Oracle Diagnostics Pack for Database
 Automatic Performance
 Real-Time Performance
 Automatic Workload Repository
Oracle Diagnostics Pack offers a comprehensive set of real time and automatic
performance diagnostics and monitoring functionality built into the core database
engine and Oracle Enterprise Manager. Whether you are managing one or many
databases, Oracle Diagnostics Pack offers a complete, cost effective, and easy to use
solution for managing the performance of your Oracle Database environment. When
used with Enterprise Manager, Oracle Diagnostics Pack additionally provides
enterprise-wide performance and availability reporting, a centralized performance
repository, and valuable cross-system performance aggregation, significantly
simplifying the task of managing large sets of databases.
 Comparing Performance Periods
 Active Session History
 Exadata Management
 Comprehensive System Monitoring
and Notification
 Automatic performance diagnostics
simplifies diagnosing performance
issues for administrators and
ensures quicker resolution of
performance bottlenecks.
 Automatically maintained workload
history facilitates historical
performance analysis.
 Comprehensive system monitoring
and event notification reduce
management cost and deliver
better quality of service.
 Ability to perform real time
performance analysis
 Enhanced ability to proactively
detect and identify the root cause of
performance issues
Automatic Performance Diagnostics
Diagnosing a slowly performing or a hung system is a time consuming task and often the activity
where database administrators (DBA) spend most of their time. A number of third party tuning tools
are available in the market but seldom do these tools provide an accurate root cause analysis. Instead
the DBA has to manually look through multiple charts trying to guess the root cause of the problem.
Oracle Diagnostics Pack takes the guesswork out of performance diagnostics. It includes a
performance-diagnostics engine built right into the Oracle database kernel, called the Automatic
Database Diagnostic Monitor (ADDM) that completely simplifies the complex and arduous task of
diagnosing performance problems for database and IT administrators.
ADDM starts its analysis by focusing on the activities that the database is spending most time on and
then drills down through a sophisticated problem classification tree to determine the root causes of
problems. ADDM’s ability to discover the actual cause behind performance problems, rather than
just reporting symptoms, is just one of the several factors that make it much superior to any other
Oracle database performance management tool or utility. Each ADDM finding has an associated
impact and benefit measure to enable prioritized handling of the most critical issues. To better
understand the impact of the findings over time, each finding has a descriptive name that allows the
application of filters and easy searching, and a link to the previous occurrences of the finding in the
last 24 hours. To enable performance diagnostics for pluggable databases (PDB) in Oracle database
12c, ADDM lists the affected PDB along with the details of the finding for quick and easy diagnosis.
For Oracle Real Application Cluster (RAC) environments, ADDM has a special mode for clusterwide performance analysis. It performs database-wide analysis of global resources, such as high-load
SQL, global cache interconnect traffic, network latency issues, skew in instance response times, I/O
capacity, etc.
Real-Time Performance Diagnostics
Diagnosing extremely slow databases or hung databases has been a big challenge for most database
administrators. With no way to connect to the hung database the administrator is often left with no
option but to bounce the entire system. This restart of the database not only causes an unplanned
outage but also gets rid of diagnostic information collected before the hung state. Without a proper
mechanism to find the root cause of the hang, the database application incurs the risk that the
problem may recur in the near future.
Real-Time ADDM provides an innovative way to analyze problems in unresponsive or hung
databases. Using a normal and a diagnostic mode connection Real-Time ADDM runs through a set
of predefined criteria to analyze the current performance and helps the DBA to resolve deadlocks,
hangs, shared pool contentions and many other exception situations that today forces the
administrator to bounce their databases, causing significant loss of revenue. Real-Time ADDM is the
only tool available in the market today that can log into a hung database, analyze the problem and
recommend a resolution.
Figure: 1 Real-Time ADDM analysis of hung database
Oracle Database 12c provides an enhanced version of the Real-Time ADDM capability. This
enhanced Real Time ADDM proactively detects transient performance issues by running in the
database automatically every 3 seconds. It uses in-memory performance data to diagnose any
performance spikes in CPU, memory, I/O etc. utilization. With this feature, Oracle Database 12c can
proactively inform an administrator a performance issue and its associated root cause even when the
system is not actively monitored.
Automatic Workload Repository
Oracle Diagnostics Pack includes a built in repository within Oracle Database, called Automatic
Workload Repository (AWR), which contains operational statistics captured at regular intervals
about that particular database and other relevant information. AWR is designed to be lightweight and
to automatically manage its use of storage space, ensuring that it does not put additional management
burden on administrators.
AWR forms the foundation for all the self-management functionality of Oracle Database. It is the
source of information that gives the database a historical perspective on how it is being used and
enables it to make decisions that are accurate and specifically tailored for the environment that
system is operating in. AWR also supports the creation of performance baselines. A moving
window baseline of 8 days is available out-of-the-box for helping compare performance to the
previous week and can be customized if needed. These AWR Baselines can then be used for
subsequent comparisons of current system performance to the baseline period to identify
performance divergences and their root-causes. The AWR report generated to analyze a period of
poor performance is really useful to look at the overall performance of the database and is the go-to
tool for most database administrators.
AWR in Oracle Database 12c also saves Real-Time SQL Monitoring, Database Operations
Monitoring and Real-Time ADDM reports inside the database, which allows the administrator to go
back in time and review a monitored execution of a query in the past. This is very useful in
determining performance inconsistencies across executions of a particular SQL query.
