Global ETD Search

Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.

Results

Showing 1 to 20 of 28 for “"AVX"”.

  1. Optimierung der Energie-Effizienz für Algorithmen der Linearen Algebra durch SIMD-Programmierung und AVX-Vektorisierung

    … die Prozessor- und Instruktionssatzerweiterung AVX zeigt Potenzial zur energie-effizienten Ausführung von Algorithmen der linearen Algebra, wobei die erzielte Energie-Effizienz von der Umsetzung der Implementierung abhängt. Für die gezeigten Untersuchungen werden drei repräsentativ ausgewählte …

    qucosa-diss

  2. JIT-Based RVV Optimization for Implicit GEMM Convolution and FlashAttention - With Cross-Architecture Autotuning, a Winograd F(4,3) Ablation, and AVX2 / AVX-512 / NEON Baselines

    … across four CPU vector backends—x86-64 with AVX2 and AVX-512, ARM with NEON, and RISC-V with the Vector Extension (RVV)—and complement it with a JIT-based FlashAttention kernel and an RVV Winograd F(4,3) ablation study. The contributions of this thesis are: 1. A lightweight Just-In-Time (JIT) …

    chalmers Repository record for JIT-Based RVV Optimization for Implicit GEMM Convolution and FlashAttention - With Cross-Architecture Autotuning, a Winograd F(4,3) Ablation, and AVX2 / AVX-512 / NEON Baselines (opens in a new tab)

  3. SIMD-fied bulk R-tree construction

    … of algorithms. One such SIMD instruction set is AVX-512, having 512-bit wide registers that can process data in parallel, significantly improving throughput compared to non-SIMD instructions. However, no work has studied the impacts of AVX-512 and of its gather instruction on the STR algorithm …

    umn Repository record for SIMD-fied bulk R-tree construction (opens in a new tab)

  4. Virus elimination from Actinidia germplasm

    … stem grooving virus (ASGV), Actinidia virus X (AVX), Citrus leaf blotch-like virus (CLBV-a), and Cucumber mosaic virus CMV). Also, the phylogenetic relationship of AMV isolated from Actinidia was compared to other extant isolates. It was found that Actinidia chinensis and Actinidia deliciosa …

    auckland-ms Repository record for Virus elimination from Actinidia germplasm (opens in a new tab)

  5. Hodnocení finanční situace společnosti a návrhy na její zlepšení

    … práce je hodnocení finanční situace společnosti AVX Czech Republic, s.r.o. od roku 2013 po rok 2018 pomocí metod finanční analýzy. V první části práce jsou rozebrány ukazatele finanční analýzy z teoretického hlediska. Druhá část obsahuje informace o uvedené společnosti spolu s analýzou současného …

    brno-tech Repository record for Hodnocení finanční situace společnosti a návrhy na její zlepšení (opens in a new tab)

  6. Asistovaná vektorizace a paralelizace kódu pomocí standardu OpenMP 4.0

    … rýchlych vyrovnávacích pamätí a modulov SSE/AVX, ktoré hrajú veľmi významnú rolu v oblasti optimalizácie. Práca demonštruje výhody optimalizovaného kódu pomocou štandardu OpenMP 4.0 oproti neoptimalozavanému kódu na sade benchmarkov zameraných na rôzne aspekty optimalizácie.

    brno-tech Repository record for Asistovaná vektorizace a paralelizace kódu pomocí standardu OpenMP 4.0 (opens in a new tab)

  7. Možnosti akcelerace symbolické regrese pomocí kartézského genetického programování

    … se architekturou SIMD a instrukční sadou SSE a AVX. Práce představuje řadu optimalizačních metod, které vedou k výraznému urychlení evoluce v kartézském genetickém programování. Metoda bitově paralelní simulace používající vektory AVX2 umožňuje paralelně pracovat s 256 vstupními kombinacemi …

    brno-tech Repository record for Možnosti akcelerace symbolické regrese pomocí kartézského genetického programování (opens in a new tab)

  8. Optimizing Convolutions in State-of-the-art Convolutional Neural Networks on Intel Xeon Phi

    … convolution for Intel Xeon Phi systems with AVX-512 support. Our strategy involves dynamic compilation approach along with standard compiler optimizations and software prefetching. We show that our JIT-based approach for direct convolution achieves close to peak performance on KNL for many …

    rice Repository record for Optimizing Convolutions in State-of-the-art Convolutional Neural Networks on Intel Xeon Phi (opens in a new tab)

