| New
tools for analyzing
SQL performance |
 |
Two new tools have been
recently introduced for analyzing Oracle SQL
performance: QBQL (pronounced "cubicle") and
SQL*Repository.
QBQL is an intuitive
graphical front-end to existing server-side
performance dictionary views, and SQL*Repository
is a powerful framework to automate the
comparison of execution plans and runtime
statistics for the same application across
Oracle Database releases.
SQL*Repository has
proven to be useful in migrating applications on
the Oracle Database from the Rule-Based to the
Cost-Based Optimizer
More
|
| |
|
Performance Tuning |
|
|
Oracle Database 10g
offers major changes in the way administrators
manage database:
- Operational data is automatically captured
and stored into the database by the
Automatic Workload Reporitory (AWR). The
data captured allows both system level and user
level analysis to be performed, reducing the
requirement to repeat the workload in order to
diagnose problems
- The Automatic Database Diagnostic Monitor
(ADDM),
a self-diagnosis engine that runs in the
database, automatically provides database-wide
performance diagnostic, analysis of root causes
and non-problem areas, and recommendations, with
impact and benefits analysis. The administrator
examines the latest ADDM report, which gives a
diagnosis and then either suggests treatment
itself or may refer users to other 10g advisory
components, such as the SQL tuning advisor
- The SQL tuning process has been automated by
introducing a new manageability feature called
the Automatic
SQL Tuning
- Automatic
memory management automates the management
of shared memory (SGA) used by an Oracle
Database instance and liberates administrators
from having to configure the shared memory
components manually. This functionality
significantly simplifies Oracle database
administration by introducing a more dynamic,
flexible and adaptive memory management scheme.
| |
