|Oracle® Database High Availability Overview
12c Release 1 (12.1)
|PDF · Mobi · ePub|
Oracle Grid Computing, Oracle Active Data Guard real time reporting and utilization, Oracle Database Consolidation using Pluggable Database or Oracle Virtualization and Oracle Global Data Services can all optimize return on investment (ROI) of any of the high availability architectures and solutions.
Oracle Real Application Clusters (Oracle RAC) is the foundation of the Server Grid Computing while Oracle Automatic Storage Management (ASM) and Exadata are the foundation for Storage Grid Computing. Oracle Active Data Guard continues to be strategic for comprehensive data protection, availability and disaster recovery and can provide highly favorable return in investment by offloading reporting, and backups, and doing tests and planned maintenance activities. You can further leverage existing system resources effectively by using database consolidation techniques such as Oracle Pluggable Databases or Oracle Virtualization. Furthermore distributed databases can be used as a collective to maximize all the resources with Global Data Services.
This chapter covers the following topics:
Grid computing is a computing architecture that effectively pools large numbers of servers and storage into a flexible, on-demand computing resource for all enterprise computing needs.
The Oracle Database captures the cost advantages of Grid enterprise computing without sacrificing performance, scalability, security, manageability, functionality, or system availability.
A Database Server Grid is a collection of commodity servers connected to run one or more databases.
A Database Storage Grid is a collection of low-cost modular storage arrays combined together and accessed by the servers in the Database Server Grid.
The same grid computing concept applies to primary as well as standby database evnironments. Figure 9-1 illustrates the Database Server Grid and Database Storage Grid in a grid enterprise computing environment.
The availability of low-cost and reliable blade servers, small multiprocessor servers, and open-source operating systems such as Linux, have made it possible to build a Database Server Grid that is highly available, scalable, flexible, and manageable.
Oracle Real Application Clusters is the technology that enables a Database Server Grid. You can drive down costs by deploying a single Oracle RAC database that spans multiple low-cost servers, each running an active Oracle database instance. Alternatively, you can use a single cluster to minimize the management and increase system utilization across multiple Oracle RAC Databases.
Oracle RAC provides the flexibility to dynamically provision resources and services in the grid as computing needs change, and to add or subtract systems from the grid as capacity demands change. In addition, Oracle RAC provides protection from system failures by automatically transitioning clients and redistributing the processing of the failed node to surviving nodes running the same Oracle RAC database. Note that the scalability and availability benefits of grid computing are not limited to lower cost servers. Any system architecture will benefit from grid computing.
The availability of low-cost Advanced Technology Attachment (ATA) disk-based storage arrays and low-cost storage networks has made it possible to use a Database Storage Grid with Oracle Database at a very low cost. One example solution is Oracle Exadata Storage Servers that offer excellent performance and availability characteristics. Each Exadata Cell can be viewed as a unit of I/O performance and capacity.
The Oracle Storage Grid is implemented using either Oracle Automatic Storage Management (Oracle ASM) and Oracle Exadata Storage Server Software or Pillar SAN Storage Systems or Oracle ASM and third-party storage. The Oracle Storage Grid with Exadata seamlessly supports MAA-related technology, improves performance, provides I/O scalability, is easy to use and manage, and delivers mission-critical availability and reliability to your enterprise.
A database administrator can use the Oracle ASM interface to specify the disks in the Database Storage Grid that Oracle ASM can manage across all server and storage platforms. Oracle ASM partitions the disk space and evenly distributes the data storage throughout the entire storage array. Additionally, Oracle ASM automatically redistributes the data as disks or storage arrays are added or removed from the Database Storage Grid.
Additionally, use I/O Resource Management to manage and meet service-level requirements. The resource manager allows you manage the grid and prioritized applications within the database or in between databases.
Data Guard standby databases are an integral part of the Grid, providing data protection, availability and disaster recovery regardless of the cause or scope of an outage. Outages can range anywhere from data corruption that can affect an individual database, to natural disasters that impact a large geographic area.
Advanced Data Guard capabilities deliver maximum ROI by enabling standby databases to be used for productive purposes-such as for read-only queries and reporting-while running in the standby role. Rather than allowing standby databases to remain idle, you can employ them to support activities that would otherwise require you to purchase additional capacity for other systems. Thus, you can defer or eliminate the need to purchase additional capacity for the primary database. This effectively reduces the cost of providing world-class disaster protection for mission critical Oracle Databases.
The following sections describe the Data Guard scenarios that provide high business utilization and a maximum return in investment:
Data Guard Redo Apply (physical standby database) is a popular solution for disaster recovery due to its relative simplicity, high performance, and superior level of data protection. The Oracle Active Data Guard optionFoot 1 (available with Oracle Database 11g Release 1 (11.1) and later releases) enables a physical standby database to be opened for read-only access while Redo Apply is active. Offload capabilities of Oracle Active Data Guard 12c have been enhanced to include: read-only reporting and ad-hoc queries including DML to global temporary tables and unique global or session sequences, data extracts, fast incremental backups, redo transport compression, efficient servicing of multiple remote destinations and the ability to extend zero data loss protection to a remote standby database without impacting primary database performance. Furthermore standby database can help mitigate downtime and risk for planned maintenance activities by using Data Guard standby-first patch apply for patching and transient logical standby for upgrades.
You can use multiple physical standby databases (using the Oracle Active Data Guard option) to deploy an Oracle Active Data Guard reader farm. An example of such a configuration is provided Figure 9-2, complete with the use of Data Guard fast-start failover to automatically fail over should the primary database fail. Note that all standby databases in the reader farm automatically recognize the new primary database after a failover occurs.
A reader farm enables an application to scale read performance of the most demanding web applications beyond what the underlying system and storage architecture can support. This provides a relatively low-cost method of scaling out using a highly redundant Oracle Active Data Guard reader farm architecture where you simply satisfy your increased reporting requirements by adding additional Oracle Active Data Guard standby databases.
The concept is a single primary database that supports read/write transactions, and multiple standby databases that provide read-only access to data. Such an approach scales read performance linearly as additional standby databases are added. It is also an effective way to isolate faults or planned maintenance such as standby-first patching, because problems that affect one standby database are isolated from the other standby databases in the configuration. If there's a standby database failure or standby system is offline for maintenance, Oracle 12c Global Data Services can be used to transparently failover to the existing standby systems or to the primary.
Creating a reader farm of physical standby databases provides the following benefits:
High performance with physical standby databases and Redo Apply
Seamless support for all DDL and data types using Redo Apply
All reader databases are kept up-to-date with changes made to the primary database
Automatic, zero or minimal data loss failover capability
Management as a unified configuration through Grid Control
Scale-out using single writer database and n reader databases
Rolling upgrade capabilities
Integrated client failover to production database or other standby databases using Global Data Services
Figure 9-2 shows a good example of how you can use Data Guard, physical standby databases, and Oracle Active Data Guard option to provide the flexibility necessary to grow your business quickly, while still providing disaster recovery. In the configuration, the primary database transmits redo data to multiple standby databases, one of which is also enabled for fast-start failover for automatic, zero, or minimal data loss failover.
If a fast-start failover is triggered in the Data Guard configuration in Figure 9-2, then:
Automatic failover occurs to the designated standby database
All standby databases accept data from the new primary database
You can perform a switchover at a convenient time in the future to return all databases to their original roles
With Database Server Grid and Database Storage Grid (described in Section 9.1.1 and Section 9.1.2), you can build standby database and testing hubs that use a pool of system resources. The system resources can be dynamically allocated and deallocated depending on various priorities. For example, if the primary database fails over to one of the standby databases in the Data Guard hub, the new primary database acquires more system and storage resources while the testing resources may be temporarily starved. With the Oracle Grid technologies, you can enable a high level of usage and low TCO without sacrificing business requirements.
A Data Guard hub can consist of:
Several standby databases in an Oracle RAC environment residing in a cluster of servers, called a grid server
Using the storage grid
The premise of the Data Guard hub is that it provides higher utilization with lower cost. The probability of failing over all databases at the same time is unlikely. Thus, when a failover occurs, you can prioritize the system resources to production activity and allocate new system resources in a grid for the standby database functions. At the time of role transition, more storage and system resources can be allocated toward that application.
For example, a Data Guard hub could include multiple databases and applications that are supported in a grid server and storage architecture. This configuration consists of a central resource supporting 10 applications and databases in the grid, rather than managing 10 separate system or storage units in a nongrid infrastructure.
Another possible configuration might be a testing hub consisting of snapshot standby databases. With the snapshot standby database hub, you can use the combined storage and server resources of a grid instead of building and managing individual servers for each application.
Database consolidation provides the ability to host multiple applications or databases on the same system platform or within the same database. For the most part, the HA architectures and solutions are still applicable. The benefits of consolidation is the reduce cost of ownership and management while leveraging all resources as effectively as possible; however the trade-off is a reduced level of isolation and independence compared to having separate database or system resources for each application and database.
Multitenant architecture is the capability that enables an Oracle database to contain a portable set of schemas, objects, and related structures that appears logically to an application as a separate database. This self-contained collection is called a pluggable database (PDB). A multitenant container database (CDB) contains PDBs.
Multitenant architecture is the most cost-effective form of database consolidation. By consolidating multiple physical databases on separate computers into a single database on an optimized engineered platform such as Exadata Database Machine, you gain the following benefits:
Cost reduction for hardware
Portability of an application's database back end
Ease of database and system administration
Centralized management of database accounts and privileges
Easier and faster upgrade paths
Multitenant architecture is especially useful when you have many databases deployed on different hardware in multiple Oracle Database installations. Each PDB might use only a fraction of the hardware resources dedicated to it, and each PDB might not require a full-time database administrator to manage it.
By combining these databases into a CDB, you can make better use of your hardware resources and database administrator resources. In addition, you can move PDBs from one CDB to another without requiring changes to the applications that depend on the PDB.
Multitenant architecture can leverage the proposed HA architectures in this book. For example, the targeted PDB can reside on Exadata Database Machine leveraging Oracle Real Application Clusters, Oracle Automatic Storage Management, and Exadata Storage Cells and have an additional Standby PDB residing on a separate Exadata Database Machine.
Oracle Database Administrator's Guide for information about creating and administering pluggable databases.
Data centers today use virtualization techniques to make abstraction of the physical hardware, create large aggregated pools of logical resources consisting of CPUs, memory, disks, file storage, applications, networking, and offer those resources to users or customers in the form of agile, scalable, consolidated virtual machines. Even though the technology and use cases have evolved, the core meaning of virtualization remains the same: to enable a computing environment to run multiple independent systems at the same time with the main intent of saving people and hardware resources.
Oracle has three main virtualization technologies:
Oracle VM for X86 and Oracle VM Manager are an enterprise-class server virtualization solution. Oracle VM Server for x86 is the most scalable x86 server virtualization solution in the market today, and it has been tested to handle mission critical enterprise workloads with support for up to 160 physical CPUs and 2 TB of memory. For virtual machines, Oracle VM 3 can support up to 128 virtual CPUs and 1TB memory per guest VM. Oracle VM supports industry standard x86 operating systems and servers from Oracle and other leading vendors, and it supports a broad range of network and storage devices, making it easy to integrate into your environment. Oracle VM Manager provides an easy-use-centralized management environment for configuring and operating your server, network, and storage infrastructure from a browser based interface (no Java client required), and it is accessible from just about anywhere.
Oracle VM Server for SPARC provides highly efficient, enterprise-class virtualization capabilities for Oracle's SPARC T-Series servers. Using the Oracle VM Server for SPARC software, you can create up to 128 virtual servers, called logical domains, on a single system. This kind of configuration enables you to take advantage of the massive thread scale offered by SPARC T-Series servers and the Oracle Solaris OS.
Oracle Solaris Zones software partitioning technology, which provides a means of virtualizing operating system services to create an isolated environment for running applications. This isolation prevents processes that are running in one zone from monitoring or affecting processes running in other zones. Zones can be used on any machine that is running the Oracle Solaris 10 or a later Oracle Solaris release. The upper limit for the number of zones on a system is 8192.
Oracle virtualization can be used in conjunction with HA features and HA architectures to reap the benefits of both target goals. Here are some of the HA benefits when integrating Oracle virtualization with HA architecture and features.
Auto restart of VMs in the event of a failure making applications HA
Oracle Real Application Clusters ensure business availability at the application layer and is integrated with Oracle VM to ensure business availability on the server as well as application data in a single or multiple geographic locations
Generally any Oracle high availability feature, such as RMAN, flashback technologies, Data Guard, and Oracle GoldenGate, that works natively in non-virtualized environments will work seamlessly in a virtualized environment.
Oracle VM accelerates the delivery of services to meet changing business need. This allows online growing of capacity
The Oracle VM website on OTN at
Oracle Solaris 11.1 Administration: Oracle Solaris Zones, Oracle Solaris 10 Zones, and Resource Management, Part II, "Oracle Solaris Zones" at
A Global Data Services (GDS) configuration is a set of databases integrated by the GDS framework into a single virtual server that offers one or more global services, while ensuring high performance, availability and optimal utilization of resources. GDS manages these virtual resources with minimum administration overhead, and allows the GDS configuration to quickly scale to handle additional client requests.
The databases that constitute the GDS configuration can be globally distributed or located within the same data center. Clients can securely connect to the gds configuration by simply specifying a service name, without needing to know anything about the components and topology of the GDS configuration, enabling a highly flexible private cloud deployment for the enterprise.
GDS enables customers to scale their disparate computing resources and heterogeneous platforms in a highly flexible way, without requiring any application changes. Geographically dispersed data centers, whether regional or global, can now be effectively utilized within a uniform framework based on business, throughput, and localized demands, without affecting run-time applications. This additional computing scale will enable a truly elastic and agile enterprise and extend the benefits of cloud computing to all employees, business partners and stakeholders.
Footnote LegendFootnote 1: Oracle Active Data Guard is referred to as real-time query in the Data Guard documentation.