No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
A Novel Performance Prediction Framework for Web Service Workflow Applications
Abstract
Web Service play an important role in the Service-oriented Architecture (SOA), which is a new paradigm to implementing dynamic e-business solution, as Web services can be composed in an orchestrated manner by using Business Process Execution Language (BPEL). In this context, the performance of a Web service workflow is a very important factor for Business Process Re-engineering (BPR). A framework for the performance prediction and analysis of service-based applications from users’ perspectives was present in this paper. A historical time series for a specific performance is evaluated first in the framework. And then Particle Swarm Optimization based Back Propagation Neural Network (PSO-BPNN) model is constructed based on time series to predict the dynamic performance of workflow systems. When the predicted value is out of the preset range, we analyze the issues according to data of Quality of Service (QoS) which is detected at runtime, to find why cause service performance failure. Thus it suggests more suitable recovery strategies for service composition. To bring this approach to fruition we analyze a simple case study.
... In addition, we carry out the analysis without the assumption of exponential distributions, and variance of the response times are computed in this paper. In [16], although the paper also studies the impact of a service on the composite one, it is different since ours is a sensitivity analysis to look for the bottlenecks while they take a correlation analysis to identify services when performance violation happens. ...
... A workflow SLA violation detection control model (WSVDC) was proposed by Sun et al. [7], in which they considered a utility function and control charts for each SLA. Their proposed framework comprised three workflow modules -SLA aggregation and assessment, calculation of SLA, and detection of violation. ...
A service level agreement (SLA) is a legal document that binds consumers and providers together for the delivery of specific services for a certain period of time. Providers need a viable SLA to maintain successful relationships with consumers. A viable SLA, based on the previous profile of a consumer, will help a service provider determine whether to accept or reject a consumer’s request and the amount of resources to offer them. In this paper we propose a soft-computing based approach to form a personalized and viable SLA. This process is carried out in the pre-interaction time phase. We build a Fuzzy Inference System (FIS) and consider a consumer’s reliability value and contract duration as the input factors to determine the amount of resources to offer to the consumer. In addition to the Fuzzy Inference System, we tested various Neural Network-based methods for viable SLA formation and compared their prediction accuracy with the output of the FIS.
... In addition, we carry out the analysis without the assumption of exponential distributions, and variance of the response times are computed in this paper. In [16], although the paper also studies the impact of a service on the composite one, it is different since ours is a sensitivity analysis to look for the bottlenecks while they take a correlation analysis to identify services when performance violation happens. ...
When combining several services into a composite service, it is non-trivial to determine, prior to service deployment, performance and reliability values of the composite service. Moreover, once the service is deployed, it is often the case that during operation it fails to meet its service-level agreement (SLA) and one needs to detect what has gone wrong (i.e., performance/reliability bottlenecks). To study these issues, we develop a Semi-Markov Process (SMP) formulation of composite services with failures and restarts. By explicitly including failure states into the SMP representation of a service, we can compute both its performance and reliability using a single SMP. We can also detect its performance and reliability bottlenecks by applying the formal sensitivity analysis technique. We demonstrate our approach by choosing a representative example that is validated using experiments on real Web services.
Cloud services have been utilized in large-scale distributed environments. As an effective service aggregation methodology, workflow technology has been used to construct composite services. Efficient and dependable workflow scheduling (WFS) is crucial for integrating enterprise systems. While WFS has been widely studied, WFS-related algorithms are mainly focused on optimizing execution time or cost. However, in cloud computing environment, WFS is up against the threats of the inherent uncertainty and unreliability to the applications. Therefore, trust service-oriented strategies must be considered in WFS. As a result, this paper proposes a trust service-oriented workflow scheduling algorithm. The scheduling algorithm adopts a trust metric that combines direct trust and recommendation trust. In addition, we provide balance policies to enable users to balance different requirements, including time, cost, and trust. A case study was conducted to illustrate the value of the proposed algorithm. The experimental results show that the proposed approach is effective and feasible.
The disqualification of third-party cloud services has caused great threat to large-scale Enterprise Systems (ESs). Service Level Agreements (SLA) management is used to enhance Quality of Service (QoS) in an effective way. Conventional service exception handling is triggered to deal with functional failure. However, in some critical business scenario, service failures and performance violations need to be proactively prevented instead of recovery triggered by the occurrence of those failures. In this paper, a proactive detective control model is proposed to prevent workflow SLA violation supported by the utilization of SLA utility functions and control charts. Emulation case study illustrates the proposed model, and simulation results show that workflow SLA management can effectively avoid cloud services composition failures.
