A prototype Taskspaces framework for grid computing of scientific computing problems that require intertask communication is presented. The Taskspaces framework is characterized by three major design choices: decentralization provided by an underlying tuple space concept, enhanced direct communication between tasks by means of a communication tuple space distributed over the worker hosts, and object orientation and platform independence realized by implementation in Java. Grid administration tasks, for example resetting worker nodes, are performed by mobile agent objects. We report on large--scale grid computing experiments for iterative linear algebra applications showing that our prototype framework scales well for scientific computing problems that require neighbor--neighbor intertask communication. It is shown in a computational fluid dynamics simulation using a Lattice Boltzmann method that the Taskspaces framework can be used naturally in interactive collaboration mode. The scalable Taskspaces framework runs fully transparently on heterogeneous grids while maintaining a low complexity in terms of installation, maintenance, application programming and grid operation. It thus offers a promising roadway to push scientific grid computing with intertask communication beyond the experimental research setting.

We define and describe a model for coordination of distributed processes or components based on associative broadcast. Associative broadcast encapsulates processes with an associative interface. The associative interface includes a profile, which specifies the current state of the component. Each message is sent with a conditional expression (selector), which evaluates to true for specific instances of profiles. Messages are broadcast but are received by only those processes where the selector of the message evaluates to true when matched with the profile of the component. Each component dynamically specifies its profile and selectors to conform to a coordination protocol.

This paper proposes Update Plans as a specification formalism for abstract machines for parallel architectures. Update Plans are a formal specification language for abstract and concrete machines. First results in using Update Plans to specify parallel architectures are illustrated, and some suggestions for further research are made.

Abstract. A development environment for applications specified in an extended version of a previously developed coordination model based on associatively broadcast interactions is presented. The previous associative broadcast coordination model is extended to incorporate more complex specifications for interactions including multiple message interactions and fault-tolerance by replication. The runtime system is extended to facilitate construction and application of distributed implementation of coordination systems. An interface definition language based on the extended coordination model and a compiler for the language are defined and described. Three example applications, a generalized readers/writers problem including replication, a “greedy reuse” algorithm and a distributed computation Google pageranks are presented. 1

The pervasive penetration of database technology may suggest that we have reached the end of the database research era. The contrary is true. Emerging technology, in hardware, software, and connectivity, brings a wealth of opportunities to push technology to a new level of maturity. Furthermore, ground breaking results are obtained in Quantum- and DNA-computing using nature as inspiration for its computational models. This paper provides a vision on a new brand of database architectures, i.e. an Organic Database System where a large collection of connected, autonomous data cells implement a semantic meaningful store/recall information system. It explores the analogy of a biological complex to charter the contours of this research vision. A concrete computational model is defined and illustrated by examples as a step into this direction. 1991 Mathematics Subject Classification: 68P20,68P15,68N99 1991 Computing Reviews Classification System: Database Logical Design(H.2.1), Datab...

Due to ever-increasing data sizes and the high computational complexity of many algorithms, there is a natural drive towards applying parallel and distributed computing to bioinformatics problems. Grid computing techniques can provide flexible, portable and scalable software solutions for parallel bioinformatics. Here we describe the TaskSpaces software framework for grid computing. TaskSpaces is characterized by two major design choices: decentralization, provided by an underlying tuple space concept, and platform independence, provided by implementation in Java. We discuss advantages and disadvantages of this approach, and demonstrate seamless performance on an ad-hoc grid composed of a wide variety of hardware for a real-life parallel bioinformatics problem. Specifically, we performed virtual experiments in RNA folding on computational grids composed of fast supercomputers, in order to estimate the smallest pool of random RNA molecules that would contain enough catalytic motifs for starting a primitive metabolism. These experiments may establish one of the missing links in the chain of events that led to the origin of