Sökning: "GPU calculations"
Visar resultat 11 - 15 av 34 uppsatser innehållade orden GPU calculations.
11. ASIC implementation of LSTM neural network algorithm
Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)Sammanfattning : LSTM neural networks have been used for speech recognition, image recognition and other artificial intelligence applications for many years. Most applications perform the LSTM algorithm and the required calculations on cloud computers. LÄS MER
12. Parallelization of ray casting for solar irradiance calculations in urban environments
Magister-uppsats, Högskolan i Gävle/Samhällsbyggnad, GISSammanfattning : The growing amount of photovoltaic systems in urban environments creates peaks of energy generation in local energy grids. These peaks can lead to unwanted instability in the electrical grid. By aligning solar panels differently, spikes could be avoided. LÄS MER
13. ART vs. NDK vs. GPU acceleration: A study of performance of image processing algorithms on Android
Master-uppsats, KTH/Skolan för datavetenskap och kommunikation (CSC)Sammanfattning : The Android ecosystem contains three major platforms for execution suitable for different purposes. Android applications are normally written in the Java programming language, but computationally intensive parts of Android applications can be sped up by choosing to use a native language or by utilising the parallel architecture found in graphics processing units (GPUs). LÄS MER
14. Experimental Evaluation of GPU Solutions to the Single Source Shortest Path Problem
Kandidat-uppsats, KTH/Skolan för datavetenskap och kommunikation (CSC)Sammanfattning : Originally designed for computer graphics, the modern graphics processing unit (GPU) has now become a potential powerhouse for parallel calculations through the use of platforms such as CUDA and OpenCL. This paper presents an experimental evaluation of GPU- and CPU-parallel as well as CPU-sequential implementations for solving the single source shortest path problem. LÄS MER
15. Implementation of the Particle Mesh Ewald method on a GPU
Master-uppsats, KTH/Numerisk analys, NASammanfattning : The Particle Mesh Ewald (PME) method is used for efficient long-range electrostatic calculations in molecular dynamics (MD). In this project, PME is implemented for a single GPU alongside the existing CPU implementation, using the code base of an open source MD software GROMACS and NVIDIA CUDA toolkit. LÄS MER