Tuesday May 28, 2013

AWR Data Extract

The Oracle RDBMS has an excellent repository of performance data that is collected automatically. It's called Automatic Workload Repository, or AWR for short. The AWR is a snapshot-based collector of database information. This includes not only the information from V$SYSSTAT, but also snapshots of other configuration information like database parameter values. This information is invaluable in diagnosing issues at the database-level. AWR information is typically exchanged via the AWR reports in HTML format (created with awrrpt.sql). While this can be useful, what would be better--and more flexible--is to have the raw data used to create the HTML report.[Read More]

Tuesday May 07, 2013

Exporting Schema Statistics

While most of us are familiar with the schema statistics used by the Cost-Based Optimizer (CBO), something not so well know is the ability to export/import these statistics using the DBMS_STATS package. This can be an invaluable aid in diagnosing query plan differences as these statistics are the primary information used by CBO. It is also a way to save and restore statistics in your own Primavera database.


[Read More]

Friday May 03, 2013

Optimizer Cost Parameters

There is nothing about query optimization that is ever black or white; just many, many shades of grey. This is the case with the two index-related database parameters: optimizer_index_cost_adj and optimizer_index_caching. The one thing I can say with confidence is that the default values are not ideal for Primavera P6 OLTP. If these parameters are set to the default values, chances are you will see some high Physical I/O (PIO) in the database.[Read More]
About

An insider view of the technology behind the Primavera product suite.

Search

Categories
Archives
« May 2013 »
SunMonTueWedThuFriSat
   
1
2
4
5
6
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
29
30
31
 
       
Today