Many modern computing platforms-notably clouds and desktop grids-exhibit dynamic heterogeneity: the availability and computing power of their constituent resources can change unexpectedly and dynamically, even in the midst of a computation. We introduce a new quality metric, AREA, for schedules that execute computations having interdependent constituent chores (jobs, tasks, etc.) on such platforms. AREA measures the average number of chores that a schedule renders eligible for execution at each step of a computation. Even though the definition of AREA does not mention any properties of host platforms (such as volatility), intuition suggests that rendering chores eligible at a faster rate will have a benign impact on the performance of volatile platforms. We report on simulation experiments that support this intuition. Earlier work has derived the basic properties of the AREA metric and has shown how to efficiently craft AREA-maximizing (A-M) schedules for several classes of significant computations. Even though A-M schedules always exist for every computation, it is not always known how to derive such schedules efficiently. In response, the current study develops an efficient algorithm that produces AREA-Oriented (A-O) schedules, which aim to efficiently approximate the AREAs of A-M schedules on arbitrary computations. The simulation experiments reported on here suggest that, in common with A-M schedules, A-O schedules complete computations on volatile heterogeneous platforms faster than a variety of heuristics that range from lightweight ones to computationally intensive ones-albeit not to the same degree as A-M schedules do. Our experiments suggest that schedules having larger AREAs have smaller completion times-but no proof of that yet exists.

Agent-Based Models (ABMs) are a class of models which, by simulating the behavior of multiple agents (i.e., ndependent actions, interactions and adaptation), aim to emulate and/or predict complex phenomena. One of the general features of ABM simulations is their experimental capacity, that requires a viable and reliable infrastructure to interact with a running simulation, monitoring its behaviour, as it proceeds, and applying changes to the configurations at run time, (the computational steering) in order to study "what if" scenarios.

We focus on agent-based simulations where a large number of agents move in the space, obeying to some simple rules. Since such kind of simulations are computational intensive, it is challenging, for such a contest, to let the number of agents to grow and to increase the quality of the simulation. A fascinating way to answer to this need is by exploiting parallel architectures. In this paper, we present a novel distributed load balancing schema for a parallel implementation of such simulations. The purpose of such schema is to achieve an high scalability. Our approach to load balancing is designed to be lightweight and totally distributed: the calculations for the balancing take place at each computational step, and influences the successive step. To the best of our knowledge, our approach is the first distributed load balancing schema in this context. We present both the design and the implementation that allowed us to perform a number of experiments, with up-to 1,000,000 agents. Tests show that, in spite of the fact that the load balancing algorithm is local, the workload distribution is balanced while the communication overhead is negligible.

We present the theoretical foundation, the design and the implementation of a library, called EulerVC to interactively handle Euler diagrams for the purposes of resource management. Fast on-line algorithms to interpret wellformed diagrams have been developed utilising a new notion of marked points to keep track of the zone sets. The interface allows the construction of overlapping ellipses to represent categories together with the drag and drop of resources in order to categories them. A visual indicator can be used to show if the diagram under construction is not wellformed to assist in reducing user mistakes, and sets of tags can be assigned to resources upon export. The generic approach is demonstrated via an integration with the bookmarking site delicious.

We have developed a novel information storage and display structure called EulerView, which can be used for the systematic management of tagged resources. The storage model is a non hierarchical classification structure, based on Euler diagrams, which can be especially useful if overlapping categories are commonplace. Keeping the constraints on the display structure relaxed enables its use as a categorisation structure which provides the user with flexibility and facilitates quick user tagging of multiple resources. As one instance of the application of this theory, in the case when the resources are photos, we have developed the Eulr tool which we have integrated with Flickr. User feedback indicates that the Eulr representation is intuitive and that users would be keen to use Eulr again.

We present a load-balancing technique that exploits the temporal coherence, among successive computation phases, in mesh-like computations to be mapped on a cluster of processors. Our method partitions the computation in balanced tasks and distributes them to independent processors through the prediction binary tree (PBT). At each new phase, current PBT is updated by using previous phase computing time (for each task) as (next phase) cost estimate. The PBT is designed so that it balances the load across the tasks as well as reduce {\em dependency} among processors for higher performances. Reducing dependency is obtained by using rectangular tiles of the mesh, of almost-square shape (i.e. one dimension is at most twice the other). By reducing dependency, one can reduce inter-processors communication or exploit local dependencies among tasks (such as data locality).Our strategy has been assessed on a significant problem, parallel ray tracing. Our implementation shows a good scalability, and improves over coherence-oblivious implementations. We report different measurements showing that granularity of tasks is a key point for the performances of our decomposition/mapping strategy.

