What s Inside This Book

What's Inside This Book

This book is specifically about the Oracle technologies that are provided with the Oracle RDBMS to help achieve a highly available database. We have grouped these technologies into four sections:

• General Availability

• Real Application Clusters

• Disaster Planning and Recovery

• Distributed Database Solutions

General Availability

General Availability refers to those features and techniques that encompass basic database configuration and administration: init.ora parameters, memory tuning, datafile storage, database object reorganization, and others immediately available to you in your existing Oracle Database 10 g database.

We begin in Chapter 1, 'Oracle and Availability: Illustrated Downtime Scenarios,' by going through a hypothetical series of downtime situations that happen to a fictional company that sells woodscrews. With each downtime scenario we offer a box that directs you to the technology that would assist the DBA for that particular problem. This is a way for us to show how the technologies discussed in Chapters 2-11 (the rest of the book) apply to very real uptime challenges. We also introduce the embarrassingly simple database tables that will be used throughout the book to illustrate examples of high availability database administration.

We embark on the journey into the world of dba-centric availability in Chapter 2, 'RDBMS Features for Availability.' Chapter 2 is dedicated to the Oracle options you can currently find hiding in the woodwork of your databases: dynamic parameter modifications, architectural availability options such as partitions and materialized views, logminer, and transportable tablespaces. We spend time discussing the benefits of Oracle's reworked Enterprise Manager suite, a one-stop interface meant to make your life easier.

Chapter 3, 'Tuning Your Database for Availability,' is a more in-depth look at tuning a database for maximum uptime. This includes a look at memory considerations for the database and utilizing the new Automatic Storage Management to get better use out of available disk space while minimizing the management of disk resources. We also spend time discussing online database performance tuning, including usage of the new Automatic Workload Repository-the offshoot of the existing Statspack software.

Real Application Clusters (RAC)

Part 2 is dedicated in its entirety to the setup, configuration, and administration of Real Application Clusters (RAC). The centerpiece of Oracle's availability offerings, RAC provides a clustering solution that found its footing in Oracle9 i and now mushrooms into the next generation of enterprise database computing in Oracle Database 10 g .

Chapter 4, 'RAC Setup and Configuration,' is the definitive guide to RAC Setup, starting with a beginner's guide to the basic architecture, as well as a hands-on lab for configuring a low-cost test cluster to allow you to get your hands dirty with minimal investment. It covers moving through the OS setup, including the setup of Oracle's Cluster File System, as well as raw files. The database configuration is coded step by step for quick RAC implementations , including the options for how to configure the archive logs.

Chapter 5, 'Database Administration in a RAC Environment,' delves deeper into the RAC stack, discussing the finer points of administration and upkeep unique to clustered databases. A closer look at RAC-specific redo and undo processing can be found here. This chapter talks to such availability requisites as rolling patch upgrades, dropping nodes, and adding nodes.

Chapter 6, 'Utility Computing: Applications as Services,' is dedicated to the implementation of services in a RAC environment-viewing applications as Services-a first step toward utility computing. The concepts of services are discussed along with the more important steps for configuration and rollout: services and work distribution.

Disaster Planning and Recovery

Part 3 deals with those technologies provided to minimize downtime due to unforeseen problems. This can be complete site loss due to natural disaster or power outage, or a smaller outage due to a burned-out hard drive or memory failure. Problems can also come in the form of user error: incorrect updates, logical application errors, or dropped tables. In the case of these types of problems, the goal must always be to have prepared successfully, and then have all the pieces in place to deal with different types of disasters appropriately.

Chapter 7, 'Oracle Data Guard: Surviving the Disaster,' is an extensive look at the configuration and administration of Oracle's most frequently overlooked feature: Data Guard. Provided as a complete disaster recovery solution, Data Guard provides a rich toolkit for using a database's existing architecture to mirror a complete database to another site. Combined with RAC and the new Flashback Recovery, Data Guard is a superior business continuity tool at times of total site loss or blackout .

Chapter 8, 'Backup and Recovery for High-Availability Environments,' takes us through the usage of Oracle's server managed recovery utility, RMAN, and how the usage of RMAN for media backups provides a necessary partner to RAC and Data Guard in the fight against down-time. Topics include a primer on RMAN configuration and usage, Enterprise Manager integration, new Oracle Database 10 g features such as the Flashback Recovery Area, and performing backups and recoveries when required. Advanced topics include the integration of RMAN into RAC clusters, and using RMAN to help build Data Guard environments.

Chapter 9, 'Oracle Flashback: Surviving User-Induced Trauma,' focuses entirely on the new suite of technologies that Oracle collectively refers to as 'Flashback Recovery.' This suite has been developed to provide minimal loss of time and data during those most dreaded of accidents: human error. Flashback Recovery comes in four related types: Flashback Query (as seen in Oracle9 i ), Flashback Transaction, Flashback Drop (for undoing a dropped object), and Flashback Database.

Distributed Database Solutions

In the brief Part 4, we discuss the benefits of exploring a distributed database solution to provide availability. In certain environments it often proves most useful to forego single-site availability and instead use a software solution to push tables to multiple sites, so that everyone sees the same table locally. This act of distributing database operations across a multitude of independent databases can provide a powerful availability solution with fringe benefits that match business requirements for independent site data.

Chapter 10, 'Oracle Streams for High Availability,' discusses Oracle Streams, a queue-based solution that provides a way to stream updates at independent sites to each other. Here we provide an overview, focusing on the use of Streams to provide a High Availability solution, as well as for Replication of Replica Databases.

Chapter 11, 'Oracle Net Configuration for Failover,' is a brief discussion of Network Configuration at the Oracle listener for various availability solutions. This includes setting up the listener for Transparent Application Failover (TAF), RAC, Data Guard, and Streams.

Similar products