# Ilum

> Ilum is a free data lakehouse platform for running Apache Spark on Kubernetes (Spark, Jupyter,
> Airflow, and MLflow in one place), plus enterprise Hadoop and Cloudera CDP migration delivered by the
> Bifrost toolkit. Bifrost provides three migration paths: in-place Cloudera CDP to open-source Hadoop
> (eliminate licensing), Hadoop to Ilum on Kubernetes, and Cloudera CDP directly to Ilum on Kubernetes.
> Migrations are auditable, reversible until an explicit finalize step, run with zero data loss, and land
> you on open formats with no vendor lock-in.

## Hadoop & Cloudera Migration (Bifrost)

- [Hadoop & Cloudera Migration](https://ilum.cloud/hadoop-migration): Business overview of migrating off
  Cloudera and Hadoop. Covers costs and ROI, the weekend in-place cutover, the Bifrost framework, a
  comparison of Bifrost vs do-it-yourself vs staying on Cloudera, how to avoid Cloudera vendor lock-in, and
  FAQs. Bifrost can replatform a Cloudera CDP cluster onto license-free open-source Hadoop in a single
  weekend (~6-8h overnight downtime), or modernize to a Kubernetes lakehouse table by table.
- [Bifrost Overview](https://ilum.cloud/docs/bifrost/overview/): What Bifrost is, design principles
  (auditable, reversible, zero data loss), and automation ceilings.
- [Migration Paths](https://ilum.cloud/docs/bifrost/migration-paths/): How to choose between Classic,
  Modernize, and Direct for your estate.
- [Classic, in-place CDP to Hadoop](https://ilum.cloud/docs/bifrost/classic-migration/): The 11-phase
  in-place procedure, stop-and-swap vs shrink-and-grow strategies, weekend timing, and rollback window.
- [Modernize, Hadoop to Ilum on Kubernetes](https://ilum.cloud/docs/bifrost/modernize-migration/):
  Table-by-table modernization with Apache Iceberg, Trino, and zero downtime.
- [Direct, Cloudera CDP to Ilum on Kubernetes](https://ilum.cloud/docs/bifrost/direct-migration/):
  Single-engagement migration from Cloudera straight to the Ilum lakehouse.
- [Component Migrations](https://ilum.cloud/docs/bifrost/component-migrations/): How HDFS, YARN, Hive, Oozie,
  Ranger, HBase, and Atlas map to modern equivalents.
- [Validation & Rollback](https://ilum.cloud/docs/bifrost/validation-and-rollback/): Row-count parity,
  value-level data diffs, benchmarks, and the reversible-until-finalize model.

## Platform

- [Ilum, Free Data Lakehouse](https://ilum.cloud): Apache Spark on Kubernetes for cloud, on-premise, and
  hybrid environments.
- [What is a Data Lakehouse?](https://ilum.cloud/data-lakehouse): Unifying data lakes and warehouses for
  scalable, flexible data solutions.
- [Data Mesh](https://ilum.cloud/data-mesh): Decentralized, domain-oriented data ownership.
- [Medallion Architecture](https://ilum.cloud/medallion-architecture): Refining raw data through Bronze,
  Silver, and Gold layers.
- [Data Sovereignty](https://ilum.cloud/data-sovereignty): Strategies for compliance, control, and
  modernization off legacy Hadoop.
- [ML Model Development](https://ilum.cloud/model-development): Production-ready Spark microservices with
  automated REST APIs and Kubernetes scaling.
- [Product Features](https://ilum.cloud/product/features): Interactive Spark sessions, cluster management,
  data lineage, and multi-environment support.
- [Architecture](https://ilum.cloud/product/architecture): How Ilum brings Apache Spark to Kubernetes.

## Resources

- [Documentation](https://ilum.cloud/docs/): Full product and API documentation.
- [Getting Started](https://ilum.cloud/resources/getting-started): Install Ilum and run your first workload.
- [Pricing](https://ilum.cloud/pricing): Plans from Free to Enterprise.
- [Get Access](https://ilum.cloud/get-access): Start with Ilum for free.
- [About Us](https://ilum.cloud/product/about-us): The team behind Ilum.

## Contact

- For Bifrost migration engagements and Ilum Enterprise: info@ilum.cloud