http://user.it.uu.se/~jarst116/slides/week4.pdf

Graph-Based Algorithms for Boolean Function Manipulation

This paper focuses on evaluating the impact of different data layouts on the computational efficiency of GPU-accelerated Inverse Distance Weighting (IDW) interpolation algorithm. First we redesign and improve our previous GPU implementation that was performed by exploiting the feature of CUDA dynamic parallelism (CDP). Then we implement three versions of GPU implementations, i.e., the naive version, the tiled version, and the improved CDP version, based upon five data layouts, including the Structure of Arrays (SoA), the Array of Structures (AoS), the Array of aligned Structures (AoaS), the Structure of Arrays of aligned Structures (SoAoS), and the Hybrid layout. We also carry out several groups of experimental tests to evaluate the impact. Experimental results show that: the layouts AoS and AoaS achieve better performance than the layout SoA for both the naive version and tiled version, while the layout SoA is the best choice for the improved CDP version. We also observe that: for the two combined data layouts (the SoAoS and the Hybrid), there are no notable performance gains when compared to other three basic layouts. We recommend that: in practical applications, the layout AoaS is the best choice since the tiled version is the fastest one among three versions. The source code of all implementations are publicly available.

/pdf/impact-of-data-layouts-on-the-efficiency-of-gpu-accelerated-2smhha4v0x.pdf

Impact of data layouts on the efficiency of GPU-accelerated IDW interpolation

The role of virtual and augmented reality in orthopedic trauma surgery: From diagnosis to rehabilitation.

Received: date / Accepted: date Abstract This paper focuses on evaluating the performance impact of different data layouts on the GPU-accelerated IDW interpolation. First, we redesign and improve our previous GPU implementation that was performed by exploiting the feature CUDA Dynamic Parallel (CDP). And then, we im- plement three versions of GPU implementations, i.e., the na¨ ive version, the tiled version, and the improved CDP ver- sion, based on five layouts including the Structure of Arrays (SoA), the Array of Sturcutes (AoS), the Array of aligned Sturcutes (AoaS), the Structure of Arrays of aligned Struc- tures (SoAoS), and the Hybrid layout. Experimental results show that: the layouts AoS and AoaS achieve better perfor- mance than the layout SoA for both the na¨ ive version and tiled version, while the layout SoA is the best choice for the improved CDP version. We also observe that: for the two combined data layouts (the SoAoS and the Hybrid), there are no notable performance gains when compared to other three basic layouts. We recommend that: in practical appli- cations, the layout AoaS is the best choice since the tiled version is the fastest one among the three versions of GPU implementations, especially on single precision.

/pdf/performance-impact-of-data-layout-on-the-gpu-accelerated-idw-2mcr3hoftv.pdf

Performance Impact of Data Layout on the GPU-accelerated IDW Interpolation

Prediction of environmental phenomena at non-observed locations is a fundamental task in geographic information science. Often, samples are taken at a limited number of sensor locations and spatial and spatio-temporal interpolation is used to generate continuous maps. The computational cost of the underlying algorithms usually grows with the number of data entering the interpolation and the number of locations for which interpolated values are needed. Thus, real-time provision and processing of large spatio-temporal datasets call for scalable computing solutions. This requires re-thinking of established (sequential) programming paradigms. In this paper, we study the implementation and behavior of inverse distance weighted interpolation (IDW) on a single graphics processing unit (GPU), as an example for spatial interpolation algorithms in massively parallel computing environments. We argue that the underlying ideas can be expanded to a framework providing highly parallel functions for geostatistics.

https://www.agile-online.org/Conference_Paper/CDs/agile_2011/contents/pdf/shortpapers/sp_157.pdf

Spatial interpolation in massively parallel computing environments

In this paper, we try to present the problems of the modern approach to treating various phobias Virtual environments created by virtual reality (VR) tools can help to make the treatment of certain types of phobias more efficient Attention to this form of phobia treatment with the help of VR is on the rise in the world, so we are also monitoring its development, as well Our paper introduces the necessary hardware and software that has been piloted in our department, but also methods of creating virtual environments, models and application designed for medical therapies of patients