EulerView incorporates a non-hierarchical classification structure to enable enhanced resource management. We apply the EulerView concept to the use of urls, constructing eli.icio.us which is a tool that realizes the EulerView concept and integrates it with del.icio.us. Currently, in del.icio.us, users can tag urls and browse for similar urls and the results can be displayed via an alphabetically sorted tag cloud. As well as facilitating easy tag management, using an EulerView display has the advantage that the user can use this representation for both tagging urls and for searching or browsing tasks, thus reducing any cognitive difficulties of users due to interacting with multiple representations.

EulerView is a resource management tool which supports operations, such as intersection, on categories. We relate EulerView displays to sets of paths in the transitive closure of the directed dual graph of an Euler diagram, and provide methods for transforming hierarchies with symbolic links into an EulerView display. Linking the display directly to the file system enabled the export of both the directory structure and resource placement which enables users to build small manageable views of their categorisation structure within which to place resources and then export this onto the larger file system. Using the transformation methods developed, we consider the importation of ACM classification into EulerView, enabling users to categorise and store resources such as articles within the imported ACM classification. EulerView is tailored to be appropriate for the ACM CCS application, and new techniques for focus and navigation within EulerView are developed.

Co-located collaboration in classroom is the topic we tackle in this paper. We present COFFEE a tailorable collaborative environment that is designed for interactive, co-located (i.e. Face2Face) collaboration in classroom. We present the requirements for tailorability that have driven our design, CoFFEE architecture, some tools that have been deployed and discuss the latecomer management issue that is offered by the core of our architecture to all the tools that can be developed within the framework.

In this paper we present a prototyped system to enable the virtual exploration of a complex virtual environment. Our approach exploits Quest3D as main rendering engine, its output is conveyed toward users PDAs to allow them to explore using the PDA as a (mobile) interface to the virtual environment. An important aspect of the system is that it relies on an off-the-shelf PC and low end wireless network. Some early results showed that the prototype is able to easily manage 5 PDAs. Suggested fields of use of our system are virtual cultural heritage, educational virtual environments, videogames.

The Treeview control is the traditional way of visualizing hierarchical information in user interfaces, but in situations such as managing bookmarks in browsers, more general classifications may be more appropriate than hierarchies. We have developed a component called EulerView which is an extension of Treeview, with a similar look and feel, addressing migration, navigation and overspecificity issues amongst 2D or 3D visualisations of classifications. It has familiar characteristics, and an underlying model, based on Euler diagrams, enabling more general classifications. We found, via user testing, that users of EulerView appear to grasp a fairly accurate internal representation of Euler diagrams and that they could organise bookmarks well within in a given classification using EulerView.

The paper presents some experiences in developing a virtual environment using a videogame engine. It provides the details and the caveats met during development, with particular attentions to the architectural reconstruction problem, the textures resolution and the programmability. Some considerations are also reported from an historical point of view, emphasizing the importance of virtual reconstructions, and from a psychological point of view stressing how a narrative context enhances the learning process.

In this paper we study state-of-the-art technologies to design interactive and cooperative 3D environments that are based on videogame 3D engines. We provide, first, a categorization of videogame 3D engines from the point of view of their usage in creating interactive 3D worlds and show a comparison of the most important characteristics. Then, we show an example of how we used a commercial videogame engine to create an interactive an enjoyable visit to an archaeological site.

In this paper we present our experience in building a visual flle manager, VENNFS2, that offers to users an adaptive interface toward access to files. Our file manager was originally designed to overcome some of limitations of hierarchical file systems, since it allows users to categorize files in such a way that files may belong multiple categories at once. Based on the past history of the files that were opened and modified by the user, VENNFS2 graphically presents the user a small number of choices of the next file the user will modify. Some preliminary testing with interesting hints are also reported.

We present a prototypal file manager, VENNFS, that is designed to overcome some of the limitations of the current desktop interfaces, that are strongly based on hierarchical file systems. VENNFS allows users to place documents and categories on a plane so that files may belong to multiple categories at once, where proximity on the plane can represent similarity and time filtering is allowed.