SQL Server

Microsoft SQL Server 2017 powers your entire data estate by supporting structured and unstructured data sources. It builds on previous versions of SQL Server, which have been industry leading for four years in a row and a leader in TPC-E. It scales to petabytes of data and allows customers to process big data through PolyBase using T-SQL over any data. SQL Server has also been the least vulnerable database during the last seven years. SQL Server 2017 brings data insights with business intelligence capabilities that provide analytics at a fraction of the cost on any device along with advanced analytics with support for R and Python.

More details about SQL Server are available in the SQL Server documentation.
If you have a technical issue, please open a post on the developer forums through Stack Overflow or MSDN.

  • Hot ideas
  • Top ideas
  • New ideas
  • My feedback
  1. Data Collection, new method for "Collect and upload now for all"

    When you have added some data collection sets to the "Data Collection" functionallity i SSMS, you can right click and select "Collect and upload now".

    Can we have the possibility to do this one level up? For example right click on "System Data Collection Sets" and click on "Collect and upload for all"

    Now you have to do this on "set" at the time and click Close on the appearing afterwards.

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Suggestions  ·  Flag idea as inappropriate…  ·  Admin →

    Upvotes: 0

    <=-=Jun 8 2008 5:51PM=-=>

    Hi Arild,
    Thank you for your suggestion regarding a shortcut for doing a bulk collect and upload now. This is an iteresting idea to consider. I would suggest learning the T-SQL API and have a common script you would load up and run to upload as needed. FWIW, doing this for all of the system collection sets is not such a good idea. For example, there is little need to update the disk usage stats in coordination with the other collection sets.
    Thank you,
    Bill Ramos

    <=-=May 8 2009 3:14PM=-=>

    Thank you for your feedback – I have converted your request to work items for our next release and resolved this bug as a duplicate.

    Thanks,

    Amy Lewis

  2. Provide support for 3D Geo-Spatial Data

    The geo-spatial technology within SQL Server 2008 is awesome although for several applications I can imagine the additional dimention in 3D would be really useful.

    In today's TechEd session flight paths were used as examples and of course without support for 3D data there is no possibility of tracking altitude.

    I'd like to use the geo-spatial support in SQL Server for a particular application that can benefit from the initial 2D data that will be available in the RTM build but would encourage you guys to provide 3D support soon.

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Suggestions  ·  Flag idea as inappropriate…  ·  Admin →

    Upvotes: 8

    <=-=Jun 18 2008 11:55AM=-=>

    Hi Doug,

    3D support would be really great, wouldn’t it? We’re in the process of identifying high-level goals for the next release, and this is certainly one of the most requested features. Unfortunately, our plan at the moment is not to include any new feature functionality with the service packs so it looks like we’ll have to wait until at least the next release. That said, the more customer scenarios and requests we get for 3D, the better our case.

    In the meantime, have you taken a look at the Z elevation coordinate? The coordinate is not used in any library calculations, but depending on your application, this could be a good way to store altitude data. Here is a link to the BOL topic on this:

    http://msdn.microsoft.com/en-us/library/bb933920(SQL.100).aspx

    If there is anything else we can do to help, please let us know. And keep…

  3. Provide Spatial Results Control Outside SSMS

    Within the current build of SQL Server 2008 shown this week at Microsoft TechEd 2008 Developers Conference there is a 'Spatial Results' control that graphically renders the results of spatial data.

    Most users are not going to be using SSMS and therefore it would be awesome if the same control was available to be embedded within client applications. Ideally, there should also be Web controls for ASP.NET and/or Silverlight.

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Suggestions  ·  Flag idea as inappropriate…  ·  Admin →

    Upvotes: 3

    <=-=Jun 9 2008 10:09AM=-=>

    Hi Doug,

    Thank you for the suggestion. We will be looking to add the map functionality to SSRS in a future release

    <=-=Jun 9 2008 10:25AM=-=>

    Adding mapping features to SSRS would be great although it would also be necessary, I believe, to provide controls for ASP.NET, Windows Forms, WPF etc that are not tied to SSRS, to provide a feature complete solution.

    <=-=Dec 1 2008 10:30PM=-=>

    Hi Doug,

    We are currently working on bringing maps into SSRS. We tentatively are planning to have them in the next release of SQL Server codenamed “Kilimanjaro”. As far as providing a map control outside of SSRS, we are investigating our options, but there is no specific timeline yet.

