Dependability Assessment of Web Service Orchestrations

Abstract

In this paper, we focus on the reliability and availability analysis of Web service (WS) compositions, orchestrated via the Business Process Execution Language (BPEL). Starting from the failure profiles of the services being composed, which take into account multiple possible failure modes, latent errors, and propagation effects, and from a BPEL process description, we provide an analytical technique for evaluating the composite process' reliability-availability metrics. This technique also takes into account BPEL's advanced composition features, including fault, compensation, termination, and event handling. The method is a design-time aid that can help users and third party providers reason, in the early stages of development, and in particular during WS selection, about a process' reliability and availability. A non-trivial case study in the area of travel management is used to illustrate the applicability and effectiveness of the proposed approach.

Fault analysis of service‐oriented systems: a systematic literature review

“…• four replicas to tolerate from fault replica [36] • optimistic replacement and replication technique [36,80] • finite state machine to model the approach [80] byzantine/arbitrary fault [52,96] implementation in standard SOAP messaging framework • Markov chain-based approach [76,82] latent errors and dormant faults [46,111] • unable to deliver successful service • coordinated atomic actions-DRIP (CAA-DRIP) framework and a simulation technique [46] • an analytical technique to calculate reliabilityavailability metrics [111] adaption faults [72,112] estimating the failure probability of each static component. In addition to [119], Ding et al [93] have proposed the online prediction of the reliability of SOS.…”

“…• probabilistic critical path identification [61] • incremental SLA agreement violation handling approach [65] reliability [14,20,23,27,42,57,71,75,82,87,88,93,101,110,111,116,117,119,125] • the black-box component is difficult to evaluate their quality • unnecessary restart to recover from failure [20] • reliability-critical workflow components [110,125] • recoverable node failures [87] • a framework and prototype tool for detecting anomalous services [116] • optimal reliable service composition and redundancy technique [23,75] • neighbourhood-based MF [119] • soft-state inference for rapid failure recovery [20] • heuristic recommendation algorithms [125] • FFDA tool and fault injection [110] • k-reliability to calculate the surviving nodes in the subnetwork [101] • reliability prediction using the running time data of the system [93] • task scheduling optimisation model [87] scalability [67,76,81] • only one error at the time [67].…”

“…• lack of quantifying the availability [76] • tested in Llama Cloud middleware [81] • a stochastic model-driven approach using monolithic models [76] • migration of the physical machines among different pools availability [76,88,89,111] • time-consuming…”

Evaluating the Reliability of Ambient-Assisted Living Business Processes

“…Within the WS-BPEL context, Mukherjee et al (2008) compute the reliability of WS-BPEL processes taking into account most of the workflow patterns that WS-BPEL can express, while the method of Distefano et al (2014) also incorporates advanced composition features such as fault, compensation, termination and event handling.…”

Incremental Syntactic-Semantic Reliability Analysis of Evolving Structured Workflows

“…Moreover, for the sake of readability and to reduce the complexity of attribute schemas, Mini workflows support only (global) boolean variables; we model invocation of external services as boolean functions with no input parameters. We remark that more complex analyses and workflow languages (see, for example, the extension of [13] in [12] for support of BPEL business processes, including parallelism and nested workflows) could be supported with richer attribute schemas.…”