Update: Microsoft will be moving away from UserVoice sites on a product-by-product basis throughout the 2021 calendar year. We will leverage 1st party solutions for customer feedback. Learn more here.

Azure Databricks

Azure Databricks is an Apache Spark-based analytics platform optimized for the Microsoft Azure cloud services platform. Designed with the founders of Apache Spark, Databricks is integrated with Azure to provide one-click setup, streamlined workflows, and an interactive workspace that enables collaboration between data scientists, data engineers, and business analysts.

We would love to hear any feedback you have for Azure Databricks.
For more details about Azure Databricks, try our documentation page.

  1. The Simba ODBC driver needs to be available as a nuget package

    The Simba ODBC driver needs to be available as a nuget package

    So we dont have to install or setup something on the target systems, and can retain "xcopy deployments"

    1 vote

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Strong Feedback  ·  Flag idea as inappropriate…  ·  Admin →
  2. This uservoice does not accept the MSFT account I've had for more than 20 years.

    This uservoice does not accept the MSFT account I've had and used for more than 20 years. Other MSFT uservoices do not have that restriction.

    Instead it wants to use a work account, something that I wont have access to or be able to follow up with if I changed employers (which happens in this industry more often than others). At that point, I cant respond, track, or followup on anything because of this.

    Using work accounts also permits users to play limited voting games and vote more than once for the same idea under different logins.

    1 vote

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Strong Feedback  ·  Flag idea as inappropriate…  ·  Admin →
  3. The documentation is absent for any kind of ad-hoc querying

    The documentation is absent for any kind of ad-hoc querying

    Is there really no way to execute ad-hoc queries (written either by hand or programmatically)?

    1 vote

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Strong Feedback  ·  Flag idea as inappropriate…  ·  Admin →
  4. Better rendering and more cell support

    Can a product be designed in a more freaking irritating way? During my script execution, a command that should take atleast an hour sometimes randomly will say "Command took 1 second". What the hell? Okay so did it run? Nooo, as when i reference the variable, it says "Not Found!"
    What the hell is happening with rendering? To solve i have detach and re-attach my notebook because as you see i had spent the whole running the slow as Databricks notebook to get to this particular point. Now, lo and behold, all my variables have been reset, my half a…

    1 vote

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Strong Feedback  ·  Flag idea as inappropriate…  ·  Admin →
  5. 1 vote

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Strong Feedback  ·  Flag idea as inappropriate…  ·  Admin →
  6. Export Databricks dashboard as pdf

    Please provide an option to export the dashboard created in databricks as PDF. Currently there is no way to export the data even in html or PDF.
    Customer wants the dashboard data as pdf for easy sharing

    1 vote

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Strong Feedback  ·  Flag idea as inappropriate…  ·  Admin →
  7. Workaround for Invalid column name 'IS_REWRITE_ENABLED' Databricks External Apache Hive metastore from HDI.

    https://stackoverflow.com/questions/63782822/setting-up-azure-sql-external-metastore-for-azure-databricks-invalid-column-na

    This is because the hive schema metastore of DBR 7.3 (Spark 3.0.1) with current hive-schema for the derby db includes the column 'ISREWRITEENABLED ' while the SPARK version 2.3.2 on HDInsight 3.6 (default HDI version) does not include it, so when trying to access it, this discrepancy is created resulting in the error.

    As an workaround to this it is necessary to add the column 'ISREWRITEENABLED ' to the TBLS table.

    Resolution:

    Create the missed column on the table TBLS from SQL server.
    ALTER TABLE TBLS
    ADD ISREWRITEENABLED char(1) NOT NULL DEFAULT 'N';

    1 vote

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Strong Feedback  ·  Flag idea as inappropriate…  ·  Admin →
  8. Plot option

    can't apply Plot option.

    1 vote

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Strong Feedback  ·  Flag idea as inappropriate…  ·  Admin →
  9. runtime

    Need email alerts about deprecated runtime versions/etc.

    My team has been moving all processing over to Azure Databricks for about the last year and we've been deploying CI/CD via Powershell API calls.

    We've been using runtime 6.5.x for probably the whole time.
    for some reason this particular version was deprecated without warning.
    I found out because our deployments started to fail and then saw the icon in the UI.

    While this didn't break Production because existing clusters are maintained for some time, it broke our CI/CD and lead to a scramble to upgrade our runtime and test our code/libraries in…

    1 vote

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Strong Feedback  ·  Flag idea as inappropriate…  ·  Admin →
  10. Can plot a multiple char out for single display (data frame) and also host the output to the running cluster

    Can plot a multiple char out for single display (data frame)

    If we get this feature that we can plot multiple charts in a single then it will help many organizations and teams build their run time dashboard just for the analysis purpose.

    Also if at the same time you can provide the feature we can host this dashboard to any of the available clusters it would be a blast.

    1 vote

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Strong Feedback  ·  Flag idea as inappropriate…  ·  Admin →
  11. Data Pipeline Feature

    We should have a data pipeline feature in data bricks, where we can build the execution flow of our notebook instead of using and running it through the Data factory or %run notebook.

    It will solve many small use cases, where company wants to build small pipeline.

    1 vote

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Strong Feedback  ·  Flag idea as inappropriate…  ·  Admin →
  12. Ray intergration with databricks

    It would be nice to have Ray integrated with Databricks so you can run python code executed on a cluster.

    Already asked it here :
    https://forums.databricks.com/questions/45772/is-there-step-by-step-guide-how-to-setup-ray-clust.html

    Git :https://github.com/ray-project/ray

    1 vote

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Strong Feedback  ·  Flag idea as inappropriate…  ·  Admin →
  13. support RAPIDS Plugin with Multiple GPU Nodes

    I tested NVIDIA Rapids Plugin with NC6s_v3 as mentioned in the documentation. https://nvidia.github.io/spark-rapids/docs/get-started/getting-started-with-rapids-accelerator-on-databricks.html It worked well.

    But, when I used NC12sv3 or NC24sv3 to create my cluster, this is not working. By the way I am using Azure Databricks with 7.0 ML DBR.

    Here is the feedback from the NVIDA team that databricks does not support creating executors, which limits the use of this plugin.

    "we don't support nodes with multiple GPUs on Databricks right now. The plugin has a restriction that each executor only has 1 GPU and it seems like the last time I tried on…

    1 vote

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Strong Feedback  ·  Flag idea as inappropriate…  ·  Admin →
  14. Azure Databricks Light Runtime

    The latest Databricks Doc (https://docs.microsoft.com/en-us/azure/databricks/runtime/light#:~:text=Databricks%20Light%20is%20the%20Databricks,benefits%20provided%20by%20Databricks%20Runtime.) indicates that Databricks Light runtime is supported.

    However, when creating a cluster, I was not able to find the light runtime version from the drop-down window.

    Really suggest to have a better UI window so both internal technical experts and end-customers will know what runtime version they are choosing when creating a cluster.

    1 vote

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Strong Feedback  ·  Flag idea as inappropriate…  ·  Admin →
  15. Is Sparkdl outdated?

    The documentation gives an impression to use Sparkdl, which is not being used in the current frameworks or distributions. If Azure The web page mentions that Databricks release alludes to the older versions (2.*), I think it would not be easy for the developers to bring their code developed on Spark3.0 to Azure Databricks. Therefore, I would request to enable some kind of code compatibility so that the Spark users and Deep Learning practitioners running different versions can flawlessly run their apps on MS Azure's platform. Just my two cents.

    1 vote

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Strong Feedback  ·  Flag idea as inappropriate…  ·  Admin →
  16. Event based job triggers for orchestration

    Currently databricks supports only cron based schedules for triggering spark jobs (using jar). It will be great if we can have a event based trigger to start and orchestrate the jobs in same workspace.
    If we have 5 spark jobs and i want to run them based on schedule, but in cases where the schedule overlaps, i had to disable the schedule of the other job. Instead if we can say start job 2 on success of job1 and so on, it will be helpful.

    This feature is available for notebooks using NotebookWorkflows.Expecting similar feature for spark jobs as well.

    1 vote

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Strong Feedback  ·  Flag idea as inappropriate…  ·  Admin →
  17. Achieve feature parity between service principals and user accounts

    Azure encourages the use of service principals to link together services, and not user accounts. Unfortunately, Databricks does not have feature parity between service principals and user accounts.

    Some examples:
    1. Service principals do not appear under users in the workspace, but do show via the RESTAPI
    2. Service principals cannot have secret scopes
    3. Service principals cannot be added or removed from the workspace over the RESTAPI or the UI.
    4. Service principals must be members of the admins group, and if you remove them, the next time you connect over the RESTAPI it simply makes them a member…

    1 vote

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Strong Feedback  ·  Flag idea as inappropriate…  ·  Admin →
  18. folders in jobs portal

    Can we have folders in jobs portal where we can categories the similar kind of jobs

    1 vote

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Strong Feedback  ·  Flag idea as inappropriate…  ·  Admin →
  19. Vnet peering

    To Azure Databricks Team:

    Scenario:
    I have an existing ADB workspace (without VNet) and want to connect with Azure Storage account with Firewall.

    In this situation, we don't have any workaround, rather then creating new ADB workspace with VNet Injection. And to allow this Vnet on Storage firewall.

    Note: Vnet Peering is not working in this scenario.

    Suggestion:
    Can you please file a future request, in order to accommodate above scenario.

    1 vote

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Strong Feedback  ·  Flag idea as inappropriate…  ·  Admin →
  20. Avoid re-execution of query to export complete results.

    Currently, default query execution returns 1000 records and to export full results it re-executes the query again. Looking for functionality something like limit function to provide the number of output records and same can be returned in first time execution.

    1 vote

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Strong Feedback  ·  Flag idea as inappropriate…  ·  Admin →
  • Don't see your idea?

Azure Databricks

Categories

Feedback and Knowledge Base