Process networks as a high-level notation for metacomputing

Abstract

Our work involves the development of a prototype Geographical Information System (GIS) as an example of the use of process networks as a well-defined high-level semantic model for the composition of GIS operations. Our Java-based implementation of this prototype is known as PAGIS (Process network Architecture for GIS).

Our process networks consist of a set of nodes and edges connecting those nodes assembled as a Directed Acyclic Graph (DAG). In our prototype, nodes represent services and edges represent the flow of data (in this case sub-processed imagery) between services. Services are pre-defined operations that can be performed on imagery, presently selected from the Generic Mapping Tools (GMT) library. In order to control the start and end-point of the DAG, we define an input node (the original image) and an output node (the result image).

To exploit potential parallelism, we extend our idea of a process network to a distributed process network, where each service may be processed on different computers. A single server coordinates computation and computation is performed by any number of workers. The server and workers together can beseen as a metacomputer. The server takes a process network from a client and distributes work to the workers. Each worker applies for work and decides if it is capable of performing the work offered. In this way, scheduling is essentially dynamic, and computation can be performed without client intervention.