Oracle Database 11g Highly Available Grid made easy with Oracle - - PowerPoint PPT Presentation

<Insert Picture Here>

Oracle Database 11g Highly Available Grid made easy with Oracle Enterprise Manager

Venkat Maddali, Director of Development, Oracle Jagan Athreya, Director of Product Management, Oracle

The following is intended to outline our general product direction. It is intended for information purposes only, and may not be incorporated into any

• contract. It is not a commitment to deliver any

material, code, or functionality, and should not be relied upon in making purchasing decisions. The development, release, and timing of any features or functionality described for Oracle’s products remains at the sole discretion of Oracle.

Oracle Highly Available Grid strategy

• Scale-Out architecture
• Commodity hardware building blocks
• Inherently highly scalable & redundant
• Scalability & Availability responsibility moves out
• f hardware/OS to scale-out savvy software
• First Web & Application server tiers
• Application servers
• Then DB tier
• Shared disk and shared nothing databases
• Then storage tier
• Scale-out storage software

Application Database Storage

Agenda

• Oracle’s Highly Available Grid in 10g
• Highly available Grid In 11g
• Setup
• Manage
• Customer perspective on Highly available Grid

Getting on to Oracle’s HA Grid – 10g recap

Single instance database Convert file system to ASM Convert SI to RAC database Create Standby database

Convert Single Instance to RAC db

• Wizard-driven interface to convert

single instance into multi-instance RAC database

• Various pre-req checks prior to

submitting the operation

• Automates all necessary steps
• Move database files/recovery area

files to shared storage (if necessary)

• Create additional undo TS, redo

logs, password files,

• ratab/registry entries etc.
• Configure listeners
• Register with cluster-ware
• Pre-requisites:
• All nodes must be running Oracle

Clusterware (CRS)

Single Instance RAC

Migrate from filesystem to ASM

• Simplify management of

Oracle database files

• Avoid single point of storage

device failure.

• Migrate database, and

recovery files to ASM diskgroups and switch

• Migration with with minimal

downtime

RAC Management

• Manage “many as one”
• RAC/Cluster Topology views
• Cluster-aware monitoring
• Cluster and RAC db metrics to

avoid duplicate monitoring

• Load average across the cluster
• Interconnect monitoring
• Global Cache Coherency
• Identify top node(s)/ instance(s)

involved in block transfers

• Identify top segments with

extreme concurrency profile across some nodes

• Monitor Clusterware
• Cluster-aware jobs
• Configure and Manage HA

Services.

Agenda

• Oracle’s Highly Available Grid in 10g
• Highly available Grid in 11g
• Setup
• Manage
• Customer perspective on Highly available Grid

Highly Available Grid in 11g

• Setup: Provision RAC, ASM and HA configurations to

implement Highly Available Grid

• Manage: Extend “Manage many as one” to
• Provide Service Centric monitoring
• Diagnose as many

Getting on to Oracle’s HA Grid – 11g

Single instance database Convert FS to ASM Provision Clusterware and Oracle software Convert SI ASM to Clustered ASM Convert SI db to RAC database

Getting on to Oracle’s HA Grid – 11g

Create Standby database Convert Standby DB to RAC standby

New Feature: Convert SI ASM to Clustered ASM

• Simple wizard to convert single

instance ASM to Clustered ASM

• Automated as an EM job
• Pre-requisite to migrating the

Single Instance Database to RAC Database on ASM storage

Demo

New Feature: Convert SI ASM to Clustered ASM

• Pre-reqs prior to performing cluster-wide operations
• Cluster verification
• Node accessibility
• Oracle Database Software on all selected nodes
• Ensure that ASM is not in use by any database that is up
• Automates
• Setup of parameter files across cluster hosts
• Listener configuration across all cluster hosts
• Mount shared diskgroups on all nodes
• Register CRS resources for ASM
• Update oratab/registry
• Enable monitoring by enterprise manager
• Restore SI ASM back if something were to fail, so retry can be

attempted after fixing the root cause (Similar to SI-RAC DB conversion)

New Feature: Setup Standby Database Grid

Challenge

• Deliver same High Availability and

Scalability as Primary Production Database Solution

• Create Standby RAC database

from Primary RAC database Benefit:

• Automated creation of standby

RAC database

• Uninterrupted scalability and

availability as primary database

Primary Database Standby Database

Standby Site Primary Site Broker

New Feature: Manage Standby Database Grid

• Manage Clustered Standby databases same as Single instance
• High Availablility for Fast-Start Failover Observer
• Grid Control will automatically restart the Data Guard Observer on an

alternative host if the primary Observer host fails

• Complete support for 11.1 DB features such as Snapshot standby,

Active Data Guard, etc.