  4. MSIT-MSO: Enable logging at stored procedure level

    Sometimes user complains that that some stored procedure is running slow. To troubleshoot this problem, we need to have idea about how much time the sp takes to run and what are the parameters with which it is called. This can be done by adding a explicit code in the stored procedure that adds the required logging. But this may not be possible always, as the stored procedure might be in production already and to do the change we need to go through whole of the DEv/Test/UAT process.

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Suggestions  ·  Flag idea as inappropriate…  ·  Admin →

    Upvotes: 0

    <=-=Jun 6 2008 9:18AM=-=>

    Hi Srikant,

    Thanks for providing with feedback on stored procedures and logging. We will certainly consider it for the next release of SQL Server. I was also wondering that have you tried using profiler and trace for this purpose and does it not help with what you are trying to achieve? You can collect diffrent events in profiler or trace and get the statistics around the user who invoked the procedure, time it took to complete etc.

    Thanks,
    Dipti.

  5. Last executed date/time on database objects

    Given a database running multiple application, changing staff, etc. it becomes difficult to know what database objects (views, procedures, functions, etc) are actually still used.

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Suggestions  ·  Flag idea as inappropriate…  ·  Admin →

    Upvotes: 0

    <=-=Jun 4 2008 7:31AM=-=>

    Josh,
    Thanks for your feedback. Updating the metadata for every execution can affect the performance significantly (extra IO). DMVs like sys.dm_exec_query_stats will give the current execution information. But, I can see the usefulness of this kind of functionality, if the DBA is willing to take the perf hit. Since we are pretty much done with SQL Server 2008 release, we will investigate this further for a future release of SQL Server.

    thanks
    Srini Acharya, SQL Engine

  6. SSIS Lookup: Duplicate values handling

    Currently, if duplicate values are found while building the lookup cache, only a warning message is generated in the execution log of the package (if enabled). The developer would need to add extra logic to the package in scenarios when duplicates values are in violation of the ETL/business logic, or when an specific instance of the duplicate row needs to be chosen when joining the incoming row with the looked up values.

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Suggestions  ·  Flag idea as inappropriate…  ·  Admin →

    Upvotes: 7

    <=-=Jun 13 2008 10:12AM=-=>

    Thank you for your feedback and support of the SSIS product. This issue doesn�t currently meet the requirements for correction in SQL 2008. We’re going to keep it open for triage in the next release.

    Again we appreciate your time and support.

  7. Make intellisense use custom code style

    Add ability to customize code style for functions please. For example I have to write (because of department's code style) @@rowCount instead of @@rowcount or @@ROWCOUNT or @@RowCount.
    Also for keywords.
    Many multipurpose editors have ability to do it.

    P.S. I have read feedbacks about lower and upper case customization of intellisense.

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Suggestions  ·  Flag idea as inappropriate…  ·  Admin →

    Upvotes: 8

    <=-=Jun 3 2008 3:49PM=-=>

    Alexey,

    Thank you for the feedback. We will start tracking this DCR as a high customer rated item.

    Regards,
    Eric

    <=-=Jun 22 2011 9:58AM=-=>

    Hi Alexey
    Thanks for taking the time to write in. We have added this item to our backlog of features for a future release. I’m resolving this as duplicate because we have several other similar requests. I’ve linked it to our work item in the backlog so we won’t lose your feedback.

    Cheers,
    -Sam Hughes

  8. Optimizer does not remove unnessecary DISTINCT in counjunction with ROW_COUNT()

    The following behaviour was already observed with SQL Server 2005 and I was expecting, it was fixed in 2008. Please look at the following example:

    use tempdb;
    create table T1
    (
    C1 varchar(80) not null default 'Empty'
    ,C2 uniqueidentifier not null default newid()
    )
    go
    -- Add some rows
    set nocount on
    insert T1(C1) values('A')
    go 5
    insert T1(C1) values('X')
    go 7
    insert T1(C1) values('Y')
    go 9
    insert T1(C1) values('Y')
    go 6
    Now run the following two queries and include the actual execution plan:
    select distinct
    top 10
    row_number() over(order by C1), C1
    from T1

    select top 10
    row_number() over(order…

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Suggestions  ·  Flag idea as inappropriate…  ·  Admin →

    Upvotes: 3

    <=-=Jun 9 2008 8:59AM=-=>

    Hello CaptainKirk!

    Thanks for reporting this issue. Because SQL Server 2008 is essentially finished, it’s too late to consider this for the 2008 release. But we’ll consider it for the next release. I see this as kind of a niche case. If you have a realistic customer scenario where this is causing you a problem, then I’d be interested to see that. It would improve the odds of us fixing this issue in the future.

