HomepageData + AI Summit 2022 Logo
Watch on demand

Databricks and Enterprise Observability with Overwatch

On Demand

Type

  • Session

Format

  • In-Person, Virtual

Difficulty

  • Beginner

Duration

  • 40 min
Download session slides

Überblick

Join us for a quick discussion and demonstration on how Overwatch can help you understand cost, utilization, workloads, and much more across the Databricks platform TODAY.

Over the past several years, Databricks has matured from a niche spark data analysis tool into a unified data platform servicing enterprises around the world. Enterprise customers share a need to understand the state of the platform, its customers, and its workloads as they evolve through time. Furthermore, it’s paramount that leaders and governance teams are able to trace and justify spend, utilization, and trends across all dimensions of the platform across all workspaces.

Overwatch is a Databricks Labs tool used by more than 80 enterprise customers worldwide to answer exactly these questions. Overwatch is designed for the customer and operates entirely within the customer account. It offers insights into spend, utilization, efficiency, workloads, and much more across all dimensions of the Databricks Platform. Some examples include:
Global costs by workspace, by tag, by user, by notebook, by job, etc
Cluster utilization
Jobs - efficiencies, runtimes, cost of failures, common failures
Identifying anti-patterns and most affected users
Streams - patterns and SLAs

We will explore several of these common observability storyboards and give you the information necessary to get started today.

Session Speakers

Daniel Tomes

Databricks

Mohan Baabu

Solution Consultant

Databricks

Das Beste des Data+AI Summits anzeigen

Watch on demand