作者: Johannes Behl , Tobias Distler , Florian Heisig , Rüdiger Kapitza , Matthias Schunter
关键词: Computer science 、 Workflow 、 Distributed computing 、 Fault tolerance 、 Web service 、 Reuse 、 Service (systems architecture) 、 Cloud computing 、 Business process management 、 Replication (computing)
摘要: With a variety of services rapidly evolving at all architectural levels cloud computing, there is an increasing demand for standardized way to coordinate their interactions. Business process management, that is, more general, the management Web-service--based workflows, could satisfy this and, indeed, first corresponding offerings have gained instant popularity. While from functional perspective, these Platform-as-a-Service (PaaS) solutions are already quite mature, support fault tolerance still very limited, making them inapplicable critical tasks.Concerning deficiencies currently existing systems, paper presents practical solution executing particularly within clouds, in fault-tolerant, highly available and configurable manner. We achieve by actively replicating workflows as well Web combined architecture, reusing standard systems coordination services. By providing automated transformation tool, replication realized transparently workflows. Measurements show our proposed architecture achieves lower response times than integration service imposes only moderate costs, while simplifying implementation leading dynamically adaptable solution.