    Best regards,
    Eric

    <=-=Jun 21 2010 10:40AM=-=>

    I’ve seen numerous cases where query performance would be significantly improved if the query optimizer’s domain assumption of ROW_NUMBER() without a PARTITION BY clause was .

    As a resource for real-world scenarios you requested, please refer to two techniques pulled from Itzik Ben-Gan’s book, "Inside SQL Server 2008: T-SQL Querying. These are run against SQL 2008 x64 SP1 CU8 on Windows…

  9. Reporting Services Date/Time Parameter to have a UTC option

    Currently, in SQL Server 2005 Reporting Services, when you enter a date/time parameter value on a report it is the locale specific time of the browser computer that gets passed to the underlying report data source.

    As databases and reporting solutions become more global in nature, the date/times being stored in a database are increasingly using UTC (Universal Time Code) format.

    So when the same report from multiple locales queries a data source based on a date/time range, either one of them will return incorrect information, or if the database uses UTC date/times. they both will return incorrect information (unless…

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Suggestions  ·  Flag idea as inappropriate…  ·  Admin →

    Upvotes: 3

    <=-=May 27 2008 3:56PM=-=>

    Thank you for your suggestion.
    We are considering looking at this more broadly (not just UTC, but actual client timezone, server timezone, and data source timezone – which could be three different zones) in a future release.

    Thanks,
    Robert Bruckner

  10. A linked dimension cannot be a reference dimension on top of another reference dimension

    I find it a big limitation that in a linked cube, a linked dimension cannot be a reference dimension on top of another reference dimension.

    The primary use case for a linked cube is to split out complex cubes. Well guess what? Complex cubes are the primary use case for reference dimensions.

    When attempting to set up this relationship in the Dimension Usage tab, it says "Chains of reference dimensions must be materialized". Ok, I can accept that, but when I check the Materialize checkbox it says "Linked dimensions that are reference dimensions cannot be materialized".

    Our organization is large…

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Suggestions  ·  Flag idea as inappropriate…  ·  Admin →

    Upvotes: 0

    <=-=May 27 2008 11:11AM=-=>

    Thank you Christian for reporting the problem.

    We are going to take a look at this limitation in the next version of Analysis Services.

    Thanks.
    The Analysis Services Team

    <=-=May 27 2008 7:06PM=-=>

    Great – I look forward to seeing what you come up with

    <=-=May 29 2008 9:28AM=-=>

    When you say you “are going to take a look at this limitation in the next version of Analysis Services”, do you mean SQL Server 2008?

    Christian

  11. Improve error message for table types and foreign keys

    When attempting to add a foriegn key to a table type, the error message returned should be more useful and explain that this is not possible. at the moment, it says the UDTT does not exist which is confusing. The errors look like this:

    Msg 4902, Level 16, State 1, Line 2
    Cannot find the object "udtt_RelationshipB" because it does not exist or you do not have permissions.
    Msg 4902, Level 16, State 1, Line 2
    Cannot find the object "udtt_RelationshipA" because it does not exist or you do not have permissions.

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Suggestions  ·  Flag idea as inappropriate…  ·  Admin →

    Upvotes: 0

    <=-=May 27 2008 5:18PM=-=>

    Dear Ben

    Thanks for your feedback and suggestion for improvement. I personally agree that the error message could be improved. However, since we are locking down on code changes for SQL Server 2008, we will have to look at this for the next release.

    Best regards
    Michael

  12. SSRS: Report Builder HiddenFields

    Sometime you want to display only one of several fields of an entity, the list "HiddenFields" does not allow to de/select all items, leading to the problem that you will have to click them all one by one.

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Suggestions  ·  Flag idea as inappropriate…  ·  Admin →
  13. Make MDX calculations work when the current coordinate is a set

    Currently MDX calculations do not work well, if at all, when the current coordinate contains a Set (with Sets in the WHERE clause).

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Suggestions  ·  Flag idea as inappropriate…  ·  Admin →
  14. BACKUP : Option to suppress info messages

    It would be nice to have a WITH option for the BACKUP command that allows us to suppress informational messages. If the backup works fine, there is no reason.

    The problem I'm specifically having is that I have a job which backs up 40 databases in one step, and when there are failures, because the SQL Agent log only keeps about 1024 characters per step, this means that unless an error happens on the very first database, I do not see the reason the step failed, since it is pushed down the stack by all the "10 percent complete" nonsense,…

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Suggestions  ·  Flag idea as inappropriate…  ·  Admin →

    Upvotes: 12

    <=-=Oct 22 2007 3:34PM=-=>

    I’m looking at the best method for accomplishing this:
    Would you see the best level to set this to be at the individual backup command (WITH NO_INFORMATIONAL)
    or at the database, or instance levels?
    If you had to choose ONE, which would it be?