• EM Management integrated with DR operations on role changes

Demo

Agenda

• Oracle’s Highly Available Grid in 10g
• Highly available Grid
• Setup
• Manage
• Customer perspective on Highly available Grid

Managing Highly Available Grid in 11g

• Menu based navigation
• HA Console to Manage

RAC/HA environment

• Fast notifications on

RAC DB/Instance Availability

• ADDM for RAC
• Service centric

monitoring

• Manage as one,

Diagnose as many

HA Console

• Configuration wide view for RAC/HA

setup instead of individual targets

• Summary/Chart view for HA Services
• Show Availability related events
• HA Configuration standards
• Show key HA information
• Launch HA Operations in context

Demo

ADDM for RAC

• A Performance Expert, now a

RAC Specialist too!

• Identifies the most “Globally

Significant” performance issues

• Automatically runs every hour
• Cluster-wide analysis of:
• Global cache interconnect issues
• Lock manager congestion issues
• Global resource contention, e.g.

IO bandwidth, hot blocks

• Globally high-load SQL
• Skew in instance response times

Self-Diagnostic Engine

Database-Level ADDM Instance-Level ADDMs

Node 1 Node 2 Node 3

Service Centric monitoring

• Setup HA Services with Preferred/Available configurations and

carry-out required operations

• Test connectivity to service
• Create tns alias files for services as well as ability to export this

to client machines.

• Summary view of health of RAC services at RAC home page as

well as HA Console

Service Centric monitoring

• Drilldown to show all HA Services with configuration, status, service

response time and CPU utilization metrics

Service Centric monitoring

• Workflows to show Service Activity by Waits and Instances

drilldowns to identify SQL/Session

Service Centric monitoring

• Drilldowns to identify SQL/Session

Manage as one, Diagnose as many

• Augment Summary/roll-up views with Tile charts to view selected metric across

all nodes/instances of the cluster

Manage as one, Diagnose as many

• Provide cluster-wide views of metrics using Tile charts
• Cluster metrics (CPU, Memory and Disk I/O) across cluster hosts
• Host Load Average

Manage as one, Diagnose as many

Manage as one, diagnose as many

• Provide cluster-wide views of metrics using Tile charts
• Wait metrics
• Cache Coherency metrics (e.g. Blocks transferred, etc.)

Manage as one, Diagnose as many

• Customizable tile charts
• Number of charts per row
• Order by
• Average over 1 hr
• Maximum over 1 hr
• Instance ID
• User defined

Demo

Agenda

• Oracle’s Highly Available Grid in 10g
• Highly available Grid
• Setup
• Manage
• Customer Perspective on Highly Available Grid

<Insert Picture Here>

HA management in Oracle Database 11g using Oracle Enterprise Manager

Julian Dontcheff, Senior DBA, Nokia

Availability

Enterprise Manager 11g Database Control

Availability

Enterprise Manager 11g Database Control

• Browser-Based Enterprise Manager Integrated

Interface for LogMiner

• Enhanced Oracle RAC Monitoring and Diagnostics in

Enterprise Manager

Availability

Enterprise Manager 11g Database Control

Availability

Enterprise Manager 11g Database Control

• Browser-Based Enterprise Manager Integrated

Interface for LogMiner

Availability

Enterprise Manager 11g Database Control

Browser-Based Enterprise Manager Integrated Interface for LogMiner

Availability

Data Guard

Availability

Data Guard

• Use OEM to create the standby database
• Use RMAN to duplicate the database for standby

Oracle Highly Available Grid

• Automate migration to

Storage Grid

• Automated migration to

Highly available Database Grid

• Management of Highly

Available Oracle grid

• Highly available

monitoring

Application Database Storage

Management Console Enterprise Enterprise Manager Manager