Home  
 

Main Menu
Home
About us
Project Description
Quantitative Results
Research Lines
Research Results
Impact on Society
Press room
Contact us
News
Secure Login
Events Calendar
« < October 2017 > »
M T W T F S S
25 26 27 28 29 30 1
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29
30 31 1 2 3 4 5
Login

Task 1 (UVEG): Performance improvements of Distributed Virtual Enviroments (DVE)

Leader: Juan Manuel Orduña; Researchers: Pedro Morillo, Silvia Rueda, Víctor Fernández

subTask 1.1 Development of an efficient awareness algorithm for Distributed Virtual Environments based on Peer-to-Peer architectures (P2P DVE)

subTask 1.2  Development of a distributed simulation tool for P2P DVEs

subTask 1.3  Performance improvement in DVE P2P systems

subTask 1.4 Performance improvement of DVE systems based on networked-server architecture

1. Brief Description of the Goals

Large scale distributed virtual environments (DVEs) have become a major trend in distributed applications nowadays. One of the reasons for this trend is the enormous popularity of multi-player online games in the entertainment industry. Distributed Virtual Environments simulate a 3-D virtual world where multiple users share the same scenario. Each user is represented in the shared virtual environment by an entity called avatar, whose state is controlled by the user through a client computer. The system renders the images of the virtual world that each user would see if he was located at that point in the virtual environment. Thousands and even hundreds of thousands of client computers can be simultaneously connected to the DVE. The client computers can be connected through different networks, and even through Internet.

 In these systems, architectures based on networked servers have been the major standard for DVEs during the last years. In these architectures, the control of the simulation relies on several interconnected servers. Client computers are assigned to one of the servers in the system. When a client computer modifies the state (usually the position, but it can also modify the appearance or other stateful information) of an avatar, it also sends an updating message to its server, which in turn must propagate this message to other servers and clients. The main reasons for the prevalence of networked-server architectures over peer-to-peer architectures have been the control of the simulation and the awareness problem. However, peer-to-peer (P2P) architectures have been proposed as a flexible scheme in order to provide good scalability for large scale DVEs, and several online games based on P2P architectures have been designed last years.

2. Scientific and Technical Developed Activities

We have filed two Spanish patents: “Sistema de ayuda al mantenimiento preventivo y reparación de maquinaria remota e in-situ” and “Sistema de seguimiento en obras en construcción, as a result of this research task. 

On the one hand, our research has focused on improving the performance of networked-server DVEs published by P. Morillo et al. in IEEE Transactions on Systems Man and Cybernetics Part A-Systems and Humans, Vol. 39, No 6, 2009. In this sense, we have developed a latency-aware partitioning method that is capable of providing the highest number of avatars with response times below any given threshold, published by Rueda et al. in Computer Communications, Vol. 30, issue 4, 2007 and by Rueda et al. in IEEE Transactions on Parallel and Distributed Systems, Vol. 18, issue 9, 2007.

On the other hand, our research has focused on DVEs based on peer-to-peer architectures. In this sense, we have characterized the behavior of P2P DVEs, published by Rueda et al. in IPDPS Conference Workshops, 2007 and by Rueda et al. in Computer Animation and Virtual Worlds, Vol 19 , No. 5, 2008. 

Based on these results, we have developed a new awareness method that provides full awareness under any conditions for both uniform and non-uniform movement patterns, published by Rueda et al. in Cyberworlds Conference, 2007. Finally, we have developed a saturation avoidance technique that limits the effects of client saturation, guaranteeing reasonable system performance levels under huge workloads, published by Rueda et al. in Future Generation Computer Systems, Vol. 26, No. 7,  2010.

From the beginning of 2011, we started to work on Collaborative Augmented Reality, a new research line not proposed in the project. In this research line, we have developed a simulator based on the characterization of current client devices, published by Fernández-Bauset et al. in IEEE DS-RT Conference, 2011. Using this simulator we have characterized CAR systems based on marker detection, as well as CAR systems based on image features. We have detected and removed different system bottlenecks, improving the performance and scalability of CAR systems based on mobile phones, by Olanda et al. in Future Generation Computer Systems, 2012, by Olanda et al. in International Conference on Computer Graphics Theory and Applications (GRAPP) 2012, by Gimeno et al. in GRAPP 2012, published by Fernández-Bauset et al. in HPCC, 2012 and by Fernández-Bauset in Journal of Supercomputing, Vol.  2010.


Publications: [Rued07], [Rued07b], [Rued07c], [Rued07d], [Rued07e], [Rued07f], [Rued08],  [Rued08b], [Rued08c], [Rued09], [Mori08], [Asen08]

Projects funded by Public Calls: [TIN2006-15516-C04-04 ], [TIN2009-14475-C04-04 ],  [TIN2007-29664-E ] by  national grants. 

External collaborations Academia: Christos BourasShun-Yun Hu

External collaborations Industry: Dragados (Spain)

Company Agreements: --

PhD dissertations:  Silvia Rueda Pascual