Re: Designing indexes

From: News <Contact_404_at_hotmail.com>
Date: 10 Feb 2006 01:39:06 -0800
Message-ID: <1139564346.347298.59060@g44g2000cwa.googlegroups.com>

hpuxrac a écrit :

> With so many tables in your environment, you need to prioritize somehow
> and start drilling into "only" the jobs/batch processes/extraction and
> load that are important enough to spend time and money improving.

I have traced all ETL sessions and got 5226 trc files. Now I want to identify consuming ones knowing by experience that a small number of queries consume most of resources. I need help with interpreting tkprof output to extract trc files/sessions/queries that need to be tuned from information below ?

call count cpu elapsed disk query current

    rows
------- ------ -------- ---------- ---------- ---------- ----------

---

Parse        1     14.87      14.61          0          0          0
       0
Execute      1      0.00       0.00          0          0          0
       0
Fetch        1      2.62       4.53       3848       3707          0
       0

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

---

total        3     17.49      19.14       3848       3707          0
       0

Is it helpful to parse output files for lines 'total' containing cpu and disk values that exceed certain limit ? Received on Fri Feb 10 2006 - 03:39:06 CST