Active Session History
A key component of AWR is Active Session History or ASH. ASH samples the current state of all
active sessions every second and stores it in memory. The data collected in memory can be accessed
by a V$ view. This sampled data is also pushed into AWR every hour for the purposes of
performance diagnostics. Like AWR, ASH is also RAC-aware and the information obtained on
session activity in the cluster wait class gives visibility into potential RAC-specific issues. ASH has
also been extended to run on standby databases to assist in analysis of Oracle Data Guard
The in-memory ASH data can be used to understand the database workload profile and proactively
diagnose any transient performance issue that occurs for a very short duration. To enable proactive
diagnosis of database performance Oracle Enterprise Manager includes ASH Analytics, a new tool to
explore the ASH data that allows the administrator to rollup, drilldown, and slice or dice
performance data across various performance dimensions. With the ability to create filters on various
dimensions, identifying performance issues has never been easier. The built-in treemap view allows
administrators to explore performance data using predefined performance dimension hierarchies.
Figure:2 ASH Analytics view of performance
Comparing Performance Periods
Oracle Diagnostics Pack also provides a performance diagnostics capability called Compare Period
ADDM that allows the administrator to answer the age-old question of why the performance today is
slower than yesterday. The administrator can compare performance between two different time
periods by choosing from either an AWR baseline or the previous AWR snapshot period or any
calendar period of choice to check why a particular period is slower than the other. Compare Period
ADDM checks both the base and compare period and generates findings that pinpoint the root cause
for the difference in performance. Examples of the types of differences identified include the
commonality of SQL statements in the base versus compare periods, regression in query
performance due to higher utilization of system resources or a runaway ad-hoc query adversely
impacting normal transaction processing.
Exadata Management
Oracle Diagnostics Pack uses a holistic approach to manage the Exadata Database Machine and
provides comprehensive monitoring and management for the entire engineered system. It provides a
unified view of hardware and software where you can view hardware components such as compute
nodes, Exadata cells, and Infiniband switches and see the placement of software running on them
along with their resource utilization. DBAs can also drilldown from the database to the storage layer
of Exadata to identify and diagnose problems such as performance bottlenecks or hardware faults.
The lights-out monitoring capability of Enterprise Manager is optimized for Exadata where metrics
and thresholds are predefined so that administrators can get timely notifications when issues arise. In
Oracle Exadata Database Machine, management is engineered together with hardware and software
to provide not just high performance and availability but also ease of management and consolidation.
Comprehensive System Monitoring and Notification
Oracle Diagnostics Pack includes a comprehensive set of monitoring and notification features to
enable administrators to proactively detect and respond to IT problems across their entire application
stack. While Enterprise Manager continues to provide out-of-the-box monitoring for newly
discovered targets, administrators can customize these monitoring settings to fit their datacenter
needs. For database targets, this includes the use of adaptive thresholds which can automatically
alert on statistically unusual values of performance metrics based on the database's own performance
history. For other target types, easy access to a target's metric history is provided, enabling
administrators to determine appropriate threshold values based on the range of typical metric values.
If there are conditions specific to the datacenter those needs to be monitored, administrators can
define new metrics for any monitored target using metric extensions. If an alert has a well-known
remediation solution, then administrators can setup corrective action scripts that will automatically
execute and resolve the alert when it is detected, thereby minimizing the need for manual
intervention. In addition, alert history is also easily accessible to enable administrators to see what
actions have been taken in previous occurrences of the alert.
The desired monitoring settings for a target can be defined in a monitoring template, one template
per target type. When a set of monitoring templates for different target types are bundled together
into a template collection and associated with an administration group, then the deployment of
monitoring settings across targets is fully automated by Enterprise Manager. Specifically, when a
target is added to an administration group, the monitoring settings associated with the group are
automatically applied to the target, thereby streamlining and simplifying the process of monitoring
setup for targets.
Once monitoring is in place and events are detected and raised on monitored targets, notifications for
these events can be sent to the appropriate administrators. Notifications include email / page
notifications, the execution of custom scripts and PL/SQL procedures, and the sending of SNMP
traps. In addition, management connectors can also be used to open helpdesk tickets for incidents
(based on important events) and/or send event information to other third party management systems.
Finally, to support planned maintenance periods on targets, a blackout capability is provided to
enable administrators to temporarily suspend monitoring of targets and prevent false alerts from
being raised during the maintenance period.
Contact Us
For more information about [insert product name], visit or call +1.800.ORACLE1 to speak to an Oracle representative.
Copyright © 2013, Oracle and/or its affiliates. All rights reserved.
This document is provided for information purposes only, and the contents hereof are subject to change without notice. This document is not warranted to be error-free, nor subject
to any other warranties or conditions, whether expressed orally or implied in law, including implied warranties and conditions of merchantability or fitness for a particular purpose.
We specifically disclaim any liability with respect to this document, and no contractual obligations are formed either directly or indirectly by this document. This document may not
be reproduced or transmitted in any form or by any means, electronic or mechanical, for any purpose, without our prior written permission.
Oracle and Java are registered trademarks of Oracle and/or its affiliates. Other names may be trademarks of their respective owners.
Intel and Intel Xeon are trademarks or registered trademarks of Intel Corporation. All SPARC trademarks are used under license and are trademarks or registered trademarks of
SPARC International, Inc. AMD, Opteron, the AMD logo, and the AMD Opteron logo are trademarks or registered trademarks of Advanced Micro Devices. UNIX is a registered
trademark of The Open Group. 0113
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF