Teruo Koyanagi

Generating transformational annotation for web document adaptation: tool support and empirical evaluation

Web annotation is crucial for providing machine-understandable descriptions of Web resources, and has a number of applications such as discovery, qualification, and adaptation of Web documents. While annotations are often embedded into a Web document, annotations can be associated externally by means of addressing expressions represented with the XPath language. However, creation of external annotation solely with a conventional editor is not easy because annotation authoring involves the maintenance and elaboration of addressing expressions as well as annotation contents. In addition, there has been little empirical study of robust pointing by XPath expressions, in spite of the increasing prevalence of the XPath language for use in emerging content adaptation systems. This paper proposes a classification of annotation tool design, taking account of differences in authoring methods and roles of annotation. On the basis of the classification, tools for generating external annotations are briefly explained along with applications of Web document adaptation for small-screen devices and portal site development. Robustness of the addressing expressions is then investigated, and practical implications to the reliable use of external annotation are drawn from empirical evaluation with evolving real-life Web documents.

Model analysis for business event processing

Business event processing requires efficiently processing live events, computing business performance metrics, detecting business situations, and providing real-time visibility of key performance indicators. Given the high volume of events and significant complexity of computation, system performance--event throughput--is critical. In this paper, we advocate model-analysis techniques to improve event throughput. In the build time, a series of model analyses of the application logic are conducted to understand such factors as runtime data-access path, data flow, and control flow. Such analyses can be used to improve throughput three ways: at build time it can be used to facilitate the generation of customized code to optimize I/O and CPU usage; information about the control flow and data flow can be used to ensure that CPU resources are used effectively by distributing event-processing computation logic evenly over time; and at runtime, knowledge gained from the model can be used to plan multithreaded parallel event-processing execution to reduce wait states by maximizing parallelization and reducing the planning overhead. This paper presents a series of model-analysis techniques and the results of experiments that demonstrate their effectiveness.

Authoring Tool for Business Performance Monitoring and Control

To monitor and warrant the performance of concerned business operations, formal representation for business performance models are needed so business analysts can describe what they expect from business service providers. This paper proposes a metamodel and an authoring tool for describing business performance for business services in the context of SOA. A business performance model explicitly captures the intentions of the consumers of the target business services. The authoring tool simplifies the modeling of the performance models. With our proposed mechanism, computational models for business performance management are generated. The separation of concerns in terms of business performance models and computational models allows for seamless transitions from business level performance models to runtime.

Agent server for a location-aware personalized notification service

Goopas is a location-aware personalized information notification service coupled with automated ticket gates in railway stations. The service was provided to 40,000 users by an earlier system implemented on J2EE, but had performance problems, because location-aware systems must finish user tasks before the users move, and such personalized systems require handling a large number of users. Agent Server was used for Goopas to replace the original system. This paper describes how Agent Server was used for a real service that provides high performance and strong capabilities to 100,000 users.

Dynamic policy management on business performance management architecture

Business performance management (BPM) is a new approach for an enterprise to improve their capabilities for sensing and responding to business situations. In a diverse and fast-changing business environment, an enterprise needs to adapt itself to any unexpected changes. For BPM, such changes imply changes of the models and services that support BPM. This paper discusses an implementation of BPM with the focus on dynamically adapting its services. We will present the motivation, concept and architecture of the dynamic change mechanisms. First we define a set of configurations as a policy, and also define its consistency through an application context. Then we propose an architectural overview including a policy management service as an implementation of consistency management.

A Hybrid Event-Processing Architecture based on the Model-driven Approach for High Performance Monitoring

Business activity monitoring (BAM) is a concept for computer systems which support monitoring situations and the performance of business processes as they execute. The monitoring system needs to capture and to process numerous events while correlating them with other events stored in a database. If the database accesses become very frequent, read-only and write-only accesses can achieve high throughputs by applying well known solutions; a result cache and a batch update. However, the event processing of the monitoring includes both simultaneously. It is the case that is hard for the existing solutions to achieve high throughputs. Here, we introduce a new architecture, HM-hybrid, to process events for monitoring efficiently. It is a hybrid of a result cache and a batch update. And our model-driven approach makes it possible for them to work together in situations that are normally regarded as being unsuited for them. It processed 1207 events per second, about 43000 queries per second. It is 6 times better performance over a naive caching architecture in our experiments. The architecture seems useful for monitoring applications where there are many update and select queries.