Measuring Change Impact Based on Usage Profiles
Measuring Change Impact Based on Usage Profiles
复制标题
DOI：
10.1109/icws.2012.35
复制
发表时间：
2012-06
期刊：
影响因子：
--
通讯作者：
Marcelo Yamashita;Bruno Vollino;K. Becker;R. Galante
中科院分区：
文献类型：
--
作者：
Marcelo Yamashita;Bruno Vollino;K. Becker;R. Galante
服务演进是一个关键问题，因为即使是很小的更改，如果不兼容，也可能影响大量的客户端应用程序。然而，特别是在大规模服务使用的情况下，根据其使用情况，更改对客户端有不同的影响。本文提出了一个变更管理框架，支持服务提供商的范围和量化的影响的基础上使用分析的变化。该框架采用了更细粒度的版本控制模型，以便轻松定位和评估服务描述中更改的兼容性。该框架还根据类似的使用模式对客户端应用程序进行集群，并在使用配置文件中对其进行总结。使用配置文件量化了相应应用程序使用的服务功能，从而能够评估不兼容更改对配置文件的影响。
Service evolution is a critical issue because even small changes, if not compatible, can potentially affect a huge number of client applications. However, particularly in the context of large scale service usage, changes have different impact on clients according to its use. This paper proposes a change management framework that supports service providers to scope and quantify the impact of changes based on usage analysis. The framework adopts a finer-grained versioning model in order to easily locate and assess the compatibility of changes in service descriptions. The framework also clusters client applications based on similar patterns of usage, summarizing them in usage profiles. A usage profile quantifies the functionality of the service used by the corresponding applications, enabling to assess the impact of incompatible changes against the profile.