Azure Cosmos DB

Have feedback for Azure Cosmos DB product? Submit your idea here or upvote other ideas. All feedback is monitored and reviewed by the Azure Cosmos DB team. 

This site is for feature suggestions only. For technical questions or issues, please submit them to StackOverflow,where we and the community can better help you.

Please use the following categories when submitting your idea.

SQL API: Query language features, syntax using SQL API, indexing, any other core features in Azure Cosmos DB.

Gremlin API: Graph features and capabilities using Gremlin API.

Cassandra API: Features and capabilities using Cassandra API.

MongoDB API: Features and capabilities using MongoDB API.

Table API: Features and capabilities using Table API.

Etcd API: Features related to using etcd as a configuration store for Kubernetes.

Azure Synapse Link: Features related to Azure Cosmos DBanalytical store and Azure Synapse Analytics run-time support

Built-in Notebooks: Features related to built-in Notebooks in Azure Cosmos DB.

SDK: Features related to Azure Cosmos DB SDKs for SQL API.

Change Feed: Features related to Change Feed.

Management: All management features, backup/restore,monitoring, ARM, PowerShell and CLI.

Portal: All features for Azure Portal and Cosmos DB Explorer.

Emulator: Features related to the Azure Cosmos Emulator.

Other: Features not related to any other category.

Security:
Authentication, authorization,permissions and encryption features.

Monitoring:
Metrics, monitoring, alerts,and diagnostics features.

Server-side: Stored procedures, Triggers,and User-Defined Functions.
  • Hot ideas
  • Top ideas
  • New ideas
  • My feedback
  1. Implement all of Gremlin's steps

    REALLY missing things like hasNext, tryNext, from, and to. It is really difficult getting around using these steps. Its hard to get around the fact that AWS already has a robust Gremlin offering that includes a more fully featured Gremlin implementation.

    1 vote
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    1 comment  ·  Gremlin API  ·  Flag idea as inappropriate…  ·  Admin →
  2. Use Order By with Group By

    In the documentation for group by (https://docs.microsoft.com/en-us/azure/cosmos-db/sql-query-group-by) you state:

    You currently cannot use GROUP BY with an ORDER BY clause but this is planned.

    Adding here so we can hopefully get some feedback on its progress and to communicate its importance to the community.

    In all my use cases for using group by, I am only interested in returning a subset of the results which without order by I cannot. I must return all results and then order & filter them in the client which is poor from a performance and cost perspective or completely unfeasible depending on…

    3 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  SQL API  ·  Flag idea as inappropriate…  ·  Admin →
  3. Ability to provide more than one PartitionKey per collection

    There are times when documents need to be queried by a secondary key field and the PartitionKey is not known.

    In order to avoid cross partition queries, the document needs to be stored twice - once with the primary PartitionKey and again with the secondary key field set as the PartitionKey.

    It'd be great if a secondary PartitionKey could be specified to allow for efficient querying without requiring saving the document multiple times.

    3 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  SQL API  ·  Flag idea as inappropriate…  ·  Admin →
  4. cosmos db metric not raising/recording http 401/403 response

    cosmos db metric not raising/recording http 401/403 response.
    It will be helpful if this is included as part of metrics

    1 vote
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    1 comment  ·  Monitoring  ·  Flag idea as inappropriate…  ·  Admin →
  5. Enable programmatic Jupyter notebook access

    I see that it's not possible to enable Jupyter Notebook support using the CLI, SDK, REST API, or an ARM template. It would be great to be able to enable Jupyter notebooks from code and even add a notebook.

    3 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  6. Azure portal Data Explorer works poorly with databases with both docdb and gremlin collections.

    We use a mix of gremlin and docdb collections in a gremlin api database (one per tenant). We initially started with 2 databases per tenant, one for gremlin and one for docdb, but this was just to much too manage, too expensive, etc. The only problem with this is the Data Explorer does not support it. It does not show documents for docdb collections so no way to add, delete, edit documents from Data Explorer. Instead we are using 3rd party tool CosmosDbExplorer that overcome this limitation, but it feels unnecessary complicated.

    1 vote
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Gremlin API  ·  Flag idea as inappropriate…  ·  Admin →
  7. Support Mongo Bulk Insert, InsertMany

    Please add InsertMany support. Saving more than 1000 documents its too slow with a simple Insert.

    3 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  MongoDB API  ·  Flag idea as inappropriate…  ·  Admin →
  8. orchestration for ETL processing for cosmos db gremlin API

    Dataware house concept - from on premise database to cosmos db germlin API to load data in to different stages into cosmos db germlin API.I didn't find orchestration component for this .Can you please suggest ?

    1 vote
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Gremlin API  ·  Flag idea as inappropriate…  ·  Admin →
  9. Improve refresh behavior in explorer UI

    In the emulator and storage explorer interfaces, browsing the contents of a container is made more difficult when checking newly-added items. If you have clicked "load more" multiple times and scrolled down the list, clicking "refresh" resets the scroll position and loses all the previously loaded items.

    It would be a nice quality of life improvement to make loading new items less disruptive. For example, "load more" should implicitly perform refresh for items created since last performed load query.

