Workflow management with service quality guarantees

Abstract

Workflow management systems (WFMS) that are geared for the orchestration of business processes across multiple organizations are complex distributed systems: they consist of multiple workflow engines, application servers, and communication middleware servers such as ORBs, where each of these server types can be replicated on multiple computers for scalability and availability. Finding an appropriate system configuration with guaranteed application-specific quality of service in terms of throughput, response time, and tolerable downtime is a major challenge for human system administrators. This paper presents a tool that largely automates the task of configuring a distributed WFMS. Based on a suite of mathematical models, the tool derives the necessary degrees of replication for the various server types in order to meet specified goals for performance and availability as well as "performability" when service is degraded due to outages of individual servers. The paper describes the configuration tool, with emphasis on how to capture the load behavior of workflows in a realistic manner. We also present extensive experiments that evaluate the accuracy of the tool's underlying models and demonstrate the practical feasibility of automating the task of configuring a distributed WFMS. The experiments use a detailed simulation which in turn has been validated through measurements with the Mentor-lite prototype system.

QoS-aware middleware for Web services composition

“…Constraint (8) indicates 2. The idea of cloning the states in a loop for the purpose of transforming a cyclic statechart into an acyclic one has been proposed by Gillmann et al [15]. that if task t k is a direct successor of task t j , then the execution of t k must start after task t j has been completed.…”

“…QoS management has been widely discussed in the area of middleware systems [3], [25], [15]. The focus of these works is essentially on the following issues: QoS specification to allow description of application behavior and QoS parameters, QoS translation and compilation to translate specified application behavior into candidate application configurations for different resource conditions, QoS setup to appropriately select and instantiate a particular configuration, and, finally, QoS adaptation to runtime resource fluctuations.…”

“…Most efforts in this area focus on specifying and enforcing temporal constraints [13], [6], [15]. Some projects such as METEOR [8] and CrossFlow [21] consider other QoS criteria than time.…”

Scalable peer-to-peer process management - the OSIRIS approach

“…Albeit navigation tasks can be distributed in a cluster, storage of process instances usually is done by using a single, centralized database instance (products like Oracle 10g [24] can also support clustered databases). Some prototype systems features similar architectures (e.g., Mentorlite [16], SDM [2]). Although executing processes in a peer-to-peer way without involving a centralized component, OSIRIS provides transactional guarantees, following the model of transactional processes [30].…”

“…In the distributed Mentor-lite [16] approach, the setting of the process engine in a cluster can be changed actively using a configuration tool, while OSIRIS distributes process execution over all available service providers, and locally runs processes on provider hosts. This requires a middleware layer -the local HDB layer-on every participating component resp.…”