One of the main attractions of Dbvisit Standby is that it takes the complication out of setting up and running Oracle standby databases.
Mike Meerding, Database Administrator, Australia

Data Distribution

Data distribution is the process of synchronizing data across multiple database servers, providing a way to maintain up-to-date copies of data in multiple databases simultaneously. It also provides the ability to share, distribute, synchronize and consolidate data in real time. This provides many opportunities to extend the availability and usefulness of data including:

Load balancing and performance improvement
Making transactional and reference data available at multiple locations
Synchronizing data between applications, organizations or sites
Providing data for test systems
Providing applications and users with subsets of data for performance and security purposes

Data distribution can be in one direction, or bi-directional. It can be between two databases (one-to-one), can publish from one source to many destinations (one-to-many) or can collect and consolidate from many sources (many-to-one).

The following diagram illustrates one possible data distribution scenario including consolidation of data from multiple remote sites, and the dissemination of data from the primary site (bi-directional):

Data distribution overview

The following list provides examples of where data distribution can be used to add value:

Production databases can be replicated across multiple database servers (local or remote), supporting the delivery of high availability and load balanced solutions.
Data collected by remote offices, regions or stores can be sent to head office and consolidated in real time.
Data can be synchronized with partners, suppliers and associated organizations.
Reference data (e.g. product information and pricing) can be maintained centrally and distributed to stores and remote offices.
High speed and high volume online transaction processing systems can be provided with a subset of the data they need to deliver their functionality, without the need to access the full, potentially slower, database.
Changes in one system can be sent to another system to initiate activity there, delivering elements of workflow control and inter-system communication. For example, data from a manufacturing system can be sent to the quality control system as each item is manufactured.
Live transactional data can be distributed to a remote database for reporting purposes. This offloads the considerable processing effort associated with report generation to a separate database and allows the online database to support higher transaction volumes (refer to Reporting Database for further details).
An up-to-date copy of a production database can be maintained on a different database platform. This allows organizations to deploy integrated solutions with different database management systems (e.g. Microsoft Reporting Services against a SQL Server database that is replicated from a source Oracle database).
By continually applying updates from the source database, large database migrations can be performed with minimal downtime (refer to Data Migration for further details).
A copy of a production database can be maintained for the purposes of solution testing, including regression and build testing, as well as user acceptance testing.
Live changes can be extracted from the source database and made available for transformation and loading into a data warehouse, significantly reducing the time delay, processing effort and shortcomings of a traditional overnight batch approach to ETL.

Dbvisit Replicate shares, distributes, migrates and offloads data processing workloads to alternative databases and server platforms. This delivers significant benefit at the infrastructure layer by minimizing downtime, delivering operational cost savings and facilitating real time business reporting.

Dbvisit Replicate provides secure real-time replication for businesses using Oracle® databases, including Standard Edition and Standard Edition One. It is the smart alternative to Oracle GoldenGate and Oracle Streams products; delivering the full power and potential of database replication at an affordable price.

How it works
Why distribute data?
Modes of operation

Dbvisit Replicate is powerful, affordable software that is simple to install and configure. Replicate provides Oracle to Oracle, MySQL and Microsoft SQL Server database replication and comes with all the tools and utilities necessary to deliver easy management, including conflict detection and resolution.

Dbvisit Replicate uses its own highly efficient change data capture (CDC) technology, and Oracle’s internal redo logs, to detect changes in the source data and to replicate and distribute these changes in real-time across Oracle and non-Oracle databases.

In addition to the built-in conflict detection and resolution processes, Dbvisit Replicate’s caching ensures all committed records are securely delivered to the target database(s) even in the event of an outage. This functionality reduces the risk of system interruptions, hardware or network unavailability and restoration delays.

Dbvisit Replicate uses public key cryptography to securely send data from one location to another. When exchanging data outside of a LAN, either over a WAN or over the Internet, public and private keys are used to encrypt and sign the data, ensuring that the data is exchanged securely and providing confidence to the recipient that the data has not been modified in transit. Within a LAN environment, this security can be tuned off.

Dbvisit Replicate provides:

Automatic recovery from errors and network issues
Encryption and compression of data sent across the network
Management of existing database triggers to avoid double firing
Conflict resolution business rules can be created with user defined PL/SQL stored procedures
Replication console to provide real time overview of the replication

With an abundance of software solutions throughout modern organizations, the expectation of high performance access to data across the enterprise is placing increasing demands on database servers and the databases they manage.

The data distribution process provides data synchronization across multiple database servers, thereby spreading the load and widening the availability of data. It also allows each of the accessing applications to connect to a copy of the data that is tailored to their usage, including the data available, its structure and the way it is stored (e.g. indexes). This allows an organization to provide multiple synchronized sources for the same data, each tuned to the individual requirements of the application or user.

The distribution process can transform data as it is transferred, allowing the data to be tailored to the specific purpose of the destination database. For example, data can be de-normalized in order to make reporting more efficient, or can be filtered to provide a subset suited to the purpose of a particular system.

Dbvisit Replicate can be operated in unidirectional and bidirectional modes.

In a unidirectional system changes from a source database are applied to one or more target databases, in a one-way flow of updates. Such an approach can be used for the distribution of reference data, or the transmission of data from a remote office to a central office.

The following diagram illustrates such a one-way flow of updates from the source database to the target.

1 Way Replication overview

In bi-directional operation, changes are shared between the databases in both directions. Changes are identified in each database by a mine process and are applied at the remote location by an apply process. This approach, which enables two databases to be continuously synchronized, can be used to share data or maintain different databases for different purposes (online versus batch data etc.).

The following diagram illustrates such a two-way flow of updates between two databases.

2 Way Replication overview