Performance and Availability Assessment for the Configuration of Distributed Workflow Management Systems

Abstract

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.

Workflow management with service quality guarantees

“…Finally, the performability model takes into account the performance degradation during transient failures, and estimates the effective mean waiting time for service requests with explicit consideration of periods during which only a subset of a server type's replicas are running. These models, which form the mathematical underpinnings of the Goliat tool, have mostly been derived in our earlier work published in [8]. The current paper extends our earlier work in two major ways: 1) We have refined and improved the load and performance model of Goliat so that it can capture the behavior of workflow executions in a more realistic manner.…”

“…Section 2 presents the architecture of the Goliat configuration tool. Section 3 reviews the suite of stochastic models that we developed in [8]. Section 4 presents new extensions of the stochastic performance model to capture the behavior of workflow executions in a more realistic manner.…”

“…To solve this problem, we have developed the Goliat autoconfiguration tool based on a suite of analytic models, using stochastic methods like continuous-time Markov chains and Markov reward models, to predict the performance, availability, and performability for given configuration and workload. These models have been developed in [8]; we will outline the general approach in Section 3 and will present new extensions in Section 4. The Goliat tool is driven by the workflow specifications that it obtains from the repository and by statistics on the workload from the monitoring tool of Mentor-lite.…”

Research on the New Enterprise Staff Performance Management Methodology based on the Performance Management Process Model

“…But the current actual situation it is difficult to have the staff of state-owned enterprises can do the goal of performance management and enterprise development together. In our country enterprise, in many cases, the solution of the management work is using administrative manner and there are very few in this performance management way to solve [1][2].…”

Formal Verification of e-Services and Workflows

“…Different from the research efforts [24,33] to model and analyze performance of workflow systems, our main goal is to verify the correctness of logic inside a workflow specification, e.g. consistency of data, avoidance of unsafe system states, and satisfaction of certain business constraints.…”