One database, copied to two identical servers. Performance bad on second server. Why?

  • Hi All,

    I'm not a DBA and I'm confused at the cause of the behavior I'm seeing. I have two identical servers (cloud servers setup from the same image) and one large database. I have an app which is running a data production process on both machines. The process truncates some tables and then imports data from a large flat file. I'm seeing that the performance of the process is not the same or approximately the same as I would expect.

    Why would this be? Originally, I created the database on Server A and then copied it to Server B. Is there anything in Server A's SqlServer that would give it more information about the db that Server B doesn't have an thus Server B's SqlServer shows worse performance?

    Thanks,

    Paul

  • Paul Koanui (11/1/2011)


    I'm not a DBA and I'm confused at the cause of the behavior I'm seeing. I have two identical servers (cloud servers setup from the same image) and one large database. I have an app which is running a data production process on both machines. The process truncates some tables and then imports data from a large flat file. I'm seeing that the performance of the process is not the same or approximately the same as I would expect.

    Why would this be? Originally, I created the database on Server A and then copied it to Server B. Is there anything in Server A's SqlServer that would give it more information about the db that Server B doesn't have an thus Server B's SqlServer shows worse performance?

    Assuming hardware, O/S and RDBMS setups are exactly the same and also assuming data is the same...

    1- Identify the process which performance differs the most.

    2- Compare server A vs. server B execution plans.

    3- If #2 are identincal then trace process on both servers and compare.

    _____________________________________
    Pablo (Paul) Berzukov

    Author of Understanding Database Administration available at Amazon and other bookstores.

    Disclaimer: Advice is provided to the best of my knowledge but no implicit or explicit warranties are provided. Since the advisor explicitly encourages testing any and all suggestions on a test non-production environment advisor should not held liable or responsible for any actions taken based on the given advice.
  • if there are different server versions, for example ServerA is SQL2005 but ServerB is SQL2008, you'll want to rebuild the indexes as well as rebuild the statistics for non-index columns; there are a lot of posts that show the updated query engine is adversely affected by stats that originate from a converted database.

    Lowell


    --help us help you! If you post a question, make sure you include a CREATE TABLE... statement and INSERT INTO... statement into that table to give the volunteers here representative data. with your description of the problem, we can provide a tested, verifiable solution to your question! asking the question the right way gets you a tested answer the fastest way possible!

  • I'd question whether the servers are the same. Could be storage is better or utilized less on one than the other. Could be that network path is different/slower.

Viewing 4 posts - 1 through 3 (of 3 total)

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