Contemporary web applications are often designed as composite services built by coordinating atomic services with the aim of providing the appropriate functionality. Although functional properties of each atomic service assure correct functionality of the entire application, nonfunctional properties such as availability, reliability, or security might significantly influence the user-perceived quality of the application. In this paper, we present CLUS, a model for reliability prediction of atomic web services that improves state-of-the-art approaches used in modern recommendation systems. CLUS predicts the reliability for the ongoing service invocation using the data collected from previous invocations. We improve the accuracy of the current state-of-the-art prediction models by considering user-, service- and environment-specific parameters of the invocation context. To address the computational performance related to scalability issues, we aggregate the available previous invocation data using K-means clustering algorithm. We evaluated our model by conducting experiments on services deployed in different regions of the Amazon cloud. The evaluation results suggest that our model improves both performance and accuracy of the prediction when compared to the current state-of-the-art models.
Web Services play an important role in the Service-oriented Architecture paradigm, as they allow services to be selected on-the-fly to build applications out of existing components. In this scenario, the Busi- ness Process Execution Language notation can be used as an orchestra- tion language which allows the user to describe interactions with Web Services in a standard way. The performance of a BPEL workflow is a very important factor for deciding which components must be selected, or to choose whether a given sequence of interactions can provide the re- quested quality of service. Due to its very dynamic nature, workflow per- formance evaluation can not be accomplished using traditional, heavy- weight techniques. In this paper we present a multi-view approach for the performance prediction of service-based applications encompassing both users and service provider(s) perspectives. As a first step towards the realization of this integrated framework we present an ecient approach for performance assessment of Web Service workflows described using the BPEL notation. Starting from annotated BPEL and WSDL specifi- cations, we derive performance bounds on response time and throughput. In such a way users are able to assess the eciency of a BPEL workflow, while service provider(s) can perform sizing studies or estimate perfor- mance gains of alternative upgrades to existing systems. To bring this approach to fruition we developed a prototype tool called bpel2qnbound, using which we analyze a simple case study.
The web services paradigm provides organizations with an environment to enhance B2B communications. The aim is to create modularized services supporting the business processes within their organization and also those external entities participating in these same business processes. Current web service frameworks do not include the functionality required for web service execution performance measurement from an organization perspective. As such, a shift to this paradigm is at the expense of the organization’s performance knowledge, as this knowledge will become buried within the internal processing of the web service platform. This research introduces an approach to reclaim and improve this knowledge for the organization establishing a framework that enables the definition of web services from a performance measurement perspective, together with the logging and analysis of the enactment of web services. This framework utilizes web service concepts, DSS principles, and agent technologies, to enable feedback on the organization’s performance measures through the analysis of the web services. A key benefit of this work is that the data is stored once but provides information both to the customer and the supplier of a web service, removing the need for development of internal web service performance monitoring.
SLAs are contractually binding agreements between service providers and consumers, mandating concrete numerical target values which the service needs to achieve. For service providers, it is essential to prevent SLA violations as much as possible to enhance customer satisfaction and avoid penalty payments. Therefore, it is desirable for providers to predict possible violations before they happen, while it is still possible to set counteractive measures. We propose an approach for predicting SLA violations at runtime, which uses measured and estimated facts (instance data of the composition or QoS of used services) as input for a prediction model. The prediction model is based on machine learning regression techniques, and trained using historical process instances. We present the architecture of our approach and a prototype implementation, and validate our ideas based on an illustrative example.
To illustrate the relationships between the performance of service and workflow, concepts of service and service-oriented workflow were defined. Characteristics and operation mechanism of service-oriented workflow were explained with an example. Then, a service-oriented workflow performance evaluation model was put forward, which combined the performances of strategic layer, business layer, service layer and IT layer. Key Performance Indicators (KPI) and their formal representations corresponding to each layer were also described, and then the concept of correlation was introduced. Finally, the improved Analytic Hierarchy Process (AHP) was explicated to calculate the correlation between different KPIs and to select the most suitable services.