    3 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Management  ·  Flag idea as inappropriate…  ·  Admin →
  10. CreateIndex behavior is confusing

    The behavior of createIndex is confusing:

    db.foo.createIndex({bar: -1})

    creates a key with the opposite sort order:

    "key" : {

                        "bar" : 1
    
    },

    I understand the sort order doesn't matter with cosmos, but then calling createIndex({bar: -1}) again should not throw an error (see attached)

    3 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  MongoDB API  ·  Flag idea as inappropriate…  ·  Admin →
  11. Configure Backup Retention Periods

    Like Azure SQL, we would like the ability to specify in the portal the retention period for PIT backups (example: PITR for 7 day period) and long term retention (full backups weekly retained for 5 weeks).

    3 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Portal  ·  Flag idea as inappropriate…  ·  Admin →
  12. Enhance TTL explanation for Rest API

    I had a hard time finding the "time to live" attribute for the REST API.

    Turns out it is named "expireAfterSeconds".

    If you could expand the description to include "time to live" + "TTL" would be great since that is what is mentioned in pretty much every other SDK and documentation Will make it easier to find for others.

    https://docs.microsoft.com/en-us/rest/api/cosmos-db-resource-provider/mongodbresources/createupdatemongodbcollection#mongoindexoptions

    1 vote
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  SDK  ·  Flag idea as inappropriate…  ·  Admin →
  13. Improve "OFFSET LIMIT" RU.

    "OFFSET LIMIT" improves client performance for pagination.
    But it doesn't reduce RU.
    It's also important to improve back-end performance.

    1 vote
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  SQL API  ·  Flag idea as inappropriate…  ·  Admin →
  14. Increase maximum properties in a composite index.

    I have a SQL query that filters 11 properties.
    But, in this page, maximum properties in a composite index is 8.
    So, I can't add a index for this query.
    I'd like greater number.

    https://docs.microsoft.com/en-us/azure/cosmos-db/concepts-limits#sql-query-limits

    1 vote
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  SQL API  ·  Flag idea as inappropriate…  ·  Admin →
  15. Make autoscale work like other services' autoscale and pricing

    Autopilot in general cost 50% more per 100 RU than the dedicated throughput and is cost effective if there is significant variation from peek load usage. This is confusing as customer that auto pilot could be costing more in some sceanrio like if you only lower loads on weekends etc. And morever this is different from all the other services available in Azure where the Auto Scale Up/down does not have a different pricing then the regular mode. this cost should not vary with provisioned throughput ...

    8 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    1 comment  ·  Management  ·  Flag idea as inappropriate…  ·  Admin →
  16. Ability to refresh(copy) Cosmos DB from one instance to another

    Development teams often need data to be copied from one database instance to another e.g. copy all the data from a production database instance to a development/stage instance every other month or so. The current configuration does not allow an easy way to do this. It would be nice if you extend the functionality of the Azure Data Migration tool to take cosmos DB as source and another CosmosDB as a target to allow this. Though this can be achieved via the workaround of using mongodump which fails for bigger data loads due to the RU limits on cosmos Collections.

    3 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  MongoDB API  ·  Flag idea as inappropriate…  ·  Admin →
  17. Ability to create dedicated containers with 100 RU/s

    I need to create 30 containers, but it becomes too expensive when I need to allocate 400 RU/s per container at minimum. If I use a shared database, then I cannot individually convert containers to dedicated containers in the future as my budget becomes better. I'd really like to see the ability to create 100 RU/s dedicated containers so that I don't have to deal with messy migration.

    10 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    1 comment  ·  Other  ·  Flag idea as inappropriate…  ·  Admin →
  18. Support multiple labels on vertices and allow searching by one of those labels

    In order to represent inheritance in the graph DB, it is useful to have multiple labels on vertices and to be able to search by one of them, e.g.:

    V1 labels = "human | man"
    V2 labels = "human | woman"

    a query for all "human" returns both V1 and V2,
    whereas a query for all "man" returns V1 only.

    Currently, this is possible in Neo4j:
    https://tinkerpop.apache.org/docs/current/reference/#multilabel

    17 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Gremlin API  ·  Flag idea as inappropriate…  ·  Admin →
  19. Gremlin bulk executor for java library

    Gremlin bulk executor is available only for .NET library, please add support for java library.

    3 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Gremlin API  ·  Flag idea as inappropriate…  ·  Admin →
  20. Add Callback handler based on partition key, so polling is not needed/charged RUs


    1. ChangeFeedProcessorBuilder does not support subscribing only to certain PartitionKeys, so it'd be great if you supported that.


    2. If not, I can get by with GetChangeFeedIterator and manual polling (pull model), but that's in preview, so please release that officially.


    3. I believe even in #1, we are charged RUs for polling behind the scenes (correct me if I'm wrong.) So idle host instances are constantly consuming RUs. Is it possible to trigger Cosmos to call the registered callback handler only when something happens on the change feed --for the specified partition key. A trigger replaces polling. Then only charge RUs for…

    1 vote
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Change Feed  ·  Flag idea as inappropriate…  ·  Admin →
← Previous 1 3 4 5 11 12
  • Don't see your idea?

Feedback and Knowledge Base