Oliver Krone

CoLa: a coordination language for massive parallelism

Introduction The emerging generation of supercomputers will be suitable for massive parallel applications. To orchestrate the communication needed to coordinate concurrent tasks, we are currently developing a new coordination language [1] called COLA. Specifically C’OLA is aimed to provide a high programming abstraction level for massively parallel computing, especially for applications in the field of distributed artificial intelligence. COLA is embedded in classical programming languages and combines both declarative as well as imperative semantic properties in its definition. COLA incorporates a high level identification concept, defining communication objects as correspondents [2], a postal mail delivery model for efficient and transparent message passing [3][4] and new communication and synchronization mechanisms.

Coordinating autonomous entities

This paper describes STL, a new coordination model and corresponding language. STLs power and expressiveness are shown through a preliminary distributed implementation of a generic autonomybased multi-agent system, which is applied to a collective robotics simulation, thus demonstrating the appropriateness of STL for developing a generic coordination platform for autonomous agents.

Coordinating autonomous entities with STL

This paper describes ECM, a new coordination model and STL its corresponding language. STLs power and expressiveness are shown through a distributed implementation of a generic autonomy-based multi-agent system, which is applied to a collective robotics simulation, thus demonstrating the appropriateness of STL for developing a generic coordination platform for autonomous agents.

Integrating Coordination Features in PVM

In this paper we focus on coordination for parallel systems controlled by the PVM software package. We describe several extensions to PVM which allow the use of extended coordination features like generative communication or client/server programming.

Load Balancing for Network Based Multi-threaded Applications

In this paper we present Lbs, a load-management-system for network based concurrent computing. The system is built on Pt-Pvm, a library based on the PVM system. Pt-Pvm provides message passing and process management facilities at thread and process level for a cluster of workstations running the UNIX operating system.