In order to accurately predict the web service Quality of Service(QoS)in a highly dynamic environment, this paper proposes a Qos dynamic prediction method based on semi-Markov processes(SMP)and case-based reasoning(CBR). This method firstly uses SMP to predict business state of web service in the future, and applies the technology of CBR to predict web service QoS such as a specific task. Experimental results show that this prediction method can improve the accuracy of web service QoS greatly. The results provide a reliable basis for the objective evaluation and successful web service composition.
In order to accurately predict the Web service Qos in a highly dynamic environment,we put forward a Qos dynamic prediction method based on Semi-Markov Processes(SMP) and Case-based Reasoning(CBR).This method firstly uses semi-Markov process to predict business state of web service in the future, and then applies the technology of CBR to predict Web service QoS, for example,when the service deals with a specific task. Experimental results show that this prediction method can improve the accuracy of Web service QoS greatly. The results provides a reliable basis for the objective evaluation and successful Web service composition.
Service Oriented Architecture (SOA) is a business-centric IT architectural approach for building distributed systems. Reliability of service-oriented systems heavily depends on the remote Web services as well as the unpredictable Internet connections. Designing efficient and effective reliability prediction approaches of Web services has become an important research issue. In this article, we propose two personalized reliability prediction approaches of Web services, that is, neighborhood-based approach and model-based approach. The neighborhood-based approach employs past failure data of similar neighbors (either service users or Web services) to predict the Web service reliability. On the other hand, the model-based approach fits a factor model based on the available Web service failure data and use this factor model to make further reliability prediction. Extensive experiments are conducted with our real-world Web service datasets, which include about 23 millions invocation results on more than 3,000 real-world Web services. The experimental results show that our proposed reliability prediction approaches obtain better reliability prediction accuracy than other competing approaches.
Availability of services on Internet has provided unique opportunity to customers as well as providers for conducting e-business. This new business paradigm can succeed provided selection of services is accomplished to customer satisfaction in terms of service delivery time as well as service quality. Instead of leaving it to service provider to declare expected service response time, it would be of great use if there is a way by which service customers could dynamically determine response time of a service at the time of service request. Hence we propose a HMM model for forecasting response time of a web service before making a call to it. The paper deals with model details and proposes a framework for implementation of the model. We also demonstrate its usability in scenarios like checkpointing orchestrated services.
Reliability prediction models for Service Oriented Architecture (SOA) based applications have thoroughly discussed system level & component level details. However, significant difference between applications based on service-oriented architecture and traditional distributed application is marshaling of remote web services, nonetheless, reliability of these web services have not been discussed significantly. Recently developed strategies did not discuss how we can establish a mechanism which helps to predict reliability of internal structure of web service that may have involved either in failure or slow response. In this paper, we have specifically focused on the area of predicting reliability parameters like response time & throughput of hidden states that exist in any remote web service. Our proposed model will help end users to predict probability of success or failure of underlying hidden states of web services. Along with reliability it shall further help us to compute probability of scalability of those hidden states.
Failures during web service execution may depend on a wide variety of causes. One of those is loss of Quality of Service (QoS). Failures during web service execution impose heavy costs on services-oriented architecture (SOA). In this paper, we seek to achieve a self-healing architecture to reduce failures in web services. We believe that failure prediction prevents the occurrence of failures and enhances the performance of SOA. The proposed architecture consists of three agents: Monitoring, Diagnosis and Repair. Monitoring agent measures quality parameters in communication level and predicts future values of quality parameter by Time Series Forecasting (TSF) with the help of Neural Network (NN). Diagnosis agent analyzes current and future QoS parameters values for diagnose web service failures. Based on its algorithm, the Diagnosis agent detects failures and faults in web services executions. Repair agent manages repair actions by using Selection agent.
Web service composition is a new paradigm to develop distributed and reactive software-intensive systems. Predicting and preventing failures of dynamic composite services is an important and challenge problem due to the dynamically evolving attribute. In previous work, we propose CASSANDRA, a novel proactive monitoring technique with the ability to predict and prevent the potential failures happening in dynamic evolvable system. In this paper, we concretize the approach into web service composition field. By combining runtime information and design-time specification of basic services, the approach can analyse future -step models ahead of the current service execution states. Then, this model can be used to check with a set of desired properties represented by property sequence chart. Initial experiments on an online medicine case study validates our approach and shows encouraging results.
