Catálogo de publicaciones - libros

This paper describes a case study of enhancing the deployment process in J2EE application servers (AS), and more precisely the services building such servers and the applications executing on the servers. We show how, by following a component-based approach to the design of the server, we address the versioning and licensing issues raised by the fact that a J2EE server is built out of heterogeneous, third-party software. As a proof of concept, we present a re-engineered version of the JOnAS J2EE server implemented using Fractal, a component model providing flexible control capabilities and hierarchical composition. We describe how Fractal packaging together with a JOnAS-specific deployment system are used to deploy and reconfigure our Fractal-based version of the JOnAS server. Finally, we show how the same model and packaging can be used to deploy applications executing on the server.

Millions of programmers use ECMA CLI-compliant languages like VB.NET and C#. The resulting bytecode can be executed on several CLI implementations, such as those from Microsoft and the open-source Mono organisation. While assemblies are the standard unit of deployment, no standard exists for the process of finding and loading assemblies at run-time. The process is typically complex, and varies between CLI implementations. Unlike other linking stages, such as verification, it is visible to programmers and can be a source of confusion. We offer a framework that describes how assemblies are resolved, loaded and used in CLI implementations. We strive for implementation-independence and note how implementations from different organisations vary in behaviour. We describe the reflection features available for dynamic loading, and give C# examples that exercise the features modelled in the framework.

The demand for component and service discovery engines to use in extensible applications is surging. No one so far has devoted much effort to metrics that aid selecting among different resolvers of the same dependency. This paper defines the Reuse Frequency: a metric that relates components or services to each other and measures their relative importance. Additionally, the ComponentGraph is presented that builds the averaged dependency graph of entities augmented with their popularity and the likelihood of each possible dependency resolver. The Reuse Frequency targets all scenarios where entities have dependencies on each other and a metric for the measurement of their relative importance is needed; the target implementation environment of the ComponentGraph is the Open Service Gateway Initiative framework, but the concepts are applicable to component or service repositories in general.

The current trend consists in deploying, on each machine, a specific version of an application, according to the choices of the enterprise and users, with constraints verified by the target site. To support automated deployment, we propose a model-based deployment framework named ORYA which allows to define and execute deployment strategies. This paper presents and illustrates the concept of deployment strategy supported by the framework.

The ability to deploy Grid infrastructure and services across organizational boundaries (rapidly, reliably, and scalably) is critical for the success of large-scale service based grids such as OGSA. We report the results of the UK-OGSA Evaluation Project infrastructure and services deployment experiments, and analytically compare application versus service deployment. The use of a 3^rd party component deployment technology to remotely automate installation and service deployment is discussed, and outstanding problems and potential solutions and benefits are presented. We conclude that grid deployment must be treated as a first-order activity by integrating secure deployment capabilities into the middleware, to enable deployment of secured infrastructure and services across organizations.