Data Typing Quandry

  • UMG Developer (11/17/2010)


    mtassin (11/17/2010)


    I agree. I got this wrong for the same reason. The Row Size doesn't change, compression just lets more data be stored on a page.

    But in this case compression actually makes it so less data fits on a page because of the metadata overhead.

    How does that change that the questioned asked about the "defined row size" not the "storage row size"?

    These are differant concepts and alters the answer completely.

  • Using SQL 2008 if one creates 4 tables (1 for each of the create statements in the question). Insert the same number of rows into each table, and then executes the procedure:

    sp_estimate_data_compression_savings 'dbo', 'Flags1', NULL, NULL, 'ROW' ;

    The answer does NOT support the "correct answer", will the questions author please explain why the difference

    If everything seems to be going well, you have obviously overlooked something.

    Ron

    Please help us, help you -before posting a question please read[/url]
    Before posting a performance problem please read[/url]

  • SanDroid (11/17/2010)


    UMG Developer (11/17/2010)


    mtassin (11/17/2010)


    I agree. I got this wrong for the same reason. The Row Size doesn't change, compression just lets more data be stored on a page.

    But in this case compression actually makes it so less data fits on a page because of the metadata overhead.

    How does that change that the questioned asked about the "defined row size" not the "storage row size"?

    These are differant concepts and alters the answer completely.

    I couldn't find any definition for "defined row size", do you have a link to a definition? Does it include metadata?

  • UMG Developer (11/17/2010)


    I couldn't find any definition for "defined row size", do you have a link to a definition? Does it include metadata?

    Exactly my point. The term in the question is "defined row size".

    Not storage row size or maximum row size.

    Compression can allow more rows to be stored on a page, but does not change the maximum row size of a table or index.

    http://msdn.microsoft.com/en-us/library/cc280449.aspx

    Also I have noticed that, as has been pointed out by others, the storage concepts in the questions explination may not factual when executed.

  • SanDroid (11/17/2010)


    UMG Developer (11/17/2010)


    I couldn't find any definition for "defined row size", do you have a link to a definition? Does it include metadata?

    Exactly my point. The term in the question is "defined row size".

    Not storage row size or maximum row size.

    Compression can allow more rows to be stored on a page, but does not change the maximum row size of a table or index.

    http://msdn.microsoft.com/en-us/library/cc280449.aspx

    Also I have noticed that, as has been pointed out by others, the storage concepts in the questions explination may not factual when executed.

    Yes, I agree with this. Compression has an effect on the storage but does not change the defined row size.

  • Cliff Jones (11/17/2010)


    Yes, I agree with this. Compression has an effect on the storage but does not change the defined row size.

    But what is the definition for the "defined row size"? Does it include metadata? (Is it a real SQL Server term, or is it just made up?)

  • UMG Developer (11/17/2010)


    Cliff Jones (11/17/2010)


    Yes, I agree with this. Compression has an effect on the storage but does not change the defined row size.

    But what is the definition for the "defined row size"? Does it include metadata? (Is it a real SQL Server term, or is it just made up?)

    Good point. I could not find a reference to that term so I suspect it is made up so therefore the ambiguity. I started to answer correctly, until I noticed that the question was referring to the "defined row size".

  • Cliff Jones (11/17/2010)


    UMG Developer (11/17/2010)


    Cliff Jones (11/17/2010)


    Yes, I agree with this. Compression has an effect on the storage but does not change the defined row size.

    But what is the definition for the "defined row size"? Does it include metadata? (Is it a real SQL Server term, or is it just made up?)

    Good point. I could not find a reference to that term so I suspect it is made up so therefore the ambiguity. I started to answer correctly, until I noticed that the question was referring to the "defined row size".

    Since, as far as I know, this is not a reserved SQL Server term, I decided to go wild and employ my English language parser.

    In my opinion, "defined row size" should refer to the size of the row as defined. And if I look at the table definition, I can't help but notice the "With (Data_Compression = Row)"

    So please stop this babbling about a perceived ambiguity, the question is perfectly clear.