As composition consists of different Web Services invocations, when one component service fails, composite Web Service will not operate appropriately. The easy solution to this problem is to reselect the service every time service fails. However, it is not feasible due to the high complexity of the reselection, which will interrupt the execution of composite service, lead to an extra delay and influence the performance of the composite service. In this paper we propose an approach on Quality of Service (QoS) aware performance prediction for self-healing Web Service Composition. In our approach, we first propose a self-healing cycle which has three phases such as monitoring, diagnostics and repair. Next, in order to minimize a number of reselections we propose Decision Tree based performance prediction approach. With our approach, the component services which have previously violated QoS parameter values can be predicted. We will demonstrate that proposed solution has better performance in supporting the self-healing Web Service composition comparing to traditional way.
Web Service has become a newly distributed computing paradigm. Modeling and analyzing the workflow of composite web services is the most important task in the process of developing service oriented software. BPEL is the industry standard used for combining a set of web services to define a new composite web service. A Petri net model- WS_Net is defined which suitable for describing the business processes of composite web services. Then, a method used for translating a BPEL process to WS_Net is presented. The reliable data of a composite web services is collected by extending UDDI model. WS_RPN which captures the reliability aspect of the web services composition and can be used for analyzing the reliability of BPEL processed through the execution paths is defined based on WS_Net. Finally an example is used to illustrate the method.
In order to manage enterprise businesses effectively, enterprise decision makers need to understand enterprise business processes from various perspectives using sophisticated process simulation and optimization tools. Evaluation of enterprise processes is the basis of enterprise process simulation and optimization research for business process re-engineering. This paper proposes a methodology for dynamic enterprise process performance evaluation with metric measurement models based on Activity-Based Costing and Activity-Based Management (ABC/ABM) for six types of process flows within manufacturing enterprises, including activity flow, product information flow, resource flow, cost flow, cash flow, and profit flow. The proposed methodology uses time, quality, service, cost, speed, efficiency, and importance as seven criteria. A prototype software system has been implemented to validate the proposed methodology.
Distributed applications are rapidly converging towards the adoption of a computing paradigm based on service-oriented architectures (SOA), according to which an application results from the composition of a set of services in execution on networked server hosts. In a SOA context, service providers are strategically interested both to describe the performance characteristics of offered services, to better qualify their offer and gain a significant advantage in the global marketplace, and to predict the level of performance that can be offered to service consumers when building composite web services that make use of services managed by various service providers. This paper introduces a model-driven approach for integrating performance prediction into service composition processes carried out by use of BPEL (Business Process Execution Language for Web Services). The proposed approach is founded on P-WSDL (Performance-enabled WSDL), a performance-oriented extension of WSDL, the language for describing the information about service capabilities and invocation mechanisms. P-WSDL is a lightweight WSDL extension for the description of performance characteristics of a web service. The approach is illustrated by use of an example application to a composite web service for travel planning.
The problem of composing services to deliver integrated business solutions has been widely studied in the last years. Besides
addressing functional requirements, services compositions should also provide agreed service levels. Our goal is to support
model-based analysis of service compositions, with a focus on the assessment of non-functional quality attributes, namely
performance and reliability. We propose a model-driven approach, which automatically transforms a design model of service
composition into an analysis model, which then feeds a probabilistic model checker for quality prediction. To bring this approach
to fruition, we developed a prototype tool called ATOP, and we demonstrate its use on a simple case study.
Virtual enterprises (VEs) have become one of main types of organisations in the manufacturing sector through which the consortium companies organise their manufacturing activities. To be competitive, a VE relies on the complementary core competences among members through resource sharing and agile manufacturing capacity. Manufacturing grid (M-Grid) is a platform in which the production resources can be shared. In this article, an M-Grid-based VE operation platform (MGVEOP) is presented as it enables the sharing of production resources among geographically distributed enterprises. The performance management system of the MGVEOP is based on the balanced scorecard and has the capacity of self-learning. The study shows that a MGVEOP can make a semi-automated process possible for a VE, and the proposed MGVEOP is efficient and agile.
Quality of service prediction of multi-agent web service integration system based on grey meural network
- M Zhang
- J Li
- J Xing
Reliability Prediction Model for Web Services Based on Control structure
- C Xie
- X Wang
Web Service QoS prediction method based on time series analysis
- Z Hua
- M Li
- J Zhao