Databricks documentation archive
Important
This documentation has been retired and might not be updated. The products, services, or technologies mentioned in this content are no longer supported.
In this archive, you can find earlier versions of documentation for Databricks products, features, APIs, and workflows.
Dev tools
- Manage libraries with
%conda
commands (legacy) - Explore and create tables in DBFS
- Transactional writes to cloud storage with DBIO
- Browse files in DBFS
- FileStore
- Hive table (legacy)
- Koalas
- Skew join optimization using skew hints
- Legacy UniForm IcebergCompatV1
- Workspace libraries (legacy)
- Databricks CLI (legacy)
- What is dbx by Databricks Labs?
- dbutils.library
- Migrate to Spark 3.x
- VSCode with Git folders
- VSCode workspace directory
- Basic authentication (End of life)
Machine learning and AI
- Load data using Petastorm
- Share feature tables across workspaces (legacy)
- MLeap ML model export
- Train a PySpark model and save in MLeap format
- Deploy MLeap model on SageMaker
- Distributed training with TensorFlow 2
- Horovod
- Model serving (legacy)
- Serverless Real-Time Inference (preview)
- Optimized LLM serving
- Set up and considerations for
ai_generate_text()
- Analyze customer reviews with
ai_generate_text()
and OpenAI - Apache Spark MLlib and automated MLflow tracking
Storage
- End-of-life for legacy workspaces
- Amazon S3 source with Amazon SQS (legacy)
- Azure Blob storage file source with Azure Queue Storage (legacy)
- Azure Cosmos DB
- Structured Streaming writes to Azure Synapse
- Connecting Databricks and Azure Synapse with PolyBase (legacy)
- Neo4j
- Read and write XML data using the
spark-xml
library - Accessing Azure Data Lake Storage Gen1 from Databricks
- Configure Delta storage credentials
- Connect to Azure Blob Storage with WASB (legacy)