  9. CUDA Enhanced Filtering In a Pipelined Video Processing Framework

    … video through SIMD instructions such as SSE2 or AVX. However, even with these instruction sets, CPUs are limited by their inherently sequential design, and can only operate on a handful of bytes in parallel. Even processors with a multitude of cores only execute on an elementary level of …

    calpoly Repository record for CUDA Enhanced Filtering In a Pipelined Video Processing Framework (opens in a new tab)

  10. Acceleration of Computational Geometry Algorithms for High Performance Computing Based Geo-Spatial Big Data Analysis

    … and Vectorization with compiler intrinsic (AVX). This dissertation has applied at least one of the mentioned techniques to the following problems. Novel method to parallelize plane sweep based geometric intersection for GPU with directives is presented. Parallelization of plane sweep based …

    marquette Repository record for Acceleration of Computational Geometry Algorithms for High Performance Computing Based Geo-Spatial Big Data Analysis (opens in a new tab)

  11. Analytical Query Processing Using Heterogeneous SIMD Instruction Sets

    … Multiple Data (SIMD) paradigm, e.g. SSE, AVX, and AVX512. Database systems have started to exploit SIMD to increase performance. However, this is still a challenging task, because existing algorithms were mainly developed for scalar processing and because there is a huge variety of …

    qucosa-diss

  12. Hardware-Assisted Dependable Systems

    … of modern CPUs, Elzar re-purposes Intel AVX extensions, and HAFT builds on Intel TSX instructions. The rest two target software bugs: SGXBounds detects vulnerabilities inside Intel SGX enclaves, and “MPX Explained” analyzes the recent Intel MPX extension to protect against buffer overflow …

    qucosa-diss

  13. Compilers for portable programming of heterogeneous parallel & approximate computing systems

    … the virtual ISA to NVIDIA GPUs, Intel’s AVX vector units, and to multicore X86-64 processors. Experimental results show that HPVM optimizations achieve significant performance improvements, HPVM translators achieve performance competitive with manually developed OpenCL code for both GPUs …

    uiuc Repository record for Compilers for portable programming of heterogeneous parallel & approximate computing systems (opens in a new tab)

  14. Hardware implementation of inference in deep neural networks

    … novel architecture and the very different AVX-512 SIMD architecture. Furthermore, I show the applicability of recurrent network architectures to a novel domain; the decoding of information encoded in the electrical spiking activity observed from ensembles of neurons. By comparing the …

    texas Repository record for Hardware implementation of inference in deep neural networks (opens in a new tab)

  15. On Parallel Computation of Large Smooth-Degree Isogeny

    … the Intel's Advanced Vector eXtensions (AVX) technology. To our best knowledge, we are the first to utilize both parallelization technologies for the isogeny computation. Also, to achieve effective implementations, we modify PCS for each execution environments and equip both …

    waterloo Repository record for On Parallel Computation of Large Smooth-Degree Isogeny (opens in a new tab)

  16. Exploiting Data-Level-Parallelism for Modern In-Memory OLAP Query Engines

    … novel instruction set extensions, like Intel's AVX-512, the functional capabilities of SIMD hardware have been continuously extended, allowing more complex operations to be efficiently executed in parallel. Consequently, this thesis outlines the opportunities and challenges of wider SIMD …

    qucosa-diss

  17. Exploiting Data-Level-Parallelism for Modern In-Memory OLAP Query Engines

    … novel instruction set extensions, like Intel's AVX-512, the functional capabilities of SIMD hardware have been continuously extended, allowing more complex operations to be efficiently executed in parallel. Consequently, this thesis outlines the opportunities and challenges of wider SIMD …

    qucosa-diss

  18. Novel materials and routes for rare-earth-free BaTiO3-based ceramics for MLCC applications.

    … on the hypotheses from this work were made by AVX Ltd in Coleraine. The devices possessed comparable TCC and better lifetimes compared with equivalent commercial products.

    whiterose Download PDF for Novel materials and routes for rare-earth-free BaTiO3-based ceramics for MLCC applications. (opens in a new tab)

  19. Enabling parallelism and optimizations in data mining algorithms for power-law data

    … vector lengths in commodity CPUs, such as AVX-512 with a vector length of 512 bits, efficient vector processing unit utilization becomes a major performance game-changer. By employing a static multi-version code generation strategy coupled with an algorithmic approximation based on the …

    gatech Repository record for Enabling parallelism and optimizations in data mining algorithms for power-law data (opens in a new tab)

Page 1 of 2