Databricks Compute

Photo by Growtika on Unsplash

Types of Databricks Compute

Regular Compute vs Serverless Compute

Regular Compute

Serverless Compute

The benefits of Serverless:

Higher user productivity:-

• User queries start instantly, with no waiting for cluster start-up.
• Add more concurrent users with instant cluster scaling.

Zero Management:-

• No configuration
• No performance tuning
• No capacity management.
• Automatic upgrades and patching.

Lower Cost:-

• Pay what you consume; eliminate idle cluster time.
• No over-provisioning of resources.
• The idle capacity was removed 10 minutes after the last query.

Features provided by Databricks Compute:

1. Runtimes — Apache Spark & Machine Learning