Claudio Zunino

An accelerated remote graphics architecture for PDAS

A new category of devices, known as Personal Digital Assistant (PDA), has become increasingly widespread since the end of the nineties. A large number of software applications have been developed for PDAs, but high-quality 3D graphics still remain beyond the computational capability of these devices.This paper tackles this issue by proposing a generic solution for hardware-accelerated remote rendering on cluster. The rendering task is submitted to a PC/workstation cluster (each cluster machine is equipped by a graphics accelerator) by means of the Chromium architecture. Each machine renders a part of the image that is then reassembled and sent to the PDA via a software bridge. On the PDA side, the user can explore the scene using an ad-hoc navigation interface.The proposed solution allows to display extremely realistic and complex models in an interactive way. Moreover, our architecture does not depend on commercial solutions/products and can be easily modified in order to better fulfill requirements of specific applications.Computer GraphicsDistributed/network graphics Distributed SystemsDistributed applications

A distributed architecture for searching, retrieving and visualizing complex 3D models on Personal Digital Assistants

Abstract Mobile devices are significantly changing the human–computer interaction. In particular, the ubiquitous access to remote resources is one of the most interesting characteristics achievable by using mobile devices such as Personal Digital Assistants, cellular phones and tablets. This paper presents an architecture that allows users to search and visualize complex 3D models over Personal Digital Assistants. A peer-to-peer network of brokers manages queries for searching objects among several data providers. The object selected for visualization is forwarded to a specialized graphics provider; this provider allows the users to investigate the object remotely rendering the scene and sending back to the Personal Digital Assistant the computed image. The user can interactively analyse objects that would not be otherwise visualizable locally on the Personal Digital Assistant.

Adding a scalar value to texture-based vector field representations by local contrast analysis

Several algorithms can effectively represent vector fields by texture-based representations, visualizing at most all information on the field: direction, orientation, and local magnitude. An open problem still remains the mapping on textures of adjunctive information such as temperature, pressure, and so on, without using colors. This article addresses this issue by proposing a technique to add a scalar value denoting streamlines by means of different levels of contrast. Both streamline starting tones and the range of tones depend on the scalar value to be mapped; in this way, areas visualized by different contrast levels are represented. Two examples show the effectiveness of the proposed technique.

A distributed JXTA-based architecture for searching and retrieving solar data

This paper describes a hybrid distributed architecture, based on the JXTA technology, for searching and discovering data in a federation of solar observation archives. This work has been developed within the European project EGSO (European Grid of Solar Observations) with the intent of studying and analyzing a set of technologies and showing the brokerage capabilities of a highly controlled distributed network topology.Three roles have been identified: providers that contain data and metadata, brokers that manage providers and user queries, and consumers/clients. JXTA is used to develop all network system functionalities, while a native XML database is used to store resource descriptions.

A Distributed Visualization Environment for Scientific Visualization Based on Jini Technology

In computational fluid dynamics (CFD), visualization is a powerful instrument for research, training and development. It has been used for data evaluation and for qualitative comparison to flow visualization originating from experiments. The design and implementation of distributed visualization systems involve a set of issues mainly concerning performance, reliability and ease of developing and upgrading the system. In this paper, we consider a new technology called Jini used for designing distributed computing environments; we point out advantages and drawbacks related to the employment of Jini in the distributed visualization of vector fields based on particle tracing.

A PVM-Based Parallel Implementation of the REYES Image Rendering Architecture

In this paper a PVM-based distributed platform of the well known REYES rendering architecture developed byPix ar is presented. This work effectivelyt ackles issues related to load balancing and memory allocation bya master-slave paradigm. In particular, the rendering is distributed performing both an image and an object space subdivision; in this way, low memory resources are necessary to the slave side. Examples show the effectiveness of the proposed work.