cancel
Showing results forย 
Search instead forย 
Did you mean:ย 
Administration & Architecture
Explore discussions on Databricks administration, deployment strategies, and architectural best practices. Connect with administrators and architects to optimize your Databricks environment for performance, scalability, and security.
cancel
Showing results forย 
Search instead forย 
Did you mean:ย 

Custom Runtime marketplace

evgenyvainerman
New Contributor

Hi! Is there a possibility to share the solution accelerator on the custom runtime via the databricks marketplace?

1 REPLY 1

Louis_Frolio
Databricks Employee
Databricks Employee

Greetings @evgenyvainerman , sorry for the delayed response. Your question is not entirely clear but I will take a swing at providing an answer.

 

Short answer: Yes, you can share a Solution Accelerator through Databricks Marketplace, but Marketplace does not distribute or enforce a custom Databricks Runtime; accelerators are delivered as Git repositories and run on the consumerโ€™s compute in their workspace. Consumers follow your instructions to use the recommended runtime or environment when they execute the notebooks.
 

What Marketplace supports for Solution Accelerators

  • Solution Accelerators are a supported asset type in Marketplace, alongside datasets, notebooks, and ML models.
  • Delivery mechanism: Solution Accelerators are shared as clonable Git repositories. Consumers click โ€œGet instant access,โ€ and the repo is added to their Databricks Git folders; they then run the notebooks on clusters in their own workspace.
  • Runtime scope: Marketplace docs enumerate supported asset types (datasets, notebooks, Solution Accelerators, ML models), not runtime images. Accelerators therefore do not ship a custom Databricks Runtime; runtime selection happens in the consumerโ€™s workspace when they choose or create clusters.

Practical ways to handle โ€œcustom runtimeโ€ needs

If by โ€œcustom runtimeโ€ you mean required packages, specific DBR versions, or initialization steps:
  • Include a clear โ€œEnvironment setupโ€ section in the accelerator repo (for example, a requirements file, init scripts, and a recommended DBR version), and reference those steps in the listing description so consumers run the accelerator on compatible compute.
  • Provide a sample cluster configuration (JSON or UI instructions) with a pinned DBR version and necessary libraries so consumers can create the right cluster before running notebooks.
If you truly need to package an application-like experience that encapsulates dependencies, note that Marketplace also surfaces โ€œapplications/solutionsโ€ as a category, which may better align with distributing a more self-contained experience than a pure repo listing.
 

How a consumer uses your accelerator (flow)

  • The consumer finds your Solution Accelerator listing, clicks โ€œGet instant access,โ€ and a Git repo is cloned into their workspaceโ€™s Git folders.
  • They open the notebooks and run them on a cluster using your recommended DBR/version and setup steps from the repoโ€™s README or listing notes.
Hope this helps, Louis.