How to %run a list of notebooks in Databricks

[Philblakeman]

I'd like to %run a list of notebooks from another Databricks notebook.

my_notebooks = ["./setup", "./do_the_main_thing", "./check_results"]
for notebook in my_notebooks:
   %run notebook

This doesn't work ofcourse. I don't want to use dbutils.notebook.run() as this creates new jobs and doesn't return anything back - I want everything executable and queryable from the main notebook.

I thought perhaps it might be possible to import the actual module and run the function.

?%run shows the command points to IPython/core/magics/execution.py and run is a method of the class ExecutionMagics in the module execution.

So perhaps, I could use execution.ExecutionMagic.run() if I created an instance of the class.

But it's beyond me - tricky and I'm doubting it's an effective solution.

How can this be done?

Am I really stuck with:-

%run ./a notebook
 
%run ./another_notebook
 
%run ./yet_another_hardcoded_notebook_name

Eternally grateful for any help!

[UmaMahesh1]

Hi @Philip Blakeman​ 

You can do this using scala or python constructs using threads and futures.

You can download the notebook archive from this link.

https://docs.databricks.com/notebooks/notebook-workflows.html#run-multiple-notebooks-concurrently

After that, based on your preference, set the number of parallel notebooks to be run using numNotebooksInParallel variable. If you just want 1 notebook at a time, you can do that too just by removing the unnecessary parts.

Be careful not to crash your driver by providing too many parallel notebooks.

Hope this helps.. Cheers.

[UmaMahesh1]

Hello @Philip Blakeman​ 

Checking if your issue got resolved or if you are facing any issues with the above approach..

Cheers..

[Philblakeman]

Hi

Thanks for the answer but it wasn't what I was after. It was about being able to provide a list as a variable.

I'll run the notebooks in series.

Basically can you %run {mynotebook}?

I think the answer is no.

https://stackoverflow.com/questions/74518979/how-to-run-a-list-of-notebooks-in-databricks

[Ajay-Pandey]

Please refer below code

import scala.concurrent.{Future, Await}
import scala.concurrent.duration._
import scala.util.control.NonFatal
 
case class NotebookData(path: String, timeout: Int, parameters: Map[String, String] = Map.empty[String, String])
 
def parallelNotebooks(notebooks: Seq[NotebookData]): Future[Seq[String]] = {
  import scala.concurrent.{Future, blocking, Await}
  import java.util.concurrent.Executors
  import scala.concurrent.ExecutionContext
  import com.databricks.WorkflowException
 
  val numNotebooksInParallel = 4 
  // If you create too many notebooks in parallel the driver may crash when you submit all of the jobs at once. 
  // This code limits the number of parallel notebooks.
  implicit val ec = ExecutionContext.fromExecutor(Executors.newFixedThreadPool(numNotebooksInParallel))
  val ctx = dbutils.notebook.getContext()
  
  Future.sequence(
    notebooks.map { notebook => 
      Future {
        dbutils.notebook.setContext(ctx)
        if (notebook.parameters.nonEmpty)
          dbutils.notebook.run(notebook.path, notebook.timeout, notebook.parameters)
        else
          dbutils.notebook.run(notebook.path, notebook.timeout)
      }
      .recover {
        case NonFatal(e) => s"ERROR: ${e.getMessage}"
      }
    }
  )
}
 
def parallelNotebook(notebook: NotebookData): Future[String] = {
  import scala.concurrent.{Future, blocking, Await}
  import java.util.concurrent.Executors
  import scala.concurrent.ExecutionContext.Implicits.global
  import com.databricks.WorkflowException
 
  val ctx = dbutils.notebook.getContext()
  // The simplest interface we can have but doesn't
  // have protection for submitting to many notebooks in parallel at once
  Future {
    dbutils.notebook.setContext(ctx)
    
    if (notebook.parameters.nonEmpty)
      dbutils.notebook.run(notebook.path, notebook.timeout, notebook.parameters)
    else
      dbutils.notebook.run(notebook.path, notebook.timeout)
    
  }
  .recover {
    case NonFatal(e) => s"ERROR: ${e.getMessage}"
  }
}