Web service response time predicting method based on time sequence

Abstract

The invention discloses a web service response time predicting method based on increment steady-state analysis, belonging to the field of software performance prediction. The method comprises the following steps: obtaining a sequence of a series of measurement values of network service response time according to a fixed unit time interval; subsequently defining and calculating observation window response time increment types; calculating a transition probability matrix among the types; then calculating a steady-state distribution probability of the response time increment based on the transition probability matrix; and finally adding a calculated increment expected value to a response time measurement value at the t moment so as to obtain the prediction value of the response time at the t+1 moment. With the adoption of the method, the defects that traditional Web prediction mold and method depend too much on SLA (Service-Level-Agreement) specifications and hypothesis parameters are constant and unchanged are avoided; and mold support and analysis means are provided for WEB service credibility research.

Description

technical field [0001] The invention belongs to the field of software performance prediction, and in particular relates to a time series-based Web service response time prediction method. Background technique [0002] As a new type of distributed component model, Web service is playing an increasingly important role in the fields of e-commerce and enterprise application integration. It is considered to be one of the core technologies in Service-Oriented Computing Architecture (SOA). Service composition technology combines individual Web services to form a complete application. However, Web services run in a dynamically changing network environment, and their response time values ​​are also changing frequently. This dynamically changing network environment will affect the runtime performance of Web services, which in turn will lead to changes in the performance of the entire Web service composition. Therefore, Predicting the performance of web services is a key technology to...

AI Technical Summary

Problems solved by technology

[0003] Although the academia and industry have proposed a series of models and methods for analyzing and predicting the performance of Web services, most of the methods use static analysis based on SLA (Service-level-agreement, Service Level Agreement) as the main means. The level agreement is a contract between the network service provider and the customer, which defines terms such as service type, service quality and customer payment, and it quantitatively constrains the minimum connection bandwidth and maximum loss of service components provided by the service provider. Packet rate and link failure rate are equal to performance-related parameters. Static performance analysis based on SLA is to directly use the limit of the above parameters constrained by the service level agreement as the model input, and assume that the performance parameters are constant, analyze and predict Web services The actual performance of components, this method of static analysis has great flaws: due to the rapidly changing network and system environment on which Web services run, various parameters that affect performance, such as bandwidth, packet loss rate, link failure rate, message It is impossible to keep constant the field length, etc., which greatly deviates from the assumption that the performance parameters are constant; since the parameters given in the service level agreement are all upper and lower limits, the corresponding amount when the Web service is running Values ​​may be above or below the limit, causing SLA-based performance prediction methods to overestimate or underestimate actual performance

Images