A5056247052- Distributed and Parallel Computing Systems
- Simulation Techniques and Applications
- Parallel Computing and Optimization Techniques
- Peer-to-Peer Network Technologies
- Evacuation and Crowd Dynamics
- Advanced Data Storage Technologies
- Network Traffic and Congestion Control
- Caching and Content Delivery
- Traffic control and management
- Bioinformatics and Genomic Networks
- Transportation Planning and Optimization
- Cloud Computing and Resource Management
- Interconnection Networks and Systems
- Mosquito-borne diseases and control
- Mathematical Biology Tumor Growth
- Data Visualization and Analytics
- Gene Regulatory Network Analysis
- Distributed systems and fault tolerance
- Artificial Intelligence in Healthcare
- Internet Traffic Analysis and Secure E-voting
- Image and Video Quality Assessment
- Computer Graphics and Visualization Techniques
- Food Safety and Hygiene
- Scientific Computing and Data Management
- Business Process Modeling and Analysis
Universitat Autònoma de Barcelona
2013-2024
Abstract Emergency plan preparation is an important problem in building design to evacuate people as fast possible. Simulation exercises fire drills are not a realistic situation understand behaviour. In the case of crowd evacuations complexity and uncertainty systems increases. Computer simulation allows us run dynamics models extract information from emergency situations. Several solve evacuation problem. Individual oriented modelling give behaviour rules individual simulate interactions...
As optimization problems become more complicated and extensive, parameterization becomes complex, resulting in a difficult task requiring significant amounts of time resources. In this paper we propose heuristic search algorithm call MCSA (Montecarlo-Clustering Search Algorithm), which is based on Montecarlo sampling, clustering strategy involving two techniques. Our objective to apply MCSA, an inherently stochastic method, address problems. To assess its performance, conducted evaluation...
R is the preferred tool for statistical analysis of many bioinformaticians due in part to increasing number freely available analytical methods. Such methods can be quickly reused and adapted each particular experiment. However, experiments where large amounts data are generated, example using high-throughput screening devices, processing time required analyze often quite long. A solution reduce use parallel computing technologies. Because does not support computations, several tools have...
Energy consumption has become one of the greatest challenges in field high performance computing (HPC). The energy cost produced by supercomputers during lifetime installation is similar to acquisition. Thus, besides its impact on environment, a limiting factor for HPC. Our research aims reduce computer systems run parallel HPC applications. In this article we analyse possible influence programming paradigms shared memory (OpenMP) and message passing (MPI), behaviour at different clock...
Crowd evacuations involve thousands of persons in closed spaces. Having knowledge about where the problematic exits will be or disaster may occur can crucial emergency planning. We implemented a simulator using Agent Based Modelling able to model behaviour people evacuation situations and workflow run it cloud. The input is just PNG image output are statistical results simulation executed on This allows provide user with system abstraction only map scenario needed. Many events held main city...
Digital technologies offer unprecedented opportunities to modernize official food safety control. Shift from paper-to digital-based systems enable recording and reporting data directly in digital format further analyze disclose inspection data. Further this, the automatization of processes through rule-based instructions, a fact that could positively influence consistency Nevertheless, extent use during control inspections by competent authorities (CAs) European Union (EU) countries is not...
Individual-oriented simulation allows us to represent the global behavior of a system through local interaction in discrete time steps. As we face up close-to-reality models and large-scale workloads, focus on turning from traditional approaches towards distributed order obtain more accurate results less time. One main problems is how distribute individuals efficiently architecture. systems can be implemented fashion by using either grid-based or cluster-based approach. On one hand, consist...
Partitioning and load balancing are highly important issues in distributed individual-oriented simulation. Choosing how to distribute individuals on the environment can be a crucial factor at moment of executing an system should efficient order reduce communication involved interaction between belong different logical processes. Furthermore, if model exhibits mobility patterns, we able maintain keep global application performance. In this work, present proximity strategy for cluster-based...
Parallel discrete event simulation (PDES) have shown to be an useful paradigm for simulating complex and large-scale models. An individual-oriented approach allows modelers capture emerging global behaviors generated by simple local interaction, like observed in self-organized systems. Usually, this type of simulations are highly expensive terms computing communications. One one hand, we can reduce the involved individual interactions means developing a robust partitioning method. On other...
Advances in network technology make multicast one of the most feasible video streaming delivery techniques for near future. However, scalability a VoD system is limited by server bandwidth. In this paper, we propose new scheme that allows every active client to collaborate with order scale performance beyond servers physical limitations. The solution combined and peer-2-peer paradigm decentralize process. able merge two or more channels using distributed collaborations between group clients....
Computational simulation has been used as a powerful tool to represent the dynamical behavior of systems based on complex analytic models. These types models have two main drawbacks: (a) limitations due degree abstraction needed simulate them, (b) high computing power heavily simplified The available today can overcome these perform quicker simulations that are closer reality. In this paper, experiments and performance analysis distributed for individual oriented model (fish schools)...
Exploring the multi-core architecture is an important issue to obtaining high performance in parallel and distributed discrete-event simulations. However, simulation features must fit on programming model order increase performance. In this paper we show our experience developing a hybrid MPI+OpenMP version of discrete- event individual-oriented fish schooling simulator. approach developed, following manner: communication between Logical Processes happens via message passing whereas...
Dengue is a febrile disease whose main vector transmitter the Aedes Aegypti mosquito. This has an annual register of 50 million infections worldwide. Simulations are important tool in helping to combat and prevent epidemic and, consequently, save lives resources. Therefore, this paper, we propose Agent-Based Model for assessment pupal productivity In model, reproduction mosquito takes into account each type container. The preliminary results show effects considering control prevention...
True VoD systems provide interactive on-demand video service by allocating each client a dedicated stream. Such scheme is inefficient when the system has large number of users. A more scalable solution to use combination multicast and unicast channels with buffering. We show design proxy server that accepts commands from users using short streams "merge" onto existing patching prefetching technique. The augmented dynamic media delivery algorithm which, guarantees without starvation or...
Data partitioning is one of the main problems in parallel and distributed simulation. Distribution data over architecture directly influences efficiency The strategy becomes a complex problem because it depends on several factors. In an Individual-oriented Model, for example, related to interactions between individual environment. Therefore, simulation should dynamically enable interchange order choose most appropriate specific context. this paper, we propose strip spatially dependent Model...