DBAsupport.com Forums - Powered by vBulletin
Results 1 to 3 of 3

Thread: How can I reduce excessive 'db file sequential read' waits

  1. #1
    Join Date
    Mar 2001
    Posts
    149

    How can I reduce excessive 'db file sequential read' waits

    Hi all, we have a java program that would perform inserts to the db. Apparently, we noticed the inserts were deadly slow. When I perform a trace on this session, it shown that it spent tons of time for 'db file sequential read' waits. What can I do to improve this? Our table has only 1 unique key index. Thanks

  2. #2
    Join Date
    Mar 2006
    Location
    Charlotte, NC
    Posts
    865
    Usually more db file sequential reads are common in insert intensive application. However, if you observe more waits then you need to look into ...

    1. storage parameters (like next extent allocation, free lists etc) of the object
    2. please work with your storage admin to make sure that there is no contenstion in disk I/O.

    what storage you are using? It must be SAN. Isn't it?

    Generate AWR report during the time of insert and check for top 5 wait events and top queries based on elepsed time. Also, look into the objects with more physical reads / writes and corelate them with top queries and so on....

    Could you please post the below details?

    1. sga_target init parameter value
    2. db_writers parameter value
    3. Is the corresponding TS is locally managed or dictionary managed?

    and finally the information in the trace file that you have generated.

    Thanks,
    Vijay Tummala

    Try hard to get what you like OR you will be forced to like what you get.

  3. #3
    Join Date
    Mar 2007
    Location
    Ft. Lauderdale, FL
    Posts
    3,555
    Quote Originally Posted by newbie View Post
    When I perform a trace on this session, it shown that it spent tons of time for 'db file sequential read' waits. What can I do to improve this? Our table has only 1 unique key index.
    Are you aware that "db file sequential read" doesn't mean process is reading sequential but actually means "a read in sequence of an index access"? e.g. those reads are taking advantage of the unique key on that table - check explain plan.

    You can force query to not use that index by adding full() hint where you should see table access by "full table scan", a decrease of "db file sequential read" and a increase of "scatered read". Sometimes FTS is better, as a rule of thumbs when more than 15% of the rows have to be accessed.
    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.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  


Click Here to Expand Forum to Full Width