Many users work with Notebook?

[Cookistador]

Hello everyone,

I'm helping a friend to organize a small class with Fabric, but I have a small question about the notebook
The idea would be to take a small licence (like a F4), but the issue I see is when they will start to work on notebook
So to get the less expesive notebook, we should set up with something like:

The minimal nodes for a notebook is 2 (I remember to meet a lot of issues when you put this value to one)
Set the node size to small  (4 cores)
So it means, just to execute this notebook, it will required 8CU, so with a F4, he won't be able to execute more than one notebook for one user, is it correct ?

The idea is to build small notebook to show to their student how they can use it to make some transformation, do you see any alternatives ? I see the High concurrency mode but it look like it is only work for one users

Session sharing conditions include:

• Sessions should be within a single user boundary.
• Sessions should have the same default lakehouse configuration.
• Sessions should have the same Spark compute properties.

  How company with 10/15 people working with notebooks do ? They are staying in databrick instead of fabric?

  Many thanks for your help 🙂

[v-veshwara-msft]

Hi @Cookistador ,

We wanted to follow up on your query to see if you were able to resolve the issue. If so, marking the response as the solution and leaving a kudos would be greatly appreciated, as it helps others in the community facing similar challenges.

If you're still experiencing difficulties or need further clarification, please let us know.

Best Regards,
Vinay.

[v-veshwara-msft]

Hi @Cookistador ,

We just wanted to check in again regarding your issue. If you’ve found a solution, marking the reply as the solution and leaving a kudos would be greatly appreciated—it helps the community and others with similar questions.

If you’re still facing challenges or have further questions, please let us know—we’re here to help and would love to assist you in resolving this.

Looking forward to hearing back from you!

Best regards,
Vinay.

[v-veshwara-msft]

Hi @Cookistador ,
We haven't heard from you since last response and just wanted to check whether the solutions provided met your requirements. If yes, please accept as solution to help others benefit. If not, please reach out.
Thank you.

[v-veshwara-msft · Accepted solution]

Hi @Cookistador ,
Thanks for posting your query in Micrososft Fabric Community,
 You're right that setting up a notebook with 2 nodes (each with 4 cores) requires 8 CUs. With an F4 capacity, providing 4 CUs, would limit you to running one notebook at a time. If multiple users try to work simultaneously, you might run into performance issues.

Alternatives you might consider:

1. High Concurrency Mode: This mode is designed to handle multiple sessions, but there are some limitations. Sessions need to be within a single user boundary and share the same configuration.

2. Optimizing Notebooks: For small-scale demonstrations, try optimizing the notebooks to be less resource-intensive. This could mean simplifying tasks, reducing data sizes, or breaking down complex operations into smaller chunks.

3. Incremental Upgrades: If budget allows, and more resources are needed down the line, consider upgrading to a higher capacity incrementally.

4. Data Flows: You can also try DataFlows for transformations which is Low/No code ETL tool and reduce dependency on Spark-based notebooks.
5. Using NotebookUtils (MSSparkUtils): Utilize NotebookUtils for efficient data management and workflow optimization. It allows you to handle files, chain notebooks, manage environment variables, and retrieve secrets securely. This can streamline the teaching process, reduce complexity, and improve overall performance, even with limited capacity like F4.
   Please refer this link for further details.
   https://learn.microsoft.com/en-us/fabric/data-engineering/microsoft-spark-utilities#reference-run-mu... 

For companies with 10-15 people:

1. Larger organizations often go for higher capacities or scalable solutions to manage multiple users and notebooks efficiently.

2. Some may prefer solutions like Databricks if it better fits their workload. However, with careful planning and optimization, Microsoft Fabric can be a cost-effective and versatile choice.

I hope this helps! If you have any more questions or need further assistance, feel free to ask.

If this post helps please consider accepting as solution to help others benefit from this and a kudos would be appreciated.

Best regards
Vinay.