Caching in a Nutshell

Sybase Data Federation uses caching to insulate production data sources from haphazard access, to maintain good performance for users and applications, to refresh data in a granular way based on business needs, and to ensure maximum data availability.

Caching makes remote data access practical by limiting the number of times a data request requires immediate communication with the original data source. Data Federation provides a variety of different caching options and features to meet diverse performance requirements. You can specify a different caching strategy for each data item, and local and remote caching options can be used separately or in combination to accomplish your goals.

Local caching enables caching of frequently requested results near the data source to reduce load on the back-end data source.
Remote caching (which is remote from the data source) caches data close to the users or applications that will use it. This cuts down on network congestion and dramatically speeds up application performance, because remote data calls are eliminated. Caches can be pre-populated and updated during off hours when network load is low, and cache configurations can be established that ensure high availability when a network is congested or unavailable for some reason.

Cache update frequency can be specified for a given data item. Database administrators can schedule how often database operations should be re-run and cached, so as to protect production databases from unexpected load. Cached data can also be configured to expire after a set time period, forcing a refresh of the data on the next request.

Data Federation’s caching mechanism is efficient and simple to deploy:

It provides one solution for multiple data types, including multiple heterogeneous database management systems, files, XML documents, and application data
No additional database licenses are needed
No special hardware or network infrastructure is needed
No application changes are needed to take advantage of the cache

An automated caching approach means fewer data stores to manage. Once you have set up caching options for a given data item, there is very little work to do. Caches are updated automatically in accordance with your specifications, and the process is completely transparent to users and applications.

For more information on caching in Data Federation, see the Data Federation Provisioning and Advanced Data Integration Guide, part of the DI Suite documentation set, which is available on the Sybase Product Manuals Web site at http://sybooks.sybase.com.