowflake", "Databricks", "Microsoft Fabric", "Amazon Redshift", "BigQuery", "MotherDuck", "AWS Glue", "DuckLake"]
        },
        {
          "name": "LakeXpress DB",
          "role": "Stores sync configurations, the history and status of every run, and incremental watermarks. Required by every command.",
          "options": ["PostgreSQL (recommended)", "SQL Server", "MySQL", "SQLite", "DuckDB"]
        }
      ],
      "workflow": [
        "lxdb init: create the LakeXpress tables in the LakeXpress DB (once; the first config create or sync also does it)",
        "config create: define the source, tables, storage and optional publishing target as a sync configuration",
        "sync: export the tables to Parquet, then publish them if a target is configured; run again for each later export"
      ]
    },
    "relationWithFastBCP": {
      "summary": "FastBCP is the arpe.io bulk export engine. LakeXpress orchestrates FastBCP extracts, table by table and in parallel, and adds configuration storage, incremental watermarks, retries, resume, logging and publishing.",
      "trialEdition": "FastBCP is bundled in the engine/ folder and found automatically",
      "fullEdition": "FastBCP is not bundled yet: install it separately, with its own FastBCP.lic licence next to the FastBCP binary, and pass its folder with --fastbcp_dir_path",
      "fastBCPWebsite": "https://fastbcp.arpe.io"
    },
    "keyDifferentiators": [
      "One configuration, then one command per run: a sync configuration is created once and run with the same sync command every time",
      "Whole schemas at once, with SQL LIKE filters on schemas and tables and row-count filters",
      "Two levels of parallelism: several tables at the same time, and several FastBCP processes inside one large table",
      "Export and publishing in one step, to eight target platforms, as external or internal tables",
      "Watermark-based incremental syncs with append, upsert or delete_insert loading strategies",
      "Run history in a database: status of every run and table, resume of failed runs",
      "Native binaries for Linux and Windows, unzip and run"
    ],
    "targetAudience": [
      "Data Engineers",
      "Data Platform and Lakehouse teams",
      "Database Administrators",
      "Teams migrating on-premises databases to Snowflake, Databricks, Microsoft Fabric, BigQuery or AWS",
      "Software vendors embedding database-to-lakehouse export (OEM)"
    ],
    "primaryUseCases": [
      "Moving on-premises database schemas to a cloud lakehouse as Parquet",
      "Feeding the bronze layer of a data lake on S3, ADLS, GCS or OneLake",
      "Publishing source tables to Snowflake, Databricks, Fabric, BigQuery, Redshift, MotherDuck, Glue or DuckLake",
      "Daily or intra-day incremental syncs of large fact tables",
      "Replacing Spark clusters or ETL pipelines used only to copy tables"
    ],
    "deployment": {
      "type": "Self-hosted command-line executable",
      "platforms": ["Linux (x64)", "Windows (x64)"],
      "installation": "Download the archive (trial: https://arpe.io/get-trial?product=LakeXpress), unzip, and on Linux chmod +x LakeXpress",
      "scheduling": "Run the same sync command from cron, Windows Task Scheduler, Airflow or another scheduler, and check the exit code"
    },
    "aiIntegrations": {
      "agentSkill": {
        "name": "lakexpress",
        "repository": "https://github.com/arpe-io/arpeio-skills/tree/main/plugins/lakexpress",
        "documentation": "https://lakexpress-docs.arpe.io/latest/ai/agent-skills",
        "install": "/plugin install lakexpress --marketplace arpe-io/arpeio-skills"
      },
      "mcpServer": {
        "name": "arpeio-mcp",
        "repository": "https://github.com/arpe-io/arpeio-mcp",
        "documentation": "https://lakexpress-docs.arpe.io/latest/ai/mcp-server",
        "hostedUrl": "https://arpe-io-arpeio-mcp.hf.space/sse",
        "localInstall": "pip install arpeio-mcp"
      }
    },
    "website": "https://lakexpress.arpe.io",
    "documentation": "https://lakexpress-docs.arpe.io/latest/",
    "documentationForLLMs": "https://lakexpress-docs.arpe.io/llms.txt",
    "trial": "https://arpe.io/get-trial?product=LakeXpress",
    "pricing": "https://lakexpress.arpe.io/pricing",
    "quoteRequest": "https://lakexpress.arpe.io/quote",
    "supportContact": "support@arpe.io",
    "vendor": {
      "name": "Architecture & Performance",
      "shortName": "arpe.io",
      "website": "https://www.arpe.io",
      "relatedProducts": [
        { "name": "FastBCP", "url": "https://fastbcp.arpe.io" },
        { "name": "FastTransfer", "url": "https://fasttransfer.arpe.io" },
        { "name": "MigratorXpress", "url": "https://migratorxpress.arpe.io" },
        { "name": "Arpeio ADBC Drivers", "url": "https://adbc-drivers.arpe.io" }
      ]
    }
  },
  "metadata": {
    "lastUpdated": "2026-10-22",
    "dataFormat": "JSON",
    "purpose": "AI Agent Knowledge Base",
    "audience": "LLMs and AI Agents",
    "schemaVersion": "1.0.0"
  }
}