DBAsupport.com Forums - Powered by vBulletin
Results 1 to 5 of 5

Thread: How to move ahead in this direction??

  1. #1
    Join Date
    Oct 2000
    Posts
    449
    I need a direction from you guys..
    My company has some data intensive tables (30-50 million records) to start with. OLTP database. Data insertion through Java application.. (No sql loader)..
    Writes, I am not bothered now.. Aswell, no DB level tuning now.. But later..
    But Performance is a problem from Programmer's level..
    What should I do first?? I need some steps..
    Any good docs or any scripts available..
    Thx..

  2. #2
    Join Date
    Nov 2000
    Location
    greenwich.ct.us
    Posts
    9,092

    1. make sure they are using bound variables. You want to make sure oracle is not parsing each and every insert statement.

    2. Use array inserts. Insert a bunch of stuff at one time instead of multiple writes.

    3. Use sql*loader. You can bypass oracle's buffer cache and insert directly into tables.

    (I don't understand why people always try to re-invent the wheel. Spend your programmers time on application logic & GUI's.)
    Jeff Hunter

  3. #3
    Join Date
    Oct 2000
    Posts
    449
    Thanx Marist89.. All those points for writes..
    How about reads and writes while writes are stillgoing on..
    A typical OLTP environment though..

  4. #4
    Join Date
    May 2000
    Location
    ATLANTA, GA, USA
    Posts
    3,135
    RAID 1+0 is a good choice for a large volume of data.
    Distribute the data on many disks. Partition may help.
    Use B-Tree in OLTP environment.
    Cache the Ref.Tables data in Buffer Pool Keep.
    Configure good amount of SORT AREA SIZE.
    Configure a large amount for DB_BLOCK_BUFFER.
    Check the application Logic.
    Avoid Group by, and SORT in SQL statement.
    Verify the EXECUTION PLAN for Nested Loops.

  5. #5
    Join Date
    Oct 2000
    Posts
    449
    How do you use B-Tree?

    What is caching Ref Tbls in Buffer Pool keep

    How do you configure sort_area_size

    what is the alternate for sorting and group by..

    ------------------------------------------------------------

    comment on the below::

    SQL> show parameters buffer

    NAME TYPE VALUE
    ------------------------------------ ------- ------------------------------
    buffer_pool_keep string
    buffer_pool_recycle string
    db_block_buffers integer 7500
    log_buffer integer 163840
    use_indirect_data_buffers boolean FALSE
    SQL> show parameters sort

    NAME TYPE VALUE
    ------------------------------------ ------- ------------------------------
    nls_sort string
    sort_area_retained_size integer 131072
    sort_area_size integer 131072
    sort_multiblock_read_count integer 2

    Suggestions, welcome .. Thx.

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