    <=-=Oct 22 2007 4:16PM=-=>

    I’m looking at the best method for accomplishing this:
    Would you see the best level to set this to be at the individual backup command (WITH NO_INFORMATIONAL)
    or at the database, or instance levels?
    If you had to choose ONE, which would it be?

    <=-=Oct 23 2007 11:02AM=-=>

    Thanks for the comments!

    If I had to choose one, it would have to be WITH NO_INFORMATIONAL at the statement level. This is both because I want the most granular control possible and also because I think this has the least impact on the…

  15. SSRS: Report Builder: Sort and Group caption is misleading

    The sort and group button in the report builder is misleading. As many people will come from Access they will probably ask themselves where they can group things. The order thing is present, but they would suspect the group thing to be in the same UI.

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Suggestions  ·  Flag idea as inappropriate…  ·  Admin →
  16. SSRS: Report Builder: Parsing of Formulars could be extended

    The Parsing of the Formulas in the expression builder is not that chatty. E.g. the following expression where TextValue is of type string will bring an error DATEDIFF(YEAR,TextValue,TODAY()) will bring an error: THe arguments of the following functions are not valid. That is not quite much as this tool should be also used by a non-developement person. A developer would surely thing of the datetype not matching the signature, but isn�t there a way to present a richer error message than just "You passed something that was not expected" ?

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Suggestions  ·  Flag idea as inappropriate…  ·  Admin →
  17. SSRS Report Builder: Datetime Cast in Expression Builder

    There is no datetime cast inthe expression builder, although this is a common thing to do.

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Suggestions  ·  Flag idea as inappropriate…  ·  Admin →
  18. XML Parameters should have better rowcount estimate

    The lack of a rowcount estimate for XML parameters reduces their usefulness.

    At present, the .nodes() method would be useful but the 10,000 rowcount estimation tends to produce horrible plans when we only have a few rows.

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Suggestions  ·  Flag idea as inappropriate…  ·  Admin →

    Upvotes: 4

    <=-=Mar 29 2007 3:15PM=-=>

    Dear Greg.

    Thanks for your report and we appreciate the feedback. Cardinality estimation on general table-valued functions is a difficult problem, especially on parameters that may change from invocation to invocation and have no statistics associated with them.

    We are however looking into some possible improvements for one of the upcoming releases.

    Best regards
    Michael

    <=-=Apr 7 2011 8:11PM=-=>

    Dear Greg

    We are closing this item as a duplicate of http://connect.microsoft.com/SQLServer/feedback/details/453982/bizarre-xml-performance

    Best regards
    Michael

  19. Provide "WITH OVERWRITE" or a similar option for ALTER XML SCHEMA COLLECTION

    Altering an XML schema collection to make a simple change to the schema is a complete pain. For example, today a customer wanted to change one of the elements from a 50-character string to a 100-character string. The only semi-good way to handle this is to create a new schema collection, then go through and update every single reference that uses the old collection -- really a lot more work than necessary. SQL Server should be able to do this for me.

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Suggestions  ·  Flag idea as inappropriate…  ·  Admin →

    Upvotes: 15

    <=-=Mar 22 2007 6:50PM=-=>

    Dear Adam. Thanks for your feedback. We will look into providing such functionality in a future release of SQL Server and appreciate your feedback.

    Thanks and best regards
    Michael

    <=-=Mar 19 2009 8:13AM=-=>

    Existing SQL Server 2005 functionality is very limited. In my case I’m just trying to add a new optional attribute to a subordinate element node. I can somewhat understand the binding to the XML column, but having to unbind the stored procedure parameters! Come on! These are just references. BTW, if a variable within a stored procedure is bound to the schema, you’re OK — which is totally inconsistent.

  20. Support for implementing incremental mining models in CLR

    It would be nice to be able to develop (in managed code) plug-in algorithms with, or even have the current algorithms provided with SQL Server updated with, the capability to do incremental update of mining models. Several discussions on this:

    http://www.sqlserverdatamining.com/DMCommunity/Newsgroup/2390.aspx

    http://forums.microsoft.com/MSDN/ShowPost.aspx?PostID=767780&SiteID=1

    0 votes
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)

    We’ll send you updates on this idea

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

SQL Server

Categories

Feedback and Knowledge Base