Instancing

A running Dataphor server has always been called an instance, but recent changes have made configuring and administering multiple instances dramatically easier. Multiple instances can be installed and managed on the same machine without the need to run command-line installations, or manage configurations using config files. A new listener service has been added to make instance discovery possible from the client, and port configuration can now be managed completely server-side. In addition, several changes have been made to consolidate the data used by an instance and enable multiple instances to use the same executable and library directories.

Native CLI

A new, lightweight, potentially stateless CLI has been built to enable a Dataphor server to be accessed from applications without having to incur the overhead required by the standard CLI. Because it is intended as an alternative, not a replacement, the new CLI is not as feature-rich, but it eliminates the dependencies on the Dataphor code base and can be used virtually stand-alone. This will enable the Dataphor server to be reached from a much broader range of clients, such as mobile applications, Silverlight clients, and other environments where resources are at a premium. In addition, the stateless potential of the new Native CLI enables applications to make scaling decisions that were previously infeasible with the standard CLI.

Cross-Instance Query

The Dataphor server now supports the configuration of Server Links that can be used to query other running instances of Dataphor servers. The Dataphor Server coordinates transactions between the current instance and any number of connected instances to provide seamless nested distributed transaction support.

SQLite Catalog Support

The catalog persistence layer was abstracted to provide the ability to plug-in different catalog store implementations, and a SQLite catalog store implementation was built. The Dataphor Server can now be configured to use SQLite, rather than SQL Server Compact Edition to provide catalog persistence.

Mono Support

In order to enable the Dataphor server to be used on a Mono platform, we have taken steps to remove any platform-specific dependencies in the code base, as well as the client tools.

All these features are currently implemented in the main development branch of the Dataphor project. To try them out, simply get latest from the open source svn repository and build the Dataphor solution. We have consolidated the projects so that it's easier than ever to build and get up and running quickly.

For more complete discussion and up-to-date documentation about the latest features available in Dataphor, see the What's New in 2.2 page of the dataphor wiki.