Pervasive DataRush

At the heart of every Pervasive DataRush-enabled application is the Pervasive DataRush Parallel Dataflow Engine. This embedded engine transforms the way your application handles large volumes of data. Developers can now build data-intensive parallel applications without specialized knowledge in parallel programming such as threading, concurrent memory access, deadlock detection, data workload partitioning/buffering or any other complex aspect of parallel thread execution. The Pervasive DataRush Parallel Dataflow Engine takes full advantage of multicore platforms, delivering outstanding performance on a single server. The Pervasive DataRush engine processes data according to a developer defined dataflow graph.

The essence of dataflow programming is the execution of a graph of computational nodes while data flows through the graph. As the data is streaming, only data required by any active operation needs to be in memory at any given time, allowing very large data sets to be analyzed. Besides offering the potential for scaling to problems larger than available memory, dataflow graphs exploit multiple forms of parallelism.

The Pervasive DataRush Java SDK and Core Libraries allow the developer to create dataflow graphs on the fly by composing a mix of Pervasive DataRush operators and custom operators that are written by the developer. The developer is in full control over what operations are performed on the streaming data.

Finally, Pervasive has built modules using the Pervasive DataRush Java SDK and Core Libraries to provide specific functionality. For example, the Pervasive DataMatcher encapsulates the functionality required to build a scalable data matching solution. Of course, developers can also build their own reusable modules.

For more information on the Pervasive DataRush technology, please visit the Developer Corner or review the Technical White Paper.