Exadata Performance tuning

Base on execution plan take the following steps

  • Improved offloading/smart scan by eliminating unnecessary indexing (simply change index to be invisible and then check execution plan,…, repeat step)
  • check and find the reason for “quarantines for sql”
  • Eliminate any reason for “reading extra block” during full table scan, like “read consistency, row chaining & migration, …
  • Set properly table storage parameters:
    • The small table threshold (_small_table_threshold parameter) is determining factor for full scans. (setting/calculation depending on version and it is relative size of segment to buffer cache).  Note: partition is segment
    •   set properly parameters for parallel processing & PX Adaptive offloading
    • force direct path read & Bloom Filter;
    • set properly parameters for parallel processing & PX Adaptive offloading;

That’s it

Andjelko

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s