Oracle® Real Application Clusters Deployment and Performance Guide 10g Release 1 (10.1) Part Number B10768-02 |
|
|
View PDF |
This chapter explains how to use Enterprise Manager to monitor Oracle Real Application Clusters (RAC) database performance. This topics in this chapter are:
Overview of Oracle Enterprise Manager for Real Application Clusters
Enterprise Manager Performance Pages for Real Application Clusters
Service Relocation and High Availability Events
See Also: Oracle Enterprise Manager Concepts for information about Enterprise Manager and the Enterprise Manager online help for more information about using Enterprise Manager |
Enterprise Manager enables you to view current and past RAC database availability, performance metrics, and response times. Enterprise Manager can display this information at a high level for multiple cluster databases without requiring you to explicitly access each individual database.
Enterprise Manager supports monitoring at all levels of a cluster environment such as the cluster, the cluster database, and the cluster database instances. Enterprise Manager also responds to metrics from across the entire RAC database and publishes alerts when thresholds are exceeded. Enterprise Manager interprets both pre-defined or customized metrics. You can also copy customized metrics from one cluster database instance to another, or from one RAC database to another.
Enterprise Manager also uses data from the Automatic Workload Repository to display performance information and to initiate database alerts. Statistics that are collected by the Automatic Workload Repository are aggregated from all instances in a RAC database and displayed on a summary Enterprise Manager page.
Enterprise Manager provides performance information for:
Top Sessions—The most time- and resource-consuming sessions, waits, and SQL statements at the cluster database instance level.
Top Consumers—This includes Top Services, Top Modules, Top Actions by Module and Service, and Top Clients. This enables statistics aggregation, client identification support, and so on.
Database Locks—Includes client identifiers, service modules, and actions.
Cluster Cache Coherency—Performance information shows Global Cache Service (GCS) and Global Enqueue Service (GCS) activity
When you install RAC and create your RAC database with the Database Configuration Assistant (DBCA), the Enterprise Manager Database Control tools are pre-authorized to monitor your RAC environment. You can use Enterprise Manager Database Control to manage a single RAC database with its instance targets, listener targets, host targets, and cluster target. You can also install Enterprise Manager Grid Control onto servers either inside or outside your cluster and manage multiple RAC databases and multiple cluster targets.
Enterprise Manager displays both aggregate and instance-specific performance statistics using color-coded histograms. On any Enterprise Manager statistics display, highlight a statistic name on a statistics chart and Enterprise Managers re-displays the chart with colors for easier viewing. To obtain details about a highlighted statistic, click it and Enterprise Manager displays a detail page that provides more information about the statistic.
From the Cluster Database Home page you can access RAC-specific detail pages that show performance trends across one or more cluster databases. The following sections describe these pages in more detail:
The Cluster Performance Page displays usage statistics for all hosts or for individual hosts. This information enables you to add, suspend, or redistribute resources as needed.
The Cluster Database Performance Page displays charts showing run queue length, paging rate, service time, and database throughput for each host or instance. This page also provides access to detailed information such as the Details for Wait Class Page for Service Time and the Top Sessions Page for Database Throughput.
The Cluster Cache Coherency Page displays cache coherency metrics for the entire cluster database. This page groups metrics into the following categories:
Block Access Statistics
Global Cache Convert, Global Cache Current Block Request, and Global Cache CR Block Request
Top 5 Library Cache Lock and Top 5 Row Cache Lock
The Cluster Cache Coherency Page can display these statistics for each instance. This page also displays the Cache Coherency vs. Session Logical Reads chart. This chart graphs the percentage of Global Cache CR Blocks Received, Current Blocks Received, Converts, and Gets against logical reads for the session.
The Cluster Cache Coherency Instances Page provides real-time monitoring of global cache statistics. The Cluster Cache Coherency Instances Page displays tables of metrics from the following groups for all cluster instances such as:
Block Access Statistics
Global Cache Convert, Global Cache Current Block Request, Global Cache CR Block Request
Top 5 Library Cache Lock and Top 5 Row Cache Lock