Using Oracle GoldenGate for Siebel Zero Downtime Upgrades

Oracle already had a comprehensive offering for Data Integration which helped move data in bulk across heterogeneous sources and targets. With the best-in-class ELT/ETL technology of ODI-EE and the best-in-class data cleansing solution: Oracle Data Profiling and Data Quality, these solutions could be integrated together with Service Oriented Architecture approaches to build re-usable data services.

Oracle GoldenGate now completes the picture with the addition of real-time change data capture and replication for high availability.

What are the technical advantages that customers can see when applying this best practice?

Highest Performance in the industry – when it comes to Non-intrusive capture of changed data, Low-latency end-to-end data movement as well as the optimized transformation. All of these are critical for end-to-end improved performance for data warehousing.

Flexibility: Heterogeneous sources and targets – applicable for Oracle Data Integrator with the Knowledge Module framework, and applicable for Oracle GoldenGate which can implement changed data capture across multiple sources in real-time

In addition, flexibility of the different architecture styles which Rob discussed, including Real-time BI/DW, SOA, data federation, and even data distribution, synchronization, consolidation which are now becoming new styles of data integration.

Lastly, the reliability of these integration movements – which is what sets Oracle apart – with support for both transaction integrity of the changed data as well as how it manages bi-directional data replication which is critical for active-active scenarios.

Oracle GoldenGate is a very multipurpose technology as you can see here. It can be deployed for feeding analytical systems with real time change data but also offer continuous data availability by synchronizing standby systems or new hardware/db/os environments companies want to migrate to. It can synchronize data across distrubuted systems in different continents for real-time data sharing and has the ability to publish changed data to JMS based messaging systems with its non invasive log based change data capture

You can capture once from the source system and move to multiple targets for different needs without impacting the performance of the source system.

Companies can standardize on Oracle GoldenGate for any real-time data movement needs. With its fast and easy deployment and low impact on the IT infrastrcuture it certainly brings savings in cost of ownership

GoldenGate captures, routes, transforms, and delivers transactional data in real time– and it works across heterogeneous environments with very low impact and preserved transaction integrity. So the key tenets and differentiators of our technology are that we provide data in real-time even with large data volumes. It works in hetereogenous environments AND maintains the commit boundaries of each transaction during the movement for transaction integrity.

Subsecond latency even with thousands of transactions per second
Heterogeneity and modular architecture for flexible configurations to meet changing IT environments
Resilient to outages and moves data with transaction integrity.

Let’s take a deeper look how GG works. As I mentioned, GG moves data between hetereogenous databases, as shown here. It does not have to Oracle. The software operates at the database level and the capture component is typically installed on the source database server outside of the DBMS. GoldenGate’s CAPTURE reads native transaction logs and captures transactions as soon as they commit and takes outside of the database system to be queued. It only moves changed, committed transactional data– therefore operates with extremely high performance and very low impact .
It allows filtering at table, column and/or row level and Transformations can be applied at this capture stage or later when delivering.

After the capture GoldenGate converts the committeds transctions into a universal data format in “trail” files. Using source and target trail files, it eliminates any single point of failure and ensures data integrity is maintained – even in the event of a system error or outage.

Another Capture component can be use to pump the data to multiple targets and for better recoverability. This configuration enhances the fault tolerance and reliability of the overall GoldenGate environment. In the event of a network failure (between the source and the target systems), the GoldenGate Capture component can continue to capture transactions since the data can be queued up locally in the Trail Files on the source, enhancing the recoverability in case of database failures.

This is NOT the database feature called Data Pump.

From the trail files the data is routed via TCP/IP to the target systems. During routing Data compression and encryption can be applied and thousands of transactions can be moved per second, without any distance limitations.

The other key component of GoldenGate is the Delivery module that is installed on the target database server. Qeued Transactions that are stored in the trail files are applied to target using native SQL calls
Basic transformations at the row level can be applied if desired at delivery or capture.

This data movement can be done bidrectionally. GoldenGate has built in loop detection to differentiate between new transactions versus the data it replicates.

GoldenGate’s modular architecture allows a number of data movement configurations to be supported. Here are popular examples. It can be used to support multi master database replication, and of course consolidation of data such as in a data warehouse, even in a cascading sytyle where you may have an ODS before feedig the DW or a data mart to feed from the DW. And with ability to publish to message quees and topics in XML and other popular formats target systems for the changed data can be a long list of databases, applications, process management engines, analytical engines, B2b partners etc.

GoldenGate is ideal for heterogeneous environments – not just supporting different versions of the same database or operation system/hardware, but replicating and integrating data across vendor systems.

We support log-based Capture of changed data from nearly all major database vendors.

We can Deliver that data to an even wider range of targets – including open source databases, several data warehouse appliances, ETL servers, and JMS message queues to support Service Oriented Architectures (SOA) and Event-Driven Architectures (EDA).

© 2010, www.oracledatabase12g.com. 版权所有.文章允许转载,但必须以链接方式注明源地址,否则追究法律责任.

相关文章 | Related posts:

1. Real-Time Business Intelligence using Oracle GoldenGate
2. Managing Performance and Availability for 25,000 Siebel Contact Center Users with Oracle Real Application Clusters
3. No Time for Downtime Patching Database using Oracle Enterprise Manager
4. Oracle Clusterware (formerly CRS) Rolling Upgrades
5. Siebel CRM with Oracle® Cost-Based Optimizer (CBO)
6. Performance Tuning Guide for Siebel CRM Application on Oracle
7. Complete Upgrade Checklist for Manual Upgrades from 8.X / 9.0.1 to Oracle9iR2 (9.2.0)
8. Complete Checklist for Manual Upgrades to 11gR2
9. Siebel Maximum Availability Architecture (MAA)
10. Oracle Database 10G – Manageability