Phobia treatment with the help of virtual reality

The global warming has an effect on changes of a snow cover over wintertime. This effect is observed in Slovak ski resorts, too. A prediction of these trends is important to build new and keep the existing ski resorts. We are able to analyze the snow cover depth in detail. This analysis is based on many continuous observations and measurements at the specific climatologic stations of Slovak Hydrometeorogical Institute. However, these climatologic stations do not render accurately all ski places, which could be examined. The aim of this work is the depth of the snow cover computing in the desired point based on the geographical characteristics of a specific geographical point in a modeled area. The main characteristics of the computing is the fact that it is time-consuming. One solution is a utilization of graphics processing units (GPUs) where the availability of enormous computational performance of easily programmable GPUs can rapidly decrease time of computing.

In our article we demonstrate how to deploy the CUDA architecture, which utilizes the powerful parallel computation capacity of GPU, to accelerate computational process of snow cover depth using the inverse-distance weighting (IDW) method. The outputs are visualized by the Grass GIS tool.

Design and performance evaluation of snow cover computing on GPUs

This paper proposes a model of neural tree architecture with probabilistic neurons. These trees are used for classification of a large amount of computer grid resources to classes. The first tree is used for classification of hardware part of dataset. The second tree classifies patterns of software identifiers. Trees are implemented to successfully separate inputs into nine classes of resources. We propose Particle Swarm Optimization model for tasks scheduling in computer grid. We compared time of creation of schedule and time of makespan in six series of experiments without and with using neural trees. In experiments with using neural tree we gained the subset of suitable computational resources. The aim is effective mapping of a large batch of tasks into particular resources. On the base of experiments we can say that improvements have been made even for middle and small batch of tasks.

/pdf/a-neural-tree-model-for-classification-of-computing-grid-vs7f792n70.pdf

A neural tree model for classification of computing grid resources using PSO tasks scheduling

The aim of this work is the depth of the snow cover computing in the desired point based on the geographical characteristics of a specific geographical point in a modeled area. The measured data are known on few places. These places are coincident with raingauge stations. These data have been collected by many continuous observations and measurements at the specific climatologic raingauging stations of Slovak Hydrometeorogical Institute. An interpolation method is necessary to obtain a representation of real situation about whole surface. The main characteristic of the interpolation computing is the fact that it is time-consuming. In paper, we present two cheap approaches of HPC. The first solution is a utilization of graphics processing units (GPUs) where the availability of enormous computational performance of easily programmable GPUs can rapidly decrease time of computing. The second one is a utilization of multithread CPUs. In our article we demonstrate how to deploy the CUDA architecture, which utilizes the powerful parallel computation capacity of GPU, to accelerate computational process of snow cover depth using the inversedistance weighting (IDW) method. The performance of GPU we face with OpenMP implementation of IDW method. We consider variable number of threads per CPU. The outputs are visualized by the GIS Grass tool.

/pdf/comparison-of-design-and-performance-of-snow-cover-computing-41sdy3fwue.pdf

Comparison of design and performance of snow cover computing on GPUs and multi-core processors

Abstract The OurGrid distributed environment is an open source grid middleware that enables the creation of P2P computational grids to speed up the execution of bag-of-tasks applications. Bag-of-tasks is an application model composed of several independent tasks not communicating with each other during their execution, tolerating network delays and faults. The paper shows that the main advantage to avoid bottlenecks is if the data gets larger or computations become more complex and the conclusion is made based on the presented measurements. Understanding such patterns sheds an insightful light on performance bottlenecks and design tradeoffs for P2P Grid applications

/pdf/analysis-of-performance-bottleneck-of-p2p-grid-applications-30hw5nxvhd.pdf

Vladimír Siládi

Papers

Phobia treatment with the help of virtual reality

Design and performance evaluation of snow cover computing on GPUs

A neural tree model for classification of computing grid resources using PSO tasks scheduling

Comparison of design and performance of snow cover computing on GPUs and multi-core processors

Analysis of performance bottleneck of P2P grid applications