Dec, 17

Parallel Level set algorithm with MPI and accelerated on GPU

Level set method has been used to capture interface motion. Narrow band algorithm is applied to localize the solving of level-set PDE on global domain to a tube around interface. Due to the unknown evolving interface, narrow band algorithm brings load balance problem for parallelizing computing. This work presents a tool for evenly distributing work […]
Dec, 14

Automating the Last-Mile for High Performance Dense Linear Algebra

High performance dense linear algebra (DLA) libraries often rely on a general matrix multiply (Gemm) kernel that is implemented using assembly or with vector intrinsics. In particular, the real-valued Gemm kernels provide the overwhelming fraction of performance for the complex-valued Gemm kernels, along with the entire level-3 BLAS and many of the real and complex […]
Dec, 14

Translating OpenMP Device Constructs to OpenCL using Unnecessary Data Transfer Elimination

In this paper, we propose a framework that translates OpenMP 4.0 accelerator directives to OpenCL. By translating an OpenMP program to an OpenCL program, the program can be executed on any hardware platform that supports OpenCL. We also propose a run-time optimization technique that automatically eliminates unnecessary data transfers between the host and the target […]
Dec, 14

GaDei: On Scale-up Training As A Service For Deep Learning

Deep learning (DL) training-as-a-service (TaaS) is an important emerging industrial workload. The unique challenge of TaaS is that it must satisfy a wide range of customers who have no experience and resources to tune DL hyper-parameters, and meticulous tuning for each user’s dataset is prohibitively expensive. Therefore, TaaS hyper-parameters must be fixed with values that […]
Dec, 14

Towards Comprehensive Parametric Code Generation Targeting Graphics Processing Units in Support of Scientific Computation

The most popular multithreaded languages based on the fork-join concurrency model (CilkPlus, OpenMP) are currently being extended to support other forms of parallelism (vectorization, pipelining and single-instruction-multiple-data (SIMD)). In the SIMD case, the objective is to execute the corresponding code on a many-core device, like a GPGPU, for which the CUDA language is a natural […]
Dec, 14

nmfgpu4R: GPU-Accelerated Computation of the Non-Negative Matrix Factorization (NMF) Using CUDA Capable Hardware

In this work, a novel package called nmfgpu4R is presented, which offers the computation of Non-negative Matrix Factorization (NMF) on Compute Unified Device Architecture (CUDA) platforms within the R environment. Benchmarks show a remarkable speed-up in terms of time per iteration by utilizing the parallelization capabilities of modern graphics cards. Therefore the application of NMF […]
Dec, 13

5th International Conference on Sustainable Development (ICSD), 2017

The 5th ICSD 2017 will be an excellent opportunity to share your ideas and research findings relevant to the Sustainability Science, through the European network of academics Papers will be published in EJSD Journal (Thompson Reuters) and Proceedings. European Center of Sustainable Development in collaboration with CIT University will organize the 5th ICSD 2017 Rome, […]
Dec, 10

cusFFT: A High-Performance Sparse Fast Fourier Transform Algorithm on GPUs

The Fast Fourier Transform (FFT) is one of the most important numerical tools widely used in many scientific and engineering applications. The algorithm performs O(nlogn) operations on n input data points in order to calculate only small number of k large coefficients, while the rest of n − k numbers are zero or negligibly small. […]
Dec, 10

Implementing and Evaluating Candidate-Based Invariant Generation

The discovery of inductive invariants lies at the heart of static program verification. This paper describes our efforts to apply candidate-based invariant generation in GPUVerify, a static checker of programs that run on GPUs. We study a set of 383 GPU programs that contain loops, drawn from a number of open source suites and vendor […]
Dec, 10

Performance Evaluation and Optimization of HPCG benchmark on CPU + MIC platform

High-performance conjugate gradient (HPCG) is the latest benchmark adopted by the TOP500 organization, and thus how to optimize the HPCG source code for different heterogeneous computing platforms to achieve a higher floating-point computation rate has already become a new hot issue in HPC field. In the paper, we used the CPU + MIC heterogeneous computing […]
Dec, 10

GPGPU Accelerated Deep Object Classification on a Heterogeneous Mobile Platform

Deep convolutional neural networks achieve state-of-the-art performance in image classification. The computational and memory requirements of such networks are however huge, and that is an issue on embedded devices due to their constraints. Most of this complexity derives from the convolutional layers and in particular from the matrix multiplications they entail. This paper proposes a […]
Dec, 10

Adaptive Work-Efficient Connected Components on the GPU

This report presents an adaptive work-efficient approach for implementing the Connected Components algorithm on GPUs. The results show a considerable increase in performance (up to 6.8x) over current state-of-the-art solutions.
Page 31 of 931« First...1020...2930313233...405060...Last »

* * *

* * *

HGPU group © 2010-2017 hgpu.org

All rights belong to the respective authors

Contact us: