Workflow management systems (WFMSs) that are geared for the orchestration of enterprise-wide or even "virtual-enterprise"-style business processes across multiple organizations are complex distributed systems. They consist of multiple workflow engines, application servers, and ORB-style communication servers. Thus, deriving a suitable configuration of an entire distributed WFMS for a given application workload is a difficult task. This paper presents a mathematically based method for configuring a distributed WFMS such that the application's demands regarding performance and availability can be met while aiming to minimize the total system costs. The major degree of freedom that the configuration method considers is the replication of the underlying software components, workflow engines and application servers of different types as well as the communication server, on multiple computers for load partitioning and enhanced availability. The mathematical core of the method consists of Markov-chain models, derived from the application's workflow specifications, that allow assessing the overall system's performance, availability, and also its performability in the degraded mode when some server replicas are offline, for given degrees of replication. By iterating over the space of feasible system configurations and assessing the quality of candidate configurations, the developed method determines a configuration with near-minimum costs.
Abstract. The HEART tool (Help for Ensuring Acceptable Response Times) has been developed by the IT Research and Innovations department of Dresdner Bank for the computation of viable message prioritization in message-based eservices, such as stock brokerage services where service requests of different customer classes with class-specific performance goals have to be served by a server. HEART determines viable message prioritizations in the sense that they satisfy the specified performance goals of customer classes. In this paper, we describe the practical problem setting we address with HEART and outline the functionality of HEART. The demo will show HEART's underlying concepts, its architecture and an example scenario.
MotivationQuality-of-Service (QoS) is a hot topic for existing Internet-based e-commerce applications and it will become "hotter" as enterprises make more and more Web Services accessible to a wider range of customers via Internet, using UDDI registries and SOAP-based messaging. Guaranteeing good performance is highly critical to the acceptance and business success of e-services [1], such as online stock brokerage where minimum response times are required by customers for their trading transactions such as buying and selling of stocks. In this setting, it is not sufficient to the customer to get "best-effort" promises about the e-service performance, but to get specific performance guarantees which may even be cast into a formal service level agreement (SLA). These performance guarantees have not only to focus on mean response times averaged over long time periods like weeks or months. Rather they have to consider mean values, standard deviations as well as the tail of response time distributions taken in short-term intervals in order to guarantee acceptable response times even in peak load situations where the number of response time outliers may be most critical to business success. Consequently, future e-service servers have to guarantee customer-class-specific performance goals. Typically, e-services are embedded in a component-based architecture where requests (e.g. in form of a SOAP message or
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.