    All regulars know that I won't hesitate te criticise a question that I feel does not make the bar. This is no such question.


    Hugo Kornelis, SQL Server/Data Platform MVP (2006-2016)
    Visit my SQL Server blog: https://sqlserverfast.com/blog/
    SQL Server Execution Plan Reference: https://sqlserverfast.com/epr/

  • Hugo Kornelis (11/17/2010)


    Cliff Jones (11/17/2010)


    UMG Developer (11/17/2010)


    Cliff Jones (11/17/2010)


    Yes, I agree with this. Compression has an effect on the storage but does not change the defined row size.

    But what is the definition for the "defined row size"? Does it include metadata? (Is it a real SQL Server term, or is it just made up?)

    Good point. I could not find a reference to that term so I suspect it is made up so therefore the ambiguity. I started to answer correctly, until I noticed that the question was referring to the "defined row size".

    Since, as far as I know, this is not a reserved SQL Server term, I decided to go wild and employ my English language parser.

    In my opinion, "defined row size" should refer to the size of the row as defined. And if I look at the table definition, I can't help but notice the "With (Data_Compression = Row)"

    So please stop this babbling about a perceived ambiguity, the question is perfectly clear.

    All regulars know that I won't hesitate te criticise a question that I feel does not make the bar. This is no such question.

    Thanks, I trust your judgment.

  • Hugo Kornelis (11/17/2010)


    So please stop this babbling about a perceived ambiguity, the question is perfectly clear.

    All regulars know that I won't hesitate te criticise a question that I feel does not make the bar. This is no such question.

    Hugo,

    We would prefer your respect by not calling our criticism babbling.

    The with clause in the Table create statements has no relation to if the question is about the Maximum Row size or the Storage Size of the row when stored in a data page.

    Some agree with your view, but most that posted about the question do not.

  • very nice question... thanks 🙂

  • Excellent question, I learned something new.

    /Håkan Winther
    MCITP:Database Developer 2008
    MCTS: SQL Server 2008, Implementation and Maintenance
    MCSE: Data Platform

  • SanDroid (11/17/2010)


    Hugo Kornelis (11/17/2010)


    So please stop this babbling about a perceived ambiguity, the question is perfectly clear.

    All regulars know that I won't hesitate te criticise a question that I feel does not make the bar. This is no such question.

    Hugo,

    We would prefer your respect by not calling our criticism babbling.

    We would prefer your respect for the effort that the author has made to create a question so that people could learn about compression, instead of calling it "dissapointing and lame".

    As I've learned something from this question, I find it an excellent question and I thank the author for his contribution.

    Need an answer? No, you need a question
    My blog at https://sqlkover.com.
    MCSE Business Intelligence - Microsoft Data Platform MVP

  • da-zero (11/18/2010)


    SanDroid (11/17/2010)


    We would prefer your respect for the effort that the author has made to create a question so that people could learn about compression, instead of calling it "dissapointing and lame".

    If you look at some of these posts several people asked the author to respond to the fact that they where not able to replicate what he described.

    Why where no responses were made?

    Did you learn something that was real?

    Did you test the concept?

    I made an opinionated response about the question and it's content and that is On Topic discussion.

    IMHO: Attacking those opinions and calling them Babble and putting Off Topic mis information in the same paragraph helps nobody learn and clarifies nothing.

  • SanDroid (11/18/2010)


    If you look at some of these posts several people asked the author to respond to the fact that they where not able to replicate what he described.

    I can't replicate showing that it actually uses more space, but not having a real definition for "defined row size" doesn't help. (Along with my lack of knowledge of internal storage.)

    But the BOL link appears to support the question with these quotes: "In some cases, the metadata overhead might be larger than the old storage format." and "bit: The metadata overhead brings this to 4 bits."

Viewing 15 posts - 16 through 30 (of 36 total)

You must be logged in to reply to this topic. Login to reply