๐”– Bobbio Scriptorium
โœฆ   LIBER   โœฆ

Engineering a BPEL orchestration engine as a multi-agent system

โœ Scribed by Mirko Viroli; Enrico Denti; Alessandro Ricci


Publisher
Elsevier Science
Year
2007
Tongue
English
Weight
956 KB
Volume
66
Category
Article
ISSN
0167-6423

No coin nor oath required. For personal study only.

โœฆ Synopsis


The BPEL language is currently the primary candidate for standardising Web Services orchestration. BPEL specifications are meant to be run by BPEL orchestration engines, which are therefore crucial components of today's business-to-business infrastructures, carrying the burden of dynamically composing existing services.

In this paper, we present the design of a BPEL orchestration engine based on a multi-agent system: while the basic BPEL activities are autonomously executed by agents, workflow aspects are realised by the mediation of ReSpecT tuple centres, a coordination model extending LINDA with the ability of declaratively programming the tuple space behaviour.

Our architecture separates the interaction, correlation, and workflow concerns into clearly identified tiers. In particular, we identify the workflow tier as the one encapsulating the core and most critical behaviour of the engine: due to its intrinsic complexity, we tackle its design formally. We introduce a core algebraic language of BPEL dealing with its workflow-related aspects, and provide it with a semantics based on a mapping into a net specification, modelling the dependencies between the activities to be executed by the engine. This mapping plays the role of a formal design, since it directly leads to an implementation of the workflow tier in the orchestration engine.


๐Ÿ“œ SIMILAR VOLUMES


A multi-agent paradigm as structuring pr
โœ Dick J.M. Saarloos; Theo A. Arentze; Aloys W.J. Borgers; Harry J.P. Timmermans ๐Ÿ“‚ Article ๐Ÿ“… 2008 ๐Ÿ› Elsevier Science ๐ŸŒ English โš– 659 KB

Although urban plans may appear simple once completed, they are the outcomes of a highly complex decision-making process. This complexity makes it difficult to develop Planning Support Systems (PSS) that are comprehensive, flexible, intelligent and, yet, understandable to users. Great improvements a