Collective Layer
- The Information Service implements mechanisms for discovering and “real-time” monitoring the whole set of resources available via infrastructure powered by the gCube system. By providing this functionality single services and virtual research environments are enabled for self-tuning resource usage and workload balancing to maximize the use of resources.
- The Keeper Service acts as the ‘orchestrator‘ of the services cpomposing the virtual research environment. It is the service in charge of bringing together the set of services belonging to a virtual research environment assuring the Quality of Service (QoS) characteristics required by its definition criteria. This will be achieved by monitoring both the status of the services and resources for the whole environment lifetime and by dynamically re-design at runtime the environment topology.
- The Broker & MatchMaker Service promotes efficient usage of the Grid infrastructure through an optimal distribution of services and resources across Grid nodes. This functionality possibly exploits the underlying Grid middleware integrating and customizing it to match the infrastructure needs.
- The Virtual Organisation Support Service provides the necessary functionalities for managing virtual organizations. A Virtual Organisation (VO) is a dynamic pool of distributed resources shared by dynamic set of users from one or more organizations in a trusted way. The infrastructure empowered by gCube will make use of this mechanism to glue together users and resources in a trusted environment.
Digital Library Creation and Management
- The Virtual Digital Library (VDL) Generator Service enables users/communities to create their own digital libraries (DLs). It permits users to specify a set of criteria expressing the expected characteristics of the new DL and then it identifies the services and information sources required to provide the requested features and content.
Process Management
- The Process Design & Verification Service provides the user with graphical tools for the design and manipulation of process definitions. Furthermore, before a compound service may be deployed, the provided validation functionality verifies that it obeys certain rules, so that the process can be considered safe for execution.
- The Process Execution & Reliability Service is responsible for all actions pertaining to the actual execution of compound services, i.e. for finding and allocating resources, starting, monitoring or aborting processes and handling process execution failures in accordance with transactional policies/guarantees.
- Within the Process Optimisation Service, user and system processes are being optimised prior to their execution by the actual services and components, under a rich, flexible cost-estimation policy which takes into consideration a multitude of parameters and values.
Content & Metadata Management
- The Content Management Service supports transparent access to the storage nodes as well as to external content providers and handles the update and the creation of new objects via various techniques such as polling and notifications, while providing support for URI maintenance, replication and partitioning.
- The Metadata Management Service manages metadata for the description and enrichment of digital objects handled by the content management service The Metadata Broker provides the broker for achieving metadata interoperability
- The Content Security components face the challenges of securing content storage and dissemination, safeguarding both Intellectual Property Rights (IPR) and content integrity via applied techniques such as signing, watermarking etc.
- The Annotation Management Service offers integrated management of annotations for digital objects available through the infrastructure, primarily multimedia content objects. Its main goal is to support international collaboration activities between distinguished users of a virtual research environment.
Indexing and Searching
- Feature Extraction offers the functionality for the analysis of media objects of various types. In particular it is capable generating metadata pertaining to image, video, audio, and text content. This generated metadata can be used to search for objects by its content and/or similar objects.
- The Index Service assists the search service by enabling efficient content-based retrieval. It will leverage state of the art search engines (like FAST’s Enterprise Search Platform) for advanced text searching, and extend search functionality to other media objects by indexing results from feature extraction as well as other metadata.
- Content Source Description and Selection Service optimises the distribution of content-based queries across search engines available on the infrastructure. Using up-to-date descriptions of the content targeted by each engine, the service automatically selects the engines which are most likely to produce relevant results for a given query.
- Data Fusion Service supports the real-time distribution of content-based queries across search engines available on the infrastructure. Using statistics about the engines and their per-query outputs, the service normalises and merges autonomously produced result rankings.
- Search Service is the top level service of the Digital Library Management System, i.e. the one that is being consumed in order to exploit the major functionality of the entire engine. It is a complex service which orchestrates the operation of the various underlying gCube services under the user's request for information retrieval. Searches over structured, semi-structured and unstructured data, have to be carried out to the infrastructure for optimal execution, and return back to the requestor the bunch of data he/she has been expecting for, potentially undergone complex processing or transformations.
- Through the Personalisation Service, search queries are being pre-processed and have personalisation information injected along with the core query data, having a direct result of adapting the returned information to specific users' needs and expectancies, by terms of cost, accuracy, content sources, formatting etc.
End-User Services
gCube provides a rich user interface to support both administration and management of the infrastructure, as well as end users in order to exploit the full potential of the Virtual Research Environment System. The technology adopted builds graphical portlets based on industrial standards, for hosting in almost any compliant Portal engine. Ultimately, users and administrators manage and use the infrastructure under the intuitive interface of a personalisable web portal.
Apart from the mechanisms to support end-user interaction with the system, gCube also provides the basic components for visualising the main information types to be met within the two application scenarios that form part of the project